Handbook of Laboratory Animal Science Third Edition VOLUME 1 Essential Principles and Practices
84550.indb 1
11/11/10 6:23:53 PM
84550.indb 2
11/11/10 6:23:54 PM
Handbook of Laboratory Animal Science Third Edition VOLUME 1 Essential Principles and Practices
Edited by
Jann Hau Steven J. Schapiro
Boca Raton London New York
CRC Press is an imprint of the Taylor & Francis Group, an informa business
84550.indb 3
11/11/10 6:23:54 PM
CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487-2742 © 2011 by Taylor and Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-13: 978-1-4200-8456-6 (Ebook-PDF) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http:// www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com
Contents Preface...............................................................................................................................................ix Editors................................................................................................................................................xi Contributors.................................................................................................................................... xiii Chapter 1 The Contribution of Laboratory Animals to Medical Progress—Past, Present, and Future.............1 John D. Harding, Gerald L. Van Hoosier, Jr., and Franziska B. Grieder Chapter 2 Ethics of Animal Research............................................................................................................... 21 I. Anna S. Olsson, Paul Robinson, and Peter Sandøe Chapter 3 An Overview of Global Legislation, Regulation, and Policies......................................................... 39 Kathryn Bayne, Bryan R. Howard, Tsutomu Miki Kurosawa, and€Maria€Eugenia€Aguilar Nájera Chapter 4 Assessment of Animal Care and Use Programs and Facilities........................................................ 65 Javier Guillen, Letty V. Medina, and James R. Swearengen Chapter 5 Education and Training..................................................................................................................... 81 Nicole Duffee, Timo Nevalainen, and Jann Hau Chapter 6 Laboratory Animal Science and Service Organizations..................................................................97 Patri Vergara and Gilles Demers Chapter 7 Laboratory Animal Allergies and Zoonoses.................................................................................. 115 Richard M. Preece and Anne Renström Chapter 8 Laboratory Animal Facilities and Equipment for Conventional, Barrier, and Containment Housing Systems............................................................................................................................. 145 Jack R. Hessler
v
84550.indb 5
11/11/10 6:23:55 PM
vi
Contents
Chapter 9 Laboratory Animal Genetics and Genetic Quality Control...........................................................209 Michael F. W. Festing and Cathleen Lutz Chapter 10 Health Status and Health Monitoring............................................................................................. 251 Axel Kornerup Hansen Chapter 11 Nutrient Requirements, Experimental Design, and Feeding Schedules in Animal Experimentation..............................................................................................................................307 Jo H. A. J. Curfs, André Chwalibog, Bart S. Savenije, and Merel Ritskes-Hoitinga Chapter 12 Impact of the Biotic and Abiotic Environment on Animal Experiments....................................... 343 Nancy A. Johnston and Timo Nevalainen Chapter 13 Experimental Design and Statistical Analysis................................................................................ 369 Michael F. W. Festing Chapter 14 Common Nonsurgical Techniques and Procedures........................................................................ 401 Vera Baumans and Cynthia A. Pekow Chapter 15 Applications of Radiotelemetry in Small Laboratory Animals...................................................... 447 Klaas Kramer and Steve Hachtman Chapter 16 Immunization for Production of Antibodies................................................................................... 467 Jann Hau and Coenraad Hendriksen Chapter 17 Laboratory Animal Analgesia, Anesthesia, and Euthanasia.......................................................... 485 Ludo J. Hellebrekers and Patricia Hedenqvist Chapter 18 Welfare Assessment and Humane End Points................................................................................ 535 David B. Morton and Jann Hau
84550.indb 6
11/11/10 6:23:55 PM
Contents
vii
Chapter 19 Surgery: Basic Principles and Procedures...................................................................................... 573 M. Michael Swindle, Heather Elliott, and Alison C. Smith Chapter 20 Microsurgical Procedures in Experimental Research.................................................................... 597 Daniel A. Steinbrüchel Chapter 21 Postmortem Procedures.................................................................................................................. 613 Ricardo E. Feinstein and Kimberly S. Waggie Chapter 22 Alternatives: Refinement, Reduction, and Replacement of Animal Uses in the Life Sciences...... 635 Lisbeth E. Knudsen, Marlies Leenaars, Bart Savenije, and Merel Ritskes-Hoitinga Chapter 23 Generation and Analysis of Genetically Modified Mice................................................................ 653 Cord Brakebusch Chapter 24 Physiological, Hematological, and Clinical Chemistry Parameters, Including Conversion Factors............................................................................................................................................. 667 Grete Østergaard, Helle Nordahl Hansen, and Jan Lund Ottesen Index............................................................................................................................................... 709
84550.indb 7
11/11/10 6:23:56 PM
84550.indb 8
11/11/10 6:23:56 PM
Preface Most of our present knowledge concerning human physiology, microbiology, immunology, pharmacology, pathology, and related disciplines has been gained from studies involving animals—from studies of genetics in fruit flies to studies of cellular processes in genetically modified mice to investigations of life-threatening infections in nonhuman primates. Biomedical research involving animals remains absolutely essential for the advancement of the medical, veterinary, agricultural, and biological sciences. All drugs prescribed for use in humans and animals have been developed and tested in laboratory animals as models. Noninvasive imaging techniques are optimized in animal models. New surgical techniques and materials are evaluated in animals before they are applied in cases that involve humans or domestic animals. The dramatic developments in genetics, including the sequencing of the human genome and the genomes of many of the most important laboratory animal species, translational research, and personalized medicines all rely on access to high-quality laboratory animals as models for humans. In The Principles of Humane Experimental Technique (1959), W. M. S. Russell and R. L. Burch counseled scientists to aim to apply the three Rs whenever possible: replacing experiments on live animals with alternative methods, reducing the number of animals necessary to obtain valid results within experiments, and refining techniques to minimize the discomfort experienced by the animal participants. These three Rs form the cornerstones of laboratory animal science, and they have been integrated into the numerous laws and guidelines that regulate the use of animals in research across the globe. The three Rs have also been integrated into almost every chapter in this handbook; replacement, reduction, and refinement are relevant to virtually all areas of laboratory science. Good science can only be performed in environments that promote good animal welfare. Efficient and humane experimental work with animals, in which subjects experience no avoidable pain and mental distress, requires skillful and conscientious staff, including specialist veterinarians. In many parts of the world, regulatory authorities require that all staff working with laboratory animals must document relevant competencies in the field, many of which are obtained through formal teaching and training programs. Universities around the world have established mandatory courses for scientists who wish to use animals in their research, and some have developed specialist education programs (often master’s level courses) for staff to achieve laboratory animal specialist competence. This handbook is a revised third edition of the handbook first edited by Per Svendsen and Jann Hau, published in 1994. Jann Hau and his old friend and colleague, Gerald Van Hoosier, in Seattle, joined forces to edit the second edition. Now, Jann Hau and Steve Schapiro, who have been friends and collaborators on numerous projects for almost 20 years, have teamed up to revise and produce the third edition of this handbook of laboratory animal science. The result is a truly international book, and we wish to thank all of the authors for their valuable contributions. Each individual chapter focuses on an important subdiscipline of laboratory animal science; the chapters can be read and used as stand-alone texts, with only limited necessity to consult other chapters for information. This approach has resulted in slight overlaps in contents in certain chapters, but we feel that this was a small price to pay in order to make the book as reader friendly as possible. It is our hope that this handbook will be useful all over the world as a textbook in laboratory animal science courses for postgraduate and undergraduate students, as a handbook for scientists who work with animals in their research, and for university veterinarians, regulators, and other specialists in laboratory animal science. Jann Hau Steven J. Schapiro ix
84550.indb 9
11/11/10 6:23:56 PM
84550.indb 10
11/11/10 6:23:56 PM
Editors Jann Hau is professor in comparative medicine at the University of Copenhagen in Denmark. Dr. Hau did his M.Sc. in reproductive biology and immunochemistry in 1977 at the University of Odense in Denmark after medical and biology bachelor studies, and specialized in laboratory animal science. Following research fellowships at the University of Odense, he did his doctorate (Dr. Med.) at this university. In 1983, he joined the Department of Pathology at The Royal Veterinary and Agricultural University (RVAU) in Copenhagen as an associate professor and Head of the Laboratory Animal Science Unit. He was later Head of the Department of Pathology and Dean of the Faculty of Animal Husbandry and Veterinary Science at the RVAU. In this period he was also the Veterinary Research Council’s member of the State Board for Animal Experimentation. In 1991, he moved to the Royal Veterinary College (RVC) in London as a professor in the London University Chair in Laboratory Animal Science and Welfare. At the RVC he was responsible for the undergraduate and postgraduate teaching in laboratory animal science and welfare, which included a specialist master of science course in laboratory animal science, which attracted a number of postgraduate students from many parts of the world. While in the U.K. he was Certificate holder responsible for all animal experimentation at the Royal Veterinary College, University of London. In 1996, Dr. Hau was appointed Professor in Comparative Medicine at Uppsala University in Sweden, and Head of the new Department of Comparative Medicine. Following amalgamations of departments at the medical faculty, comparative medicine became integrated as a division of the Department of Physiology, of which Dr. Hau became chairman. The division became integrated into the Department of Neuroscience in 2002. In Uppsala he established a number of courses for undergraduate students and postgraduate students, including specialist education programs. He has supervised many postgraduate students (M.Sc. and Ph.D.) from a number of countries in Denmark, the U.K., and Sweden. In 2003, Dr. Hau was John H. Blaffer Visiting Professor at the University of Texas M.D. Anderson Cancer Center, Bastrop, Texas, in the United States. On September 1, 2004, Dr. Hau was appointed Professor in Comparative Medicine and Head of the Department of Experimental Medicine at the University of Copenhagen, The Panum Institute in Denmark. Dr. Hau has published more than 200 peer-reviewed papers in comparative medicine and chapters in books, and he is frequently invited to speak at international conferences and symposia. He is the recipient of prizes awarded for his research and contributions to animal welfare. Together with Dr. P. Svendsen he wrote the first Danish textbook on laboratory animals and animal experiments published in 1981, 1985, and 1989, and they co-edited the first edition of the Handbook of Laboratory Animal Science published in 1994. Together with Dr. G. Van Hoosier he edited the 2nd edition of the Handbook, which was published in 2003 and 2004. Dr. Hau has organized several international meetings and courses on laboratory animal science in many different countries and is frequently invited to lecture at international courses and educational symposia. He is the editor-in-chief of the Scandinavian Journal of Laboratory Animal Science, editor of the laboratory animals section of the UFAW journal Animal Welfare and member of the editorial board of the journal In Vivo. He is a member of a number of laboratory animal science and primatology organizations, and former president of the Scandinavian Society of Laboratory Animal Science (ScandLAS) and the Federation of European Laboratory Animal Science Associations (FELASA). Dr. Hau is a member of AAALAC council, and former chairman of the FELASA Accreditation Board for European laboratory animal science courses. He is a member of several EU and international advisory boards and working groups on various aspects of comparative medicine, laboratory animal science, and primatology. xi
84550.indb 11
11/11/10 6:23:56 PM
xii Editors
Steven Schapiro is associate professor of comparative medicine in the Department of Veterinary Sciences at the Michale E. Keeling Center for Comparative Medicine and Research at the University of Texas M. D. Anderson Cancer Center. Dr. Schapiro received his PhD from the University of California at Davis in 1985, after receiving his BA in behavioral biology from Johns Hopkins University. He completed a postdoctoral research fellowship at the Caribbean Primate Research Center of the University of Puerto Rico. In 1989, he joined the Department of Veterinary Sciences at M. D. Anderson’s Keeling Center and has been there ever since. In 2009, Dr. Schapiro was a visiting professor in the Department of Experimental Medicine at the Panum Institute, University of Copenhagen, Denmark. He is one of the founding faculty members of the Primate Training and Enrichment Workshops— educational programs that have been offered to over 600 individuals from around the globe involved in all aspects of caring for and working with captive nonhuman primates. Dr. Schapiro has published approximately 100 peer-reviewed papers and book chapters examining various aspects of nonhuman primate behavior, management, and research. He has participated in several international meetings and courses on laboratory animal science in different countries and served as coeditor for one issue of the ILAR Journal. He is a member of a number of primatology and animal behavior societies and is currently the treasurer and vice president for membership of the International Primatological Society. He is also a past president and former treasurer of the American Society of Primatologists. Dr. Schapiro is a consultant for a number of primate facilities in the United States and abroad that focus on the production, management, and use of nonhuman primates in biomedical research.
84550.indb 12
11/11/10 6:23:56 PM
Contributors Vera Baumans Utrecht University Utrecht, the Netherlands Kathryn Bayne AAALAC International Frederick, Maryland Cord Brakebusch Biomedical Institute University of Copenhagen Copenhagen, Denmark André Chwalibog The Royal Veterinary and Agricultural University Frederiksberg, Denmark Jo H. A. J. Curfs Central Animal Laboratory Radboud University Nijmegen Medical Center Nijmegen, the Netherlands Gilles Demers Canadian Council on Animal Care Ottawa, Ontario, Canada Nicole Duffee Washington University School of Medicine St. Louis, Missouri Heather Elliott GlaxoSmithKline Hertfordshire, United Kingdom
Franziska B. Grieder National Center for Research Resources National Institutes of Health Bethesda, Maryland Javier Guillen Association for the Assessment and Accreditation of Laboratory Animal Care International Pamplona, Spain Steve Hachtman Data Sciences International St. Paul, Minnesota Axel Kornerup Hansen The Royal Veterinary and Agricultural University Frederiksberg, Denmark Helle Nordahl Hansen Novo Nordisk A/S Måløv, Denmark John D. Harding National Center for Research Resources National Insitutes of Health Bethesda,€Maryland Jann Hau University of Copenhagen Copenhagen, Denmark Patricia Hedenqvist Karolinska Institutet Stockholm, Sweden
Ricardo E. Feinstein The National Veterinary Institute Uppsala, Sweden
Ludo J. Hellebrekers Department of Equine Sciences Department Clinical Sciences of Companion Animals Utrecht, the Netherlands
Michael F. W. Festing MRC Toxicology Unit University of Leicester Leicester, United Kingdom
Coenraad Hendriksen National Institute of Public Health and Environment Bilthoven, the Netherlands xiii
84550.indb 13
11/11/10 6:23:56 PM
xiv
Jack R. Hessler Hessler Consulting, LLC Laytonsville, Maryland
Timo Nevalainen University of Kuopio Kuopio, Finland
Bryan R. Howard Private consultant Sheffield, United Kingdom
I. Anna S. Olsson Institute for Molecular and Cell Biology Porto, Portugal
Nancy A. Johnston University of Washington Seattle, Washington
Grete Østergaard Department of Experimental Medicine University of Copenhagen Copenhagen, Denmark
Lisbeth E. Knudsen Department of Public Health University of Copenhagen Copenhagen, Denmark Klaas Kramer Department of Health, Safety and the Environment Free University Amsterdam Amsterdam, the Netherlands Tsutomu Miki Kurosawa The Institute of Experimental Animal Sciences Osaka University Medical School Osaka, Japan Marlies Leenaars Nijmegen Medical Center Radboud University Nijmegen, the Netherlands Cathleen Lutz The Jackson Laboratory Bar Harbor, Maine Letty V. Medina GPRD, Development Sciences, Abbott Laboratories Abbott Park, Illinois
84550.indb 14
Contributors
Jan Lund Ottesen Novo Nordisk A/S Måløv, Denmark Cynthia A. Pekow Veterans Affairs Puget Sound Health Care System Seattle, Washington Richard M. Preece AstraZeneca Macclesfield, Cheshire, United Kingdom Anne Renström Karolinska Institutet Stockholm, Sweden Merel Ritskes-Hoitinga University of Southern Denmark Odense, Denmark Paul Robinson Lexicon Editorial Surrey, England Peter Sandøe The Royal Veterinary and Agricultural University Frederiksberg, Denmark
David B. Morton University of Birmingham Edgbaston, Birmingham, United Kingdom
Bart S. Savenije Nijmegen Medical Center Radboud University Nijmegen, the Netherlands
Maria Eugenia Aguilar Nájera ESM-IPN Central Animal Facility Mexico City, Mexico
Alison C. Smith Medical University of South Carolina Charleston, South Carolina
11/11/10 6:23:57 PM
Contributors
84550.indb 15
xv
Daniel A. Steinbrüchel Rigshospitalet University of Copenhagen Copenhagen, Denmark
Gerald L. Van Hoosier, Jr. University of Washington Seattle, Washington
James R. Swearengen National Biodefense Analysis and Countermeasures Center Frederick, Maryland
Patri Vergara Universidad Autonoma de Barcelona Barcelona, Spain
M. Michael Swindle Medical University of South Carolina Charleston, South Carolina
Kimberly S. Waggie Zymogenetics Seattle, Washington
11/11/10 6:23:57 PM
84550.indb 16
11/11/10 6:23:57 PM
Chapter 1
The Contribution of Laboratory Animals to Medical Progress—Past, Present, and Future
John D. Harding, Gerald L. Van Hoosier, Jr., and Franziska B. Grieder Contents Introduction.........................................................................................................................................2 Past Role of Laboratory Animals.......................................................................................................2 Overview........................................................................................................................................2 Examples from the Past.................................................................................................................3 Specific Examples from the Past—Nonhuman Primates.........................................................4 Specific Examples from the Past—Nontraditional Species......................................................4 Public Concerns.............................................................................................................................5 Present Role of Laboratory Animals..................................................................................................5 Mouse Models................................................................................................................................5 Genetically Altered Mice and Knockout Mouse Projects.........................................................5 Large-Scale Mouse Programs...................................................................................................7 Repositories...............................................................................................................................8 Specific Mouse Models.............................................................................................................9 Rat Models................................................................................................................................... 10 Nonhuman Primate Models......................................................................................................... 11 AIDS....................................................................................................................................... 11 Hepatitis B and C.................................................................................................................... 12 Neurobiology........................................................................................................................... 12 Genetic Technologies and NHPs............................................................................................. 13 Fish Models.................................................................................................................................. 14 Other Animal Model Species....................................................................................................... 14 Future Directions for Laboratory Animals in Biomedical Research............................................... 15 Genomics..................................................................................................................................... 15 Proteomics, the Interactome, and Systems Biology..................................................................... 16 Knockout and Knockdown of Gene Expression.......................................................................... 16 Stem Cells and Assisted Reproductive Technologies.................................................................. 17 Phenotyping................................................................................................................................. 17 References......................................................................................................................................... 18
1
84550.indb 1
11/11/10 6:23:57 PM
2
Handbook of Laboratory Animal Science
Introduction Animal-based research has played a key role in understanding infectious diseases, neuroscience, physiology, and toxicology. Experimental results from animal studies have served as the basis for many key biomedical breakthroughs. Open-heart surgery provides an illustrative example. What transpired that allowed a surgeon to perform the first open-heart procedure? In fact, there were probably thousands of animal-based research steps that led to the first surgery, including research in management of heart failure asepsis, blood pressure, surgical instruments and materials, relief of pain, wound healing, surgical techniques, ventilation of an open thorax, transfusions, blood groups and typing, anticoagulants, antibiotics, and the pump oxygenator, to name just a few. Since the 1970s, molecular-based approaches have increasingly complemented and extended traditional animal research. New approaches include sophisticated techniques for cell and tissue culture, statistical analysis, and computer modeling. Perhaps most revolutionary has been the application of genetic technologies to animal studies. The techniques of genetic engineering, coupled with advances in assisted reproductive technologies, have made possible the derivation of mice in which the activity of virtually any gene can be manipulated in a tissue- and temporally specific manner. Studies of gene regulation of “lower” animals, such as the fruit fly, Drosophila melanogaster, and the nematode, Caenorhabditis elegans, have made major contributions toward understanding the relationship of gene structure to function—a relationship that underlies all physiological processes. The genomic sequences of humans and many animals have been determined, leading to an unprecedented level of knowledge of genetic structure, regulation, and evolution, which can be used to better understand the basis of disease in both humans and animals. Animal models are required to connect all of these technologies in order to understand whole organisms, both in healthy and diseased states. In turn, these animal studies are required for understanding and treating human disease. While this handbook discusses numerous aspects of the laboratory animal sciences, scientists also seek alternatives to the use of animals in research: reduce, refine, and replace, or the “three Rs”: Reduce the number of animals needed. Refine tests. Replace the use of animals whenever possible.
All animal-based experiments are designed such that animals are treated humanely and there are many regulatory systems in place to ensure this. Humane treatment of animal subjects is a primary concern of all scientists who use animals in research. Past Role of Laboratory Animals Overview Any attempt to assemble a comprehensive or detailed chronological overview of all the contributions to biomedical research based on experiments involving laboratory animals would be neither complete nor useful. Therefore, this section will start with a brief overview of past contributions of laboratory animals to medical progress and will highlight examples that have significantly influenced human health and reduced suffering. The use of animals in biomedical research dates as far back as the 1600s. During subsequent centuries, as scientific methods and research practices evolved, laboratory animals have continuously contributed to virtually all aspects of biomedical progress. This fact can be illustrated with history-making contributions from diverse areas of science, as represented in the following list: development of anesthetics diagnosis of infectious diseases (e.g., rabies, yellow fever)
84550.indb 2
11/11/10 6:23:57 PM
The Contribution of Laboratory Animals to Medical Progress
3
understanding of infectious processes leading to antimicrobial agents transplantation technologies (e.g., kidneys, heart) development of modern anesthesia and neuromuscular blocking agents development of vaccines (e.g., smallpox, polio, whooping cough, feline leukemia) cardiovascular medicine (e.g., open-heart surgery, coronary bypass procedure, heart transplant) development of cancer treatments development of effective drug therapies (e.g., for HIV, depression, leukemia, control of transplant rejection, ulcers)
Examples from the Past A few specific examples will underscore that many past accomplishments based on scientific observations or experiments performed on laboratory animals formed the foundation for medical progress, thus leading to some of today’s sophisticated health-science technologies. In 1877, the German scientist Robert Koch built upon observations that were already over a decade old: Anthrax could be transmitted from animal to animal. Koch’s animal experiments proved that the Bacillus anthracis bacterium caused a particular disease and that the isolated and purified bacterium from an initial host could cause the same disease in a new, second host. Hence, Koch’s postulates were born: four criteria designed to establish a causal relationship between a causative microbe and a disease (Koch 1877): The microorganism must be found in abundance in all organisms suffering from the disease, but not in healthy organisms. The microorganism must be isolated from a diseased organism and grown in pure culture. The cultured microorganism should cause disease when introduced into a healthy organism. The microorganism must be from the inoculated, diseased experimental host and identified as being identical to the original specific causative agent.
Koch’s observations gained general acceptance and helped lay the foundation for more intensive use of laboratory animals, especially for investigations of infectious diseases. Even today, over a century later, Koch’s postulates form a cornerstone for infectious-disease biology and research. Without animal research, most of the effective vaccines against infectious microbes or their toxins would not have been developed (Hendriksen 1996). Two major milestones were the independent developments of the first vaccines against smallpox and rabies— centuries-old human diseases that resulted in severe, potentially fatal illness or in 100% mortality, respectively. In 1798, English physician Edward Jenner noticed that people infected with cowpox were unaffected by smallpox. He conducted experimental vaccinations for the prevention of smallpox by inoculating with the cowpox virus (Jenner 1798). He found that persons inoculated with cowpox virus showed complete resistance to a challenge with the deadly smallpox virus. A century and a half later, Jenner’s smallpox vaccine formed the basis for the World Health Organization’s 1958 program of global eradication of smallpox. Without insights gained from the study of cowpox, the smallpox vaccine might have taken much longer to develop. In Paris, Louis Pasteur adapted the wild-type or street rabies virus to laboratory animals, resulting in a change in viral properties that today would be called attenuated virus strains. Pasteur and his colleagues subsequently developed concepts and experimental approaches that led to the first protective vaccination against rabies. Much of our knowledge about the structure of immunoglobulins or antibodies, molecules of central importance in immunology and host defense, was derived from extensive investigations into neoplastic plasma cells derived from mouse myelomas. Data on the Y-shaped protein
84550.indb 3
11/11/10 6:23:57 PM
4
Handbook of Laboratory Animal Science
structure and the potential for large-scale production of homologous antibodies were put forth by Kohler and Milstein in 1976. The hybridoma technique that these two scientists developed has provided a method of antibody production that is now widely utilized. The basic principle of hybridoma technology relies on the fusion of immortal myeloma cells with antibody-producing spleen cells harvested from a previously immunized mouse against the antigen of interest. Successful hybridoma clones will produce one type of target-specific or monoclonal antibody in unlimited quantities. Such monoclonal antibodies have been used for a wide range of applications, including diagnosis of pathogens, identification of physiological cell components, treatments of diseases, and purification of biological materials. More recently, the traditional production method of monoclonal antibody as mouse ascites is being replaced by alternative in vitro techniques that abolish the need for using live mice. Another application of our expanded understanding of the immune system is in the field of tissue transplantation, which was significantly advanced by experimental findings in mice. At the Jackson Laboratory, scientists performed pioneering work by restoring the health of a mouse with a blood disorder after performing a bone marrow transplant (Russell, Smith, and Lawson 1956). Working at the same institution, George Snell discovered genetic factors recognized by the immune system that determine the possibilities of transplanting tissue from one individual to another. This pioneering work on the concept of the H antigen and the major histocompatibility complex (Snell 1981) later resulted in the shared 1980 Nobel Prize in physiology and medicine. Specific Examples from the Past—Nonhuman Primates As noted previously, the eradication of poliomyelitis from the human population was dependent upon the development of an effective vaccine. Rhesus macaques were critical to this process, providing an ideal animal model to study the pathogenesis of the disease and to test the efficacy of candidate vaccines. Rhesus monkeys are currently used in the development and testing of candidate AIDS vaccines (see later discussion), although in greatly reduced numbers relative to previous use in the development of polio vaccines. Several types of viral hepatitis are of major importance as global human-health problems. Again, nonhuman primates are essential in the development of control methods for these conditions. Chimpanzees, susceptible to hepatitis B virus, were of critical importance in the development of an effective vaccine against this agent (Chisari 2000). The hepatitis B vaccine is now widely used and has helped bring this disease under control. As close genetic relatives of humans, chimpanzees are also susceptible to other human hepatitis viruses (e.g., hepatitis C virus). Because nonhuman primates are closely related to humans, they have been important in neurobiological studies in a wide range of fields, including perception, behavior, and basic neurologic studies (see “Present Role of Laboratory Animals—Nonhuman Primate Models” section). Many of these experiments have been of relatively long duration and require training of animals in controlled experiments that can help pinpoint the effects of specific treatments on various parts of the complex primate brain. Nonhuman primates are of special value in determining the addictive potential of specific compounds and in dissecting the mechanisms of neurobiological processes, such as addictions. Specific Examples from the Past—Nontraditional Species Several unlikely species are of significance in biological and medical studies. The armadillo, native to the southern United States, is one of few species susceptible to the mycobacterium that causes human leprosy. Thus, the animals can serve as sources of this agent and also can be used to identify mechanisms of pathogenesis and to assess potential therapies (Scollard et al. 2006).
84550.indb 4
11/11/10 6:23:58 PM
The Contribution of Laboratory Animals to Medical Progress
5
The chinchilla, a South American rodent raised for its pelt and used in the fur industry, has a large and accessible acoustic system. This system has been exploited to study basic mechanisms of hearing and to assess the effects of factors that are toxic or otherwise detrimental to hearing (Biebink 1999). Ferrets, now common as pets, are of importance in the study of influenza (Bodewes, Rimmelzwaan, and Osterhaus 2010), which continues to be a major threat to the human population. The development of efficacious vaccines is dependent upon a susceptible host and the ferret acts as such a host. Public Concerns Over the past decade, the use of cats and dogs in biomedical research has significantly diminished for a variety of reasons. First and foremost is the changing focus of much research to basic biologic questions that are best investigated in rodent models. However, there is also continuing and increasing public concern about the use of dogs and cats in research. A primarily urban public views these animals as members of the family and thus vocally opposes their use in research. Public concern has also focused on the use of nonhuman primates for research studies. Cultural differences contribute to these concerns, as reflected by intense and vocal objections to animal-based research in Europe and the United States compared to other regions. It is important to point out that dogs, cats, and nonhuman primates now make up less than 1% of all research animals. In part, this is a result of the active pursuit of the three Rs and the increased development and use of alternatives. It is worthy of note that dogs and cats (and their owners) have benefited greatly from research done to improve human health. Numerous therapeutic technologies and drugs developed for human medicine are also of critical importance in the veterinary arena, which itself does not have the financial resources to support such research.
Present Role of Laboratory Animals Mouse Models Genetically Altered Mice and Knockout Mouse Projects Today, rodents such as mice and rats still account for the vast majority of all laboratory animals (Maher 2002). In part, this is due to the fact that they offer advantages to scientists that are beyond their small size and relatively short reproductive cycle. Laboratory mice used in today’s research settings are very well characterized; of particular importance are the facts that their entire genome has been sequenced (Mouse Genome Sequencing Consortium 2002), they are genetically close to humans (Maher 2002), and they exist in lines and strains that are inbred (Morse 1978), resulting in animals that are genetically identical to their parents and siblings. Further, these strains can carry different mutations, both spontaneous and artificially induced by genetic manipulations. This has resulted in a collection of mouse strains that has grown enormously over time: During the early part of the twentieth century, several hundred strains of mice were generated (Maher 2002). With the development of sophisticated new technologies, this number increased to thousands of strains between the early 1970s and the present. At that time, the Mouse Genome Database (http://www.nervenet.org/main/dictionary.html) had a listing of approximately 6,000 strains. By August 2008, that number had grown to 12,370 (http://www.informatics. jax.org), with new strains being added at an accelerating rate. This escalating growth is a result of several mutagenesis programs and gene targeting and trapping efforts that aim to produce hundreds
84550.indb 5
11/11/10 6:23:58 PM
6
Handbook of Laboratory Animal Science
of mouse strains and over 200,000 embryonic stem cell (ESC) lines for generation of new mouse strains useful for biomedical research in the future. Spontaneous mutants, discovered as deviant animals among the hundreds of thousands of mice born in laboratory breeding facilities, have resulted in some well characterized and successfully utilized models. After careful phenotyping and mapping of their mutations to a chromosome locus, these models have served scientists well. As mentioned, most of the spontaneous mutation mouse models are identified in breeding colonies because of their abnormal phenotype. In most cases, a natural or spontaneous mutant mouse displays a phenotype that resembles a human disease, either phenotypically or clinically, or both. Some specific examples of spontaneous mutant models are described next, including mouse models with completely absent leptin and models for human Marfan syndrome and synpolydactyly. Not just one but many genes influence the onset and outcome of obesity, one of our society’s greatest challenges. However, the leptin gene, with many different mutations, has been identified as one major contributor, and models with mutations in this gene have been used to study the pathogenesis of obesity and to test therapeutic interventions. A nonsense mutation in codon 105 resulting from a C to T point mutation results in a complete absence of leptin in homozygous mice. The obesity of these mice is characterized by an increase in both number and size of adipocytes. Phenotypically, these mice exhibit hyperphagia, hyperglycemia, glucose intolerance, elevated plasma insulin, reduced fertility, and impaired wound healing. To date, this mouse is the most frequently studied mouse model for human obesity and diabetes (Ioffe 1998). A mutation in B10.D2 congenic mice, a 30–40 kilo base tandem duplication within a gene named Fbn1, results in a large frame shift in transcription. This mutation yields mice that show many of the typical features of human Marfan syndrome, including tight skin because of changes in collagen in the dermis, abnormal cutaneous microfibril morphology, and abnormally small collagen fiber diameter. The mutant mice also have an increased lung capacity accompanied by emphysematous lesions in the lungs as well as cardiovascular abnormalities (Green 1976; Gayraud 2000). Another spontaneous mutation, a 21 base-pair in-frame duplication within the poly-A track of the Hoxd13 gene, is genotypically very similar to the genetic alteration observed in patients with synpolydactyly. Both clinical phenotype and the molecular basis of the abnormality are nearly identical in the mouse model and the human condition. Synpolydactyly describes webbing and/or the presence of an extra number of fingers or toes. The mouse model displays shortened, fused, and duplicated digits in both rear and front feet. Because of the involvement in ossification, delayed joint formation, and reduced rate of proliferation of chondrocytes, this mutant mouse has also been used to study mammalian embryo development (Johnson 1998; Albrecht 2004). Although such spontaneous mutants have contributed in significant ways in the past and present, today’s focus has shifted to genetically engineered mouse strains. Frequently, they are also referred to as genetically modified, genetically altered, or mutant strains and the methods to generate them are called transgenic technology. Specifically, the ascent of these new powerful genetic methodologies applied to molecular biology allowed scientists to delete or insert genes, and attention has shifted from the original inbred strains developed by Castle and Lathrop (Maher 2002) almost a century ago to technologies that produce tissue- and disease-specific mouse models. These transgenic technologies give scientists the tools to insert foreign genes into the mouse genome and allow them to gain insight into the functions of those genes. The refined step of the early technique was the “knockout” strategies, also called targeted mutagenesis or gene replacement technologies. These permit scientists to inactivate existing genes by replacing or disrupting them with a newly introduced artificial piece of DNA. Seminal to this revolutionary development were studies conducted by three scientists who received the 2007 Nobel Prize for their achievements: Drs. Mario Capecchi, Oliver Smithies, and Martin Evans
84550.indb 6
11/11/10 6:23:58 PM
The Contribution of Laboratory Animals to Medical Progress
7
(http://nobelprize.org/nobel_prizes/medicine/laureates/2007/press.html; http://nobelprize.org/nobel_ prizes/medicine/laureates/2007/announcement.html). Capecchi, Smithies, and Evans developed the immensely powerful technology that allowed scientists to create animal models of human disease in mice using this “knockout” approach. Other scientists are now applying this technology, in addition to its application in mice, to other animal model species. For example, investigators have developed swine in which the gene for production of the cell surface antigen, galactosyl-alpha-1,3-galactose, has been knocked out. Organs from these genetically modified swine are expected to be useful for xenotransplantation studies, since the antigen is normally responsible for hyperacute rejection of swine organs by primates. The organs derived from the knockout swine lack the antigen (Kolber-Simonds et al. 2004). Today, these knockout technologies are broadly employed to produce transgenic mice. Starting with embryonic stem cells (ESCs) harvested from an early-stage mouse embryo 4 days after fertilization, scientists use in vitro methods to insert the foreign DNA into the nucleus of the ESCs. The first approach, called gene targeting, relies on homologous recombination. The foreign piece of DNA shares a part of its sequence (homologous sequences) with a sequence in the targeted ESCs’ genome. These homologous sequences flank the gene of interest both upstream and downstream. The cell’s own machinery will exchange the corresponding portion of the existing gene or the entire gene for the newly introduced artificial piece. By design, the artificial piece is inactive or under the control of an element that allows external control of its activation state; additionally, the artificial insert contains a genetic tag or reporter that allows the scientist to track the insertion (e.g., confirm its presence, location, and orientation within the genome). The second approach is called gene trapping. Instead of targeting a specific gene of interest, as described in the first approach, scientists employ a random strategy using artificial DNA pieces with reporter genes that insert arbitrarily into any genes in the host ESCs’ genome. The inserted foreign piece of DNA prevents the proper processing of the host cells’ genes, therefore knocking out specific gene functions. Both gene targeting and gene trapping have obvious strengths. In gene targeting, the gene that is knocked out is known and efficiency is generally high. In gene trapping, knowing a specific gene [sequence] is not required, and the use of a single type of vehicle or vector to transport the artificial DNA into the ESCs’ nucleus results in high-throughput capacity. Large-Scale Mouse Programs The powerful utility of these knockout mice is obvious: Inactivation of a given gene in a knockout mouse model provides valuable clues to the functions of that gene, specifically for genes that are shared between humans and mice. Therefore, observing and analyzing the characteristics of knockout mice afford scientists access to information that can help understand specific gene functions or disease pathways. Recognizing the power of the knockout mouse technology and its general widespread benefit to biomedical scientists, discussions on an international level were held in the fall of 2003 at the Banbury Conference Center at Cold Spring Harbor Laboratories. The participants began exploring the feasibility of a coordinated effort to produce knockout mouse strains to generate a comprehensive and publicly accessible resource of mouse ESCs containing knockout or null mutations in every protein-coding gene in the mouse genome (Austin 2004). The driving force behind this effort was the recognition that, although a significant proportion of the approximately 22,000–25,000 mouse genes have been knocked out and published, many of these knockout mouse models are not readily available to scientists other than to those in whose laboratory the knockout mouse was created. It was recognized that a coordinated effort would save time and money, and ensure that the knockout mouse models would be made available in a standardized format (e.g., mouse strain background, genotype testing, and control of contamination). Thus, the foundation of several knockout mouse projects was born.
84550.indb 7
11/11/10 6:23:58 PM
8
Handbook of Laboratory Animal Science
Three independent yet collaborating efforts have been launched to address the original challenge posed during the Banbury Conference: the Knockout Mouse Project (KOMP), funded by the National Institutes of Health (NIH) in the United States; the North American Conditional Mouse Mutagenesis (NorCOMM) Project, funded by Genome Canada and its partners; and the European Conditional Mouse Mutagenesis (EUCOMM) Program, funded by the European Union (International Mouse Knockout Consortium 2007). In a collaborative effort among different institutes and centers within the NIH, a 5-year mutant mouse resource initiative (>$50,000,000) was started in 2006; it aims to (1) use gene targeting to make a resource of null alleles marked with reporters, (2) support a repository to archive the products of this resource, (3) develop improved and robust ESCs on the inbred mouse strain C57BL/6, and (4) implement a data coordinating center that allows all scientists easy access to data relevant to this effort (www.komp.org). The two partner projects in Canada (NorCOMM; http://norcomm.phenogenomics.ca/) and Europe (EUCOMM; http://www.eucomm.org/) state their goals as “a large-scale research initiative focused on developing and distributing a library of mouse ESC lines carrying single conditional knockout mutations across the mouse genome” and “a collection of up to 20,000 mutated genes in mouse C57BL/6N ESCs using conditional gene trapping and gene targeting approaches,” respectively. During the initial period in which the three projects have been funded, numerous joint meetings have been held. Strong efforts are in place to collaborate and to coordinate activities in order to avoid duplication of efforts. The three projects have agreed to share the gene lists and data in order to help with the coordination. Ideally, resources produced by one project would be available to scientists on a different continent, thereby enabling scientists simply to order all mouse strains from the local site without the potential difficulties associated with international transport. The future will tell if this will become reality and if sharing of mutant mouse resources can happen across international borders. Repositories A conservative estimate lists the number of existing mutant mouse lines available for biomedical research scientists at over 4,000. Additionally, over 27,000 ESCs have been generated with the potential to produce mutant mice. These resources are available to scientists for a distribution fee only, signifying the realization of earlier mouse genetics research “dreams.” Recognizing that this large number of mutant mouse strains would be difficult for not-for-profit and for-profit organizations to make available to scientists, the NIH established repositories at different locations across the United States. In addition to archiving the mouse strains and ESCs generated by large-scale projects, the importation of mutant mouse strains made by scientists in individual laboratories represents a major challenge. Over the past 10 years, significant progress has been achieved in establishing several mutant mouse repositories and associated catalogues. These resources offer scientists a wide selection of mouse models for their research and provide expert advice and assistance with questions related to mouse biology and reproduction, cell culture techniques, molecular biology, and the selection of the most appropriate model or approach for research projects. Similar resources and repositories have been established in Canada, Europe, and Japan (see http://www.mmrrc.org/about/resources.html and references therein). To promote mouse resource interactions on an international level, the Federation of International Mouse Resources (FIMRe) was established in 2005. FIMRe is a collaborating group of mouse repository and resource centers spanning four continents with the collective goal of archiving and providing strains of mice as cryopreserved embryos and gametes, ESC lines, and live breeding stock to the research community (http://www.fimre.org/).
84550.indb 8
11/11/10 6:23:58 PM
The Contribution of Laboratory Animals to Medical Progress
9
Specific Mouse Models Thousands of mutant mouse models are available today, mainly through the repositories described earlier; many have successfully contributed to advancements of our knowledge in biomedical science to the next level. A few such models are showcased here for demonstration purposes, but many more can be reviewed in the literature and in electronic catalogs (e.g., http://wmc. rodentia.com; http://www.jax.org; http://www.taconic.com; http://www.mmrrc.org). Rett syndrome is a neurodevelopmental disorder that occurs once in 10,000–15,000 births. Its clinical features include a reduced growth rate of the head, hands, and feet that can result in microcephaly. Frequently, patients show stereotypic, repetitive hand movements combined with cognitive impairment, problems with socialization, and seizures. Rett syndrome is caused by sporadic mutations in the gene called MEP2, located on the X chromosome. It almost exclusively affects females; male fetuses with the disorder rarely survive to term. Scientists created Mecp2-deficient mice by generating a null mutation or by knocking out the gene in the brain only (Chen 2001; Guy 2001). Both mutant mice exhibit phenotypes that share some key features with Rett patients. Mecp2-null mice are normal until about 5 weeks of age, when they begin to develop disease symptoms. Their brains show substantial reduction in both weight and neuronal cell size, but no signs of neurodegeneration. The cystic fibrosis knockout mouse provides an outstanding animal model for cystic fibrosis (CF). Human CF is caused by mutations in the gene that encodes the CF transmembrane regulator (CFTR). A knockout mouse in which both copies of the CF gene have been inactivated dies due to intestinal obstruction during the first month of life. This defect can be repaired by introducing a gene that contains the human CFTR coding sequence under control of a rat intestinal fatty-acidbinding protein gene promoter. The human CFTR gene is therefore expressed in the intestinal epithelial cells of the mice, leading to normal intestinal development and survival. These mutant mice should also be useful for examining the effects of the pulmonary CF phenotype, while correcting the effect of mutation in the gastrointestinal tract (Zhou et al. 1994). In addition to mouse models that can be generated by deleting or altering one gene, mouse models exist that help scientists understand pathways and processes of human disease that are influenced by different mutations or mutations in different genes. Such models cannot be generated by deleting a single gene. Specifically, hereditary hearing impairment can be caused by different mutations in the same gene or on different genes; over 70 protein-coding chromosomal genes and tRNA- or rRNA-coding mitochondrial genes have been identified as contributory (http://webhost.ua.ac.be/hhh/). A better understanding of the function of these protein-coding and regulatory genes is needed in order to elucidate inner ear development and, thereby, the mechanisms by which specific mutations lead to the condition of hereditary hearing impairment. With a prevalence of 1 in 1,000 cases, hearing impairment or hearing loss is the most common hereditary disability in children. Therefore, mutant mouse models that help clarify the contributions of these different genes serve as ideal models to identify and study the role of these genes in the function of the inner ear. A relatively recent review summarizes many different mutant mouse models and associated achievements (Friedman 2007). Finally, another type of mouse model permits scientists to study a human disease or condition in a species that is not affected by or susceptible to the condition or the disease. In these cases, scientists produce a transgenic genome within the mouse that subsequently expresses the gene of interest. One such example is the study of prion disease in mice. A number of animal and human disorders are recognized as being caused by prions, including bovine spongiform encephalopathy, scrapie, and Creutzfeldt–Jakob disease. An important observation related to these diseases is that the inoculation of brain material into individuals of the same species typically reproduces the
84550.indb 9
11/11/10 6:23:58 PM
10
Handbook of Laboratory Animal Science
condition, while passage to another species is generally ineffective. Most of our understanding of this block or barrier comes from studies of transgenic mice that revealed that the sequences of PrPsc (prion protein in its conformation-changed, disease-associated form) in the inoculum and the host’s PrP should be isogeneic to result in transmission (Pruisner 1990). Rat Models For over 150 years, various strains of the laboratory rat (Rattus norvegicus) have been successfully used in biomedical research for investigations into many human diseases and pathologic conditions, including diabetes, hypertension, heart failure, immune-mediated disorders (autoimmune diseases), and cancer (Jacob 1999). Over 1,000 different rat strains, including some 500 inbred strains, are listed in the Rat Genome Database (RGD; http://rgd.mcw.edu); all of these strains are available to scientists. In addition to the inbred lines, many of these strains are transgenics, spontaneous mutants, or complex trait strains. Similarly to the repositories developed for mouse models, several centralized resources have been established to collect, preserve, and supply high-quality rat models to requesting scientists. In the United States, the Rat Resource and Research Center (RRRC; http://nrrrc.missouri.edu/) maintains over 300 strains. An even larger scale repository exists in Japan: the National Bio Resource Project for the Rat (NBRP-Rat; http://www.anim.med.kyptp-u.jp/NBR/), with approximately 570 strains archived. Two smaller European repositories, holding some 140 strains, exist in Germany (http://mh-hannover.de/2652.html) and the Czech Republic (http://www.euratools.eu). During the past 10–15 years, information and knowledge on rat genomics has increased, primarily for three reasons. First, the analysis of a comprehensive rat sequence (Gibbs 2004) that enabled researchers to conduct a careful comparison between the human, mouse, and rat genome resulted in detailed new information on genome evolution. Second, there has been a significant increase in genetic markers for the rat, including millions of single nucleotide polymorphisms (SNPs). Based on these newly established and accessible resources containing well-characterized SNPs for quantitative trait loci (QTLs), scientists have been able to construct high-density genetic maps and associated disease gene mapping tools. Last, and clearly not least, the recent publication of a report describing the successful culture of germline-competent rat ESCs (Shinobu 2008) represents a significant breakthrough certain to open many new avenues. Specifically, these two new rat ESCs lines display undifferentiated characteristics, maintain their pluripotency and a normal karyotype, and are capable of producing chimeras. These new cells will provide scientists the long-awaited tool they need to produce new, state-of-the-art transgenic rat models to study human disease; explore new avenues in regenerative medicine; and advance applications in pharmaceutical and therapeutic research. Numerous rat models have been developed and successfully utilized in biomedical research, including models for human conditions such as rheumatoid arthritis (Oelzner et al. 2010), polycystic kidney disease (Wu et al. 2009), breast cancer (Ting et al. 2007), and multiple sclerosis (Thessen et al. 2009). Specific information for these and other models can be obtained from the Rat Resource and Research Center Web site (http://www.nrrrc.missouri.edu/) and the rat genome database (http://rgd.mcw.edu/wg/portals?100). These recent accomplishments may advance the rat to a more prominent position as an animal model for studies in biology and pathology used to understand human diseases, particularly those with complex genetic backgrounds. Rat models of the future will be used in translational research to explore the basis of clinical disorders, as well as the influence of genetic and environmental factors, thus increasing success rates for the development of new drugs to cure human diseases.
84550.indb 10
11/11/10 6:23:59 PM
The Contribution of Laboratory Animals to Medical Progress
11
Nonhuman Primate Models Of all laboratory animals, nonhuman primates (NHPs), such as old-world monkeys and chimpanzees, are evolutionarily closest to human primates and thus exhibit many aspects of physiology and development that are relevant to understanding human biology and disease. NHPs also exhibit many behaviors, both normal and aberrant, that are similar to those of humans. Therefore, NHPs are often critical animal models for translational research. A classic example is the extensive use of rhesus monkeys in the development of polio vaccines (Salk and Salk 1984; Sabin 1985). However, there are inherent difficulties in the use of NHPs for medically related translational research. NHPs require expensive, complex husbandry, which is accomplished most readily at large primate centers. For this reason, the U.S. National Institutes of Health (NIH) established a National Primate Research Center (NPRC) Program more than 40 years ago (see http://www.ncrr.nih.gov/). Currently, eight U.S. NPRCs house more than 25,000 NHPs, including about 16,000 rhesus monkeys. NIH also supports more specialized primate centers, including those that house new-world monkeys, such as owl monkeys and squirrel monkeys. In addition to facilities in the United States, primate centers are located in Europe, Japan, China, India, Indonesia, and Africa. The cost and complex husbandry of NHPs places a premium on careful experimental design and consideration of the statistical power of experiments. Virtually all aspects of human medicine and physiology have been studied in parallel in NHPs, including infectious disease; neurobiology; metabolic conditions, such as diabetes and obesity; aging; transplantation; and reproductive biology (see following discussion). In the process of drug development, NHPs are used to test safety and efficacy of new approaches, in addition to toxicology. Some major current examples of the use of NHPs for medically related investigations include AIDS, hepatitis B and C, neurobiology, and genetic technology research. AIDS The failure of AIDS vaccine trials in 2008 emphasized the need to understand the pathogenesis of HIV viruses better. Monkeys are very important experimental hosts for such investigations. Many parameters can be controlled in experiments using monkeys that cannot be controlled in studies of human populations, including variation in viral sequences, the genotype of the host, and the mode of infection. Early stages of infection can be studied in monkeys, and tissues such as the gut and brain can be sampled readily, unlike in studies using human patients. Therefore, the monkey system not only permits studies of lentiviral infections occurring through sexual or blood-borne routes, but also has contributed to studies of mother-to-child transmission and neuro-AIDS—again, difficult problems to investigate with human patients (reviewed in Lackner and Veazey 2007). Studies of AIDS pathogenesis have been advanced by the development of the simian immunodeficiency virus (SIV) monkey model for AIDS. Many African monkey species in captivity and in the wild are infected by SIVs and may develop high titers of virus in the blood, but do not ultimately progress to AIDS. In contrast, Asian monkeys, such as rhesus macaques, can be infected with certain strains of SIV and then rapidly develop AIDS. An SIV strain that causes AIDS in rhesus macaques (termed SIVmac) was originally identified at the New England NPRC in the 1980s (Letvin et al. 1985), but appears to have originated in the primate centers in the 1970s (Apetrei et al. 2006). Like one of the human AIDS viruses, HIV-2, SIVmac originated from sooty mangabeys, an African monkey species. Cross-species conversion of retroviruses to a virulent state is extremely rare in captive NHP colonies. It has been hypothesized, with good supporting evidence, that the virulence of SIVmac was enhanced when it was inadvertently serially passed through rhesus macaques during experiments examining induction of prion diseases in monkeys (Apetrei et al. 2006).
84550.indb 11
11/11/10 6:23:59 PM
12
Handbook of Laboratory Animal Science
SIVmac, related virulent SIVs, and modified versions that contain specific human HIV sequences (viruses termed SHIVs) have been used to study many aspects of AIDS using NHP models. Some of the key features of the SIV monkey model include: • SIV and HIV are very similar in genetic structure and biological properties (reviewed in Desrosiers 1990). • Infection of rhesus macaques with SIVmac causes a fatal disease remarkably similar to human AIDS in an experimentally useful time frame (3 months to 1–2 years, depending on the viral strain and types of animals that are used). • Both SIV and HIV interact with two surface molecules in combination in order to enter a target cell: the CD4 molecule and a chemokine receptor (the most common being CCR5 or CXCR4). • Both SIV and HIV cause a steady decrease in CD4+ lymphocytes, leading to AIDS. • A major source of loss of immune cells early in infection occurs in the mucosal immune system, a large part of which resides in the gut. This was first discovered in monkeys and then verified in humans; it has major implications for the design of vaccines and therapeutic strategies (reviewed in Lackner and Veazey 2007). • A small percentage of HIV-infected humans and SIV-infected monkeys are able to control the infection, despite high levels of virus in the blood early in infection. These individuals are termed “long-term nonprogressors” or “elite controllers.” There is great interest in understanding this phenomenon, which can be analyzed in detail in monkeys (reviewed in Lackner and Veazey 2007). • Several studies have suggested that the genetic composition of the host can influence infection with HIV or SIV. Recent investigations demonstrate that specific alleles (different forms of genes) of the major histocompatibility complex (MHC) in both humans and monkeys are associated with “elite controller” status (Zhang et al. 2002; Loffreo et al. 2007). This phenomenon can be studied in detail in monkeys because of the ability to control precisely the type of infecting virus used (through clonal preparations of SIV), the route of infection, and the genetics of the animal.
Hepatitis B and C Over 500 million people worldwide are infected with the hepatitis B (HBV) and C (HCV) viruses, which cause acute and chronic inflammatory liver disease and can lead to hepatocellular carcinoma. Development of vaccines for these agents has been a high priority. The chimpanzee is the only natural animal host for these viruses and was central to the successful development of a vaccine for HBV. HCV was initially cloned from the plasma of an infected chimpanzee, leading to its identification in humans (reviewed in Bukh 2004). To find an alternative to chimpanzees, considerable effort has gone into the development of cell lines and transgenic mice in which aspects of the biology of HCV infection can be analyzed. However, a successful HCV vaccine has yet to be developed and continued research is needed to achieve this goal. Neurobiology NHPs have played a major role in many areas of the investigation of the nervous system. The brains of NHPs are organized very similarly to those of humans and NHPs have complex cognitive capabilities and social interactions. The development of noninvasive, high-resolution magnetic resonance imagery (MRI) methods of mapping brain structure and function has given rise to a new generation of specific digital brain atlases. These atlases enable quantitative neuroanatomical integration of gene expression and neurophysiological and behavioral data with an order of magnitude of greater precision than was previously possible (Bowden and Martin 1977). Some current areas in neurobiology of particular emphasis include the following (for a review, see Capitanio and Emborg 2008): • Neurodegenerative diseases such as Parkinson’s and Alzheimer’s diseases. Parkinsonism can be induced in monkeys using the neurotoxin MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine),
84550.indb 12
11/11/10 6:23:59 PM
The Contribution of Laboratory Animals to Medical Progress
•
•
•
•
13
which specifically causes dopaminergic neuronal loss in the substantia nigra, as occurs in naturally occurring Parkinson’s disease in humans. Monkeys with MPTP-induced Parkinsonism respond to anti-Parkinsonian drugs and display the same side effects as humans (Capitanio and Emborg 2008). Thus, monkeys serve as a model for modifying existing drug regimens or designing new ones. The MPTP model of Parkinson’s disease has been used as the basis for several studies using gene therapy or tissue replacement approaches for treating the disease. Huntington’s disease (HD). HD is a genetic disease caused by the increase in dinucleotide repeats in a specific region of the Huntingtin gene. The resulting mutant protein increases the rate of neuronal cell death in certain regions of the brain. Investigators have recently derived monkeys that contain mutant Huntingtin genes and that show the phenotype of the disease (Yang et al. 2008). These monkeys should be very useful for investigating disease pathogenesis and for testing therapeutic strategies. This is the first monkey model that has been genetically engineered to have a human disease. Addiction. Monkeys show addictive behaviors very similar to those of humans to a wide range of substances of abuse, including cocaine and alcohol. Studies using monkeys have identified many aspects of the neurotransmitter circuits involved in addiction and provide leads for drugs that can be used to influence these circuits as a means of combating addiction (Capitanio and Emborg 2008; Howell and Murnane 2008). Neuro-AIDS. As discussed before, the SIV monkey model very closely replicates human AIDS, including its neurological aspects. Studies using monkeys have demonstrated that the perivascular macrophage appears to be the infected cell in the brain and that the degree of infection is directly related to viral load (Williams et al. 2001). It is virtually impossible to examine this type of question in HIV-infected humans. Behavioral genetics. Since monkeys exhibit many of the same behaviors as humans, there is great interest in correlating behaviors with genetic variation as a means to identify similar genetic effects in humans. An example of this type of investigation is the demonstration that infant rhesus monkeys show heritable variability in anxious behavior (Williamson et al. 2003). The investigators used tests modeled after those used in human studies. Anxious behavior is considered to be correlated with the probability of developing mood disorders (and is therefore termed an endophenotype) in humans. It is now possible to characterize the genetic components of the endophenotype in monkeys, using genetic mapping techniques and the rhesus genomic DNA sequence (see following discussion). In turn, these results can be correlated with human studies to understand the origins of mood disorders better and to design early interventions in humans.
Genetic Technologies and NHPs Investigating the relationship of genotype to phenotype is critical for understanding most human diseases. For example, there appear to be many genes that affect susceptibility to a given complex human disease or condition, such as diabetes, Crohn’s disease, heart disease, and obesity (Eleftherohorinou et al. 2009; McCarthy 2008). The precise effect of a potential deleterious genetic variant on an individual will also depend on the environment. NHPs are important animal models for understanding these genetic and gene–environment effects because they are genetically similar to humans and because their environment can be controlled. Development of requisite genetic tools and assays is necessary to realize this potential. Considerable progress has been made in the development of genetic tools for NHPs in the past several years. Investigators have derived fairly high-resolution genetic maps for the baboon (Cox et al. 2006) and rhesus macaque (Rogers et al. 2006). The maps have been used to identify genetic regions (termed quantitative trait loci [QTLs]) that influence conditions such as heart disease, atherosclerosis, diabetes, and osteoporosis. In order to exploit this mapping data fully, as well as for many other uses, a full genome sequence is necessary. The full genome sequence for the rhesus macaque was achieved in 2007 (Gibbs et al. 2007), and whole genomic sequencing of the baboon, marmoset, and other NHPs is currently in progress (see http://www.genome.gov/25521747).
84550.indb 13
11/11/10 6:23:59 PM
14
Handbook of Laboratory Animal Science
Another important aspect of genetic analysis is the investigation of gene expression patterns in tissues of normal and diseased animals. Microarrays have become a method of choice for this type of experiment and have been developed and commercialized for rhesus monkeys (Spindel et al. 2005; Wallace et al. 2007). Fish Models Increasing attention has been devoted to studies using fish models, which are attractive in biomedical research because many species produce a large number of eggs very frequently. In most instances, these eggs are transparent and thus early developmental stages and events can be observed and closely monitored. Investigators can efficiently detect the effects of a variety of chemicals or other physical stimuli, and the pathogenesis of the resulting abnormalities can be dissected in a controlled environment. A single pair of fish can produce an extraordinary number of progeny, allowing scientists to test targeted approaches in a large-scale and high-throughput manner. The major types of fish used in biomedical studies have been derived from species formerly based in the aquarium trade; the most frequent and numerous species is the zebrafish (Danio rerio). Scientists use zebrafish to study many processes, including early embryonic development, developmental pathways, cancers, and blood disorders (Zhu and Zon 2002). However, new areas of interest to biomedical scientists are explored and described continuously. Furthermore, advances in fish genomics, combined with the modern transgenic technologies described before for rodents, have enabled scientists to generate large numbers of mutant zebrafish. In the United States, a national resource—the Zebrafish International Resource Center (ZIRC) (see http://zebrafish.org)—serves as a central repository for both wild-type and mutant zebrafish strains and provides scientists with access to almost 1,000 different strains of zebrafish. Strains of wild-type fish carrying mutations and transgenes are available to the research community and can be obtained as live adults, embryos, or frozen sperm for use in other laboratories. However, fish-based investigations pose unique challenges. Water quality is of paramount importance, and intercurrent diseases that affect the various fish populations in a colony or resource are often not well defined. Advances continue to be made in these areas, especially in large laboratory settings. Furthermore, significant progress has been made in the area of diagnostic fish pathology and disease, but continued efforts are still necessary in this area of research. Other Animal Model Species In addition to the three major groups of laboratory animals discussed previously (rodents, nonhuman primates, and zebrafish), other species have served and will continue to serve an important function as laboratory animal models. Their contributions as comparative and complementary model species are critical to the advancement of scientific knowledge in many different research areas. Research investigations using the fruit fly Drosophila melanogaster have contributed in the past and will continue to contribute to major advances in biomedical knowledge. The deciphering of the Drosophila genome (Adams et al. 2000), the high degree of conservation of many proteins, and the cellular and developmental pathways between insects and vertebrates—combined with the Drosophila’s extensive set of well-developed genetic tools (http://flybase.org/)—make this species one of the foremost disease model organisms for investigations into many fundamental problems in biology. As with many other nonmammalian species, the costs of maintaining individual animals are greatly reduced and large numbers of individual flies can be used in studies. Similarly, the small nematode worm Caenorhabditis elegans has served as a research animal model in the past and continues to contribute in significant ways in current investigations. Biomedical investigators’ continued research on the genetics of C. elegans has resulted in enhanced understanding of embryogenesis, morphogenesis, development, nerve function, behavior, and a number of
84550.indb 14
11/11/10 6:23:59 PM
The Contribution of Laboratory Animals to Medical Progress
15
age-associated neurodegenerative diseases, including Alzheimer’s, Parkinson’s, and Huntington’s diseases (Culetto and Sattelle 2000; Shim and Paik 2010). Other nonmammalian species have more recently gained prominence as research animal models and their use and importance have expanded over the past decade. Dedicated resources of these species allow scientists access to well-characterized, disease-free animals. Among them are cephalopods (including squid, cuttlefish, and nautilus), swordtails and platyfish (Xiphophorus sp.), medaka, and the sea snail (Aplysia). Large numbers of inbred stocks of animals have been developed and are used in genetic studies, including determination of those changes that correspond to genetic abnormalities in humans and other mammalian populations. Additionally, they are used in large studies of chemical mutagenesis and carcinogenesis. The African clawed frogs (Xenopus sp.) comprise another aquatic animal genus that is of considerable importance in studies of embryogenesis, early development, and factors that can affect these events (Amaya 2005; Beck and Slack 2001). Embryos of Xenopus are accessible for surgical and chemical manipulations, and the effects of such manipulations on specific cells can be determined. Because events in early development in Xenopus occur more slowly than in other species, these processes can be more easily observed and dissected. These advantages enable scientists engaged in studies on early embryonic development to perform their studies under these favorable conditions and to address experimental hypotheses more directly. Future Directions for Laboratory Animals in Biomedical Research The current uses of animals to understand basic physiological properties and to test drugs should continue into the future. In some cases (e.g., toxicology), cell-based systems may supplement and perhaps eventually supplant the use of animals. In many cases, though, there will be no substitute for whole-animal studies because of the involvement of multiple tissue and organ systems in both normal and aberrant physiological conditions. In the recent past, technologies based on new genetic knowledge have revolutionized our understanding of both animal and human biology. Methods for characterizing the complement of genes in a given animal and patterns of gene expression in specific tissues and organs have become relatively facile. Informatics systems have developed in parallel, making information accessible to most researchers. One striking feature of this revolution in genetic information is that one type of data invariably leads to the development of technologies to gain more sophisticated insights and to ask new questions. As an example, catalogs of gene and mRNA sequences have led to the development of the microarrays used to measure differential gene expression. This trend of developing technologies that progressively build on one another both to ask and to answer increasingly sophisticated questions about animal and human systems should continue. A few examples are discussed next. Genomics Whole genome sequences are now available for the major laboratory mammals, including mice, rats, and rhesus monkeys, as well as for important experimental invertebrates, such as Drosophila and C. elegans. The zebrafish genome sequence and many other NHP sequences, such as that of baboons, should be available in the near future. Genome sequences make it possible to catalog all of the genes in an organism and to compare the organization and content of the genome with those of humans and other animals. Furthermore, the sequences can be used to design high-throughput assay platforms for genetic and physical mapping (as is currently done for humans and mice) and for characterizing gene expression patterns using microarrays. New, very high-throughput DNA sequencing technologies (often termed “next-gen” technologies) are in an advanced state of development and should greatly decrease the cost and increase the speed of genomic sequencing.
84550.indb 15
11/11/10 6:23:59 PM
16
Handbook of Laboratory Animal Science
Therefore, within the next several years, the research community should have not only “index” sequences, as are currently available (derived as a composite from a small number of individuals), but also access to a large amount of high-quality data on sequence variation of individual humans and animals. These data should be directly comparable with phenotypes of interest, such as disease states. As has already occurred for genomic sequencing, these advances will require parallel improvements in informatics capabilities, not only to store data, but also to make data accessible (in a user-friendly fashion) to those who are not genome specialists. Proteomics, the Interactome, and Systems Biology The genome sequence is a catalog of the genes that an organism contains. Microarray experiments provide information on sequences that are expressed differentially at the level of mRNA in various tissues or disease states (Schena et al. 1995). The next steps are to understand how differences in mRNA levels are reflected in protein levels (proteomics) and how these proteins interact with each other (analysis of the interactome). These studies represent a subset of the discipline of systems biology. It is necessary to understand the cell at the systems biology level because both normal and aberrant metabolic processes, such as disease states, function at the protein level. A classic example of the importance of understanding cellular networks based on protein–protein interaction is the delineation of the many signal transduction pathways in various cell types. These pathways are the molecular circuits by which signals on the outside of cells (e.g., hormones) cause a cascade of events that influence gene expression in the nucleus or affect other metabolic processes. Protein interactions—detailed maps of the interactome—are being developed (Li et al. 2004, reviewed in Figeys 2008) based on the revolution in genomics and improvements in methods for detecting and quantitating proteins in laboratory animals and in humans. The significance of the interactome resides in the fact that most proteins affect cellular functions through their interaction with other proteins or participation in macromolecular complexes and that diseases and drugs often affect these interactions. The application of genomics, proteomics, and informatics is giving rise to comprehensive, Webaccessible, anatomically indexed knowledge bases that foster communication and theoretical integration across the diverse disciplines of genetics, physiology, and behavior, as well as across animal species ranging from insects to primates. Knockout and Knockdown of Gene Expression The use of knockout mice to understand the function of particular genes has been discussed previously. The systems used to construct knockout mice have become increasingly sophisticated, and it is now possible to control the time and place in which the expression of a given gene will be abolished in the knockout animal. Progress in this field has been greatly facilitated by the development of inducible gene expression systems and the discovery of tissue-specific promoter sequences in the transcriptional control regions of mammalian genes. The use of the knockout strategy at this high level of sophistication is currently possible only in mice, although it would be expected to be extended readily to rats. Because knockout animals can only be produced on a large scale in rodents, there is great interest in using other techniques that can “knock down” (decrease) gene expression in other animal models. Many of these approaches make use of “antisense” RNAs that can bind to specific mRNA sequences and inhibit their expression in the cell, thus decreasing the level of gene expression specific to the mRNA (reviewed in McManus and Sharp 2002; Kim and Rossi 2007). Antisense RNAs can be delivered as the molecules or as genetic constructs, the expression of which can be regulated.
84550.indb 16
11/11/10 6:24:00 PM
The Contribution of Laboratory Animals to Medical Progress
17
One type of antisense compound is called a morpholino, which has a different chemical backbone than normal RNA and thus is not degraded within the cell. Morpholinos are microinjected into embryos or eggs and have been used to knock down gene expression in zebrafish, Xenopus, and sea urchins (reviewed in Corey and Abrams 2001). The use of antisense technologies to knock down gene expression in many animal models should increase greatly during the next several years. Stem Cells and Assisted Reproductive Technologies ESCs are derived from the early developmental stages of animals and can be used to create new, genetically modified animals (as for knockout mice) as well as specific cell types for therapies based on cell replacement (part of the discipline of regenerative medicine). At present, the very powerful technique used to make knockout animals using ESCs is limited to certain strains of mice. However, there is a great deal of activity aimed at isolating ESCs from other animals, such as NHPs, that can be used for similar purposes. In parallel, ESCs isolated from humans and many experimental animals are being tested for their potential to develop into specific cell types for the purposes of regenerative medicine. Examples include derivation of pancreatic beta cells from ESCs potentially to treat type 2 diabetes (Borowiak and Melton 2009) and derivation of specific types of neuronal cells potentially to treat Parkinson’s disease (Fricker-Gates and Gates 2010). ESC-based approaches to create genetically modified animals and specific cell types should continue as major research efforts in the future. Because of the difficulty of deriving ESCs that can be used for creating genetically modified animals, alternative procedures using the technique of somatic cell nuclear transfer (SCNT) have been developed. In this technique, the nucleus from a differentiated cell is introduced into an unfertilized egg. The resulting embryo is introduced into a “foster” mother and offspring can result. This is the basis of the creation of the sheep Dolly (Wilmut et al. 1997). SCNT has been used successfully for creation of several different animals, including cats, pigs, goats, horses, and monkeys (Wilmut, Sullivan, and Taylor 2009). However, the efficiency of this procedure is very low. One critical aspect of creation of animals by SCNT is that the genome of the donor somatic cell nucleus can potentially be modified (analogous to procedures used to create knockout mice), thus creating genetically modified animals. Many investigators are examining methods to increase the efficiency of SCNT to accomplish this goal. Phenotyping Characterizing the phenotype of an animal as it relates to a specific property or disease must go hand in hand with genetic analysis. As genetic technologies become more sophisticated, phenotyping must improve in parallel. Therefore, a major trend that should continue and increase is the development of phenotyping screens of increasing sophistication and the automation of these assays so that large numbers of animals can be characterized. Because of the current medical emphasis on infectious diseases such as AIDS, much effort is aimed at increasing the ability to characterize immune cells by techniques such as multicolor flow cytometry. Among the many topics of current interest in the area of phenotyping are improvements in imaging of intact animals (e.g., for use in neurobiology), large-scale behavioral testing of mutant mice to identify specific phenotypes, and advances in high-throughput approaches to pathology. Phenotyping technologies will be improved and many new ones will be developed in the future, based on high-throughput approaches. The goal will be to correlate increasingly sophisticated knowledge of animal phenotypes with detailed knowledge of variations in gene sequence and gene expression, leading to enhanced knowledge of both normal physiology and disease. In turn, this knowledge will be used to develop new and improved therapies for many human conditions.
84550.indb 17
11/11/10 6:24:00 PM
18
Handbook of Laboratory Animal Science
References Adams, M. D., S. E. Celniker, R. A. Holt, et al. 2000. The genome sequence of Drosophila melanogaster. Science 287:2185–2195. Albrecht, A. N. 2004. A molecular pathogenesis for transcription factor associated poly-alanine tract expansion. Human Molecular Genetics 13:2351–2356. Amaya, E. 2005. Xenomics. Genome Research 15:1683–1691. Apetrei, C. et al. 2006. Kuru experiments triggered the emergence of pathogenic SIVmac. AIDS 20 (3): 317–321. Austin, C. P. 2004. The knockout mouse project. Nature Genetics 36:921–924. Beck, C. W., and J. M. Slack. 2001. An amphibian with ambition: A new role for Xenopus in the 21st century. Genome Biology 2: reviews1029.1–reviews 1029.5. Biebink, G. S. 1999. Otitis media: The chinchilla model. Microbial Drug Resistance 5:57–72. Bodewes, R., G. Rimmelzwaan, and A. Osterhaus. 2010. Animal models for the preclinical evaluation of candidate influenza vaccines. Expert Reviews, Vaccines 9:59–72. Borowiak, M., and D. A. Melton. 2009. How to make beta cells? Current Opinion in Cell Biology 21:727–732. Bowden, D., and R. Martin. 1977. A digital Rosetta stone for primate brain terminology. In Handbook of chemical neuroanatomy, vol. 13, ed. A. Bjorklund, T. Hokfelt, and F. E. Bloom, 1–37. Amsterdam: Elsevier. Bukh, J. 2004. A critical role for the chimpanzee model in the study of hepatitis C. Hepatology 39 (6): 1469–1475. Capitanio, J., and M. Emborg. 2008. Contributions of nonhuman primates to neuroscience research. Lancet 317:1126–1135. Chen, R. Z. 2001. Deficiency of methyl-CpG binding protein-2 in CNS neurons results in a Rett-like phenotype in mice. Nature Genetics 27 (3): 327–331. Chisari, F. V. 2000. Viruses, immunity and cancer: Lessons from hepatitis B. American Journal of Pathology 156:1117–1132. Corey, D., and J. Abrams. 2001. Morpholino antisense oligonucleotides: Tools for investigating vertebrate development. Genome Biology 2 (5): 1015.1–1015.3. Cox, L. et.al. 2006. A second-generation genetic linkage map of the baboon (Papio hamadryas) genome. Genomics 88:274–281. Culetto, E., and D. B. Sattelle. 2000. A role for Caenorhabditis elegans in understanding the function and interactions of human disease genes. Human Molecular Genetics 9:869–877. Desrosiers, R. C. 1990. The simian immunodeficiency viruses. Annual Review of Immunology 8:557–578. Eleftherohorinou, H., V. Wright, C. Hoggart, et al. 2009. Pathway analysis of GWAS provides new insights into genetic susceptibility to three inflammatory diseases. PLoS One 4:e8068. Figeys, D. 2008. Mapping the human protein interactome. Cell Research 18:716–724. Fricker-Gates, R. A., and M. A. Gates. 2010. Stem cell-derived dopamine neurons for brain repair in Parkinson’s disease. Regenerative Medicine 5:267–278. Friedman, L. M. 2007. Mouse models to study inner ear development and hereditary hearing loss. International Journal of Developmental Biology 51:609–631 Gayraud, B. 2000. New insight into the assembly of extracellular microfibrils from the analysis of the fibrillin 1 mutation in the tight skin mouse. Journal of Cell Biology 50:667–680. Gibbs, R. et al. 2007. Rhesus Macaque Genome Sequencing and Analysis Consortium, evolutionary and biomedical insights from the rhesus macaque genome. Science 316:222–234. Gibbs, R. A. 2004. Genome sequence of the brown Norwegian rat yields insight into mammalian evolution. Nature 428:493–521. Green, M. C. 1976. Tight-skin, a new mutation of the mouse causing excessive growth of connective tissue and skeleton. American Journal of Pathology 82:493–512. Guy, J. 2001. Breeding and maintenance of a Mecp2-deficient mouse model of Rett syndrome. Nature Genetics 27 (3): 322–326. Hendriksen, C. 1996. A short history of the use of animals in vaccine development and quality control. Developments in Biological Standardization 86:3–10.
84550.indb 18
11/11/10 6:24:00 PM
The Contribution of Laboratory Animals to Medical Progress
19
Howell, L. L., and K. S. Murnane. 2008. Nonhuman primate neuroimaging and the neurobiology of psychostimulant addiction. Annals of the New York Academy of Sciences 1141:176–194. International Mouse Knockout Consortium. 2007. A mouse for all reasons. Cell 128:9–13. Ioffe, E. 1998. Abnormal regulation of the leptin gene in the pathogenesis of obesity. Proceedings of the National Academy of Sciences USA 95:11852–11857. Jacob, H. 1999. Functional genomics and rat models. Genome Research 9:1013–1016. Jenner, E. 1798. An inquiry into the causes and effects of the variolae vaccine. London: Samson, Low. Johnson, K. R. 1998. A new spontaneous mouse mutation of Hoxd13 with a polyalanine expansion and phenotype similar to human synpolydactyly. Human Molecular Genetics 7:1033–1038. Kim, D., and J. Rossi. 2007. Strategies for silencing human disease using RNA interference. Nature Reviews Genetics 8:173–184. Koch, R. 1877. Die Aetiologie der Milzbrand-Krankheit, begrundet auf die Entwicklungsgeschichte des Bacillus anthracis. Beiträge zur Biologie der Pflanzen 2:277–310. Kohler, G., and C. Milstein. 1976. Derivation of specific antibody-producing tissue culture and tumor lines by cell fusion. European Journal of Immunology 6:511–519. Kolber-Simonds, D., L. Lai, S. R. Watt, et al. 2004. Production of alpha-1,3-galactosyltransferase null pigs by means of nuclear transfer with fibroblasts bearing loss of heterozygosity mutations. Proceedings of the National Academy of Sciences USA 11:7335–7340. Lackner, A., and R. Veazey. 2007. Current concepts in aids pathogenesis: Insights from the SIV/macaque model. Annual Review of Medicine 58:461–476. Letvin, N. L. et al. 1985. Induction of AIDS-like disease in macaque monkeys with T-cell tropic retrovirus STLV-III. Science 230:71–73. Li, S. et al. 2004. A map of the interactome network of the Metazoan C. elegans. Science 23: 303(5657):540–543. Loffreo J. et al. 2007. CD8+ T cells from SIV elite controller macaques recognize Mamu-B*08-bound epitopes and select for widespread viral variation. PLoS ONE 2 (11): e1152. Maher, B. A. 2002. Test tubes with tails. Scientist 16:22. McCarthy, M. I., G. R. Abecasis, L. R. Cardon, et al. 2008. Genome-wide association studies for complex traits: Consensus, uncertainty and challenges. Nature Reviews Genetics 9:356–369. McManus, M., and P. Sharp. 2002. Gene silencing in mammals by small interfacing RNAs. Nature Reviews Genetics 3:737–747. Morse, H. C. 1978. Origin of inbred mice. New York: Academic. Mouse Genome Sequencing Consortium. 2002. Initial sequencing and comparative analysis of the mouse genome. Nature 420:520–562. Oelzner, P., S. Fleissner-Richter, R. Brauer, et al. 2010. Combination therapy with dexamethasone and osteoprotegerin protects against arthritis-induced bone alterations in antigen-induced arthritis of the rat. Inflammation Research Epub ahead of print, March 20, 2010. Pruisner, S. B. 1990. Transgenetic studies implicate interactions between homologous PrP isoforms in scrapie prion replication. Cell 63:673–686. Rogers, J. et al. 2006. An initial genetic linkage map of the rhesus macaque (Macaca mulatta) genome using human microsatellite loci. Genomics 87 (1): 30–38. Russell, E., L. Smith, and F. Lawson. 1956. Implantation of normal blood-forming tissue in radiated genetically anemic hosts. Science 124:1076–1077. Sabin, A. B. 1985. Oral poliovirus vaccine: History of its development and use and current challenge to eliminate poliomyelitis from the world. Journal of Infectious Diseases 151:420–436. Salk, D., and J. Salk. 1984. Vaccinology of poliomyelitis. Vaccine 2:59–74. Schena, M., D. Shalon, R. W. Davis, and P. O. Brown. 1995. Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270:467–470. Scollard, D. M., L. B. Adams, T. P. Gillis, et al. 2006. The continuing challenges of leprosy. Clinical Microbiology Reviews 19:338–381. Shim, Y.-H., and Y.-K. Paik. 2010. Caenorhabditis elegans proteomics comes of age. Proteomics 10:846–857. Shinobu, U. 2008. Establishment of rat embryonic stem cells and making of chimera rats. PLoS ONE 3:1–9. Snell, G. 1981. Studies in histocompatibility. Science 213:172–178.
84550.indb 19
11/11/10 6:24:00 PM
20
Handbook of Laboratory Animal Science
Spindel, E. et al. 2005. Leveraging human genomic information to identify nonhuman primate sequences for expression array development. BMC Genomics 6:160. Thessen, H. M., A. Gillett, T. Olsson, et al. 2009. Characterization of multiple sclerosis candidate gene expression kinetics in rat experimental autoimmune encephalomyelitis. Journal of Neuroimmunology 210:30–39. Ting, A. Y., B. F. Kimler, C. J. Fabian, and B. K. Petroff. 2007. Characterization of a preclinical model of simultaneous breast and ovarian cancer progression. Carcinogenesis 28:130–135. Wallace, J. et al. 2007. High-density rhesus macaque oligonucleotide microarray design using early-stage rhesus genome sequence information and human genome annotations. BMC Genomics 8:28. Williams, K. C., S. Corey, S. V. Westmoreland, et al. 2001. Perivascular macrophages are the primary cell type productively infected by simian immunodeficiency virus in the brains of macaques: Implications for the neuropathogenesis of AIDS. Journal of Experimental Medicine 193:905–915. Williamson, D. et al. 2003. Heritability of fearful-anxious endophenotypes in infant rhesus macaques: A preliminary report. Society of Biological Psychiatry 53:284–291. Wilmut, I., A. E. Schnieke, J. McWhir, et al. 1997. Viable offspring derived from fetal and adult mammalian cells. Nature 385:810–813. Wilmut, I., G. Sullivan, and J. Taylor. 2009. A decade of progress since the birth of Dolly. Reproduction, Fertility and Development 21:95–100. Wu, M., A. Arcaro, Z. Varga, et al. 2009. Pulse mTOR inhibitor treatment effectively controls cyst growth but leads to severe parenchymal and glomerular hypertrophy in rat polycystic kidney disease. American Journal of Physiology and Renal Physiology 297:597–605. Yang, S. et al. 2008. Towards a transgenic model of Huntington’s disease in a nonhuman primate. Nature 453:921–924. Zhang, Z. et al. 2002. Mamu-A-*01 allele-mediated attenuation of disease progression in simian-human immunodeficiency virus infection. Journal of Virology 76 (24):12845–12854. Zhou, L., C. R. Dey, S. E. Wert, et al. 1994. Correction of lethal intestinal defect in a mouse model of cystic fibrosis by human CFTR. Science 266:1705–1708. Zhu, H., and L. I. Zon. 2002. Use of zebrafish models for the analysis of human disease. In Current protocols in human genetics, Chapter 15, Unit 15.3.
84550.indb 20
11/11/10 6:24:00 PM
Chapter 2
Ethics of Animal Research
I. Anna S. Olsson, Paul Robinson, and Peter Sandøe Contents Introduction....................................................................................................................................... 21 What Are Our Duties to Animals?................................................................................................... 22 Does Animal Species Matter?..........................................................................................................26 How Can Benefits Be Maximized?...................................................................................................28 How Can Harm Be Minimized?....................................................................................................... 31 How Can Standards Be Maintained?................................................................................................ 33 Notes................................................................................................................................................. 35 References......................................................................................................................................... 35 Introduction Antibiotics, anesthetics, vaccines, insulin for diabetes, open heart surgery, kidney dialysis and transplants, treatments for asthma, leukemia and high blood pressureâ•›…â•›these are just some of the major medical advances that have depended on the use of animals in medical research and testing. Research Defense Society, 2008
In the quoted defense of animal-based research, a number of examples where experiments on animals have played a role in the development of vaccines and therapeutic treatments are listed. Most researchers, including the authors of this chapter, agree that research with animals has contributed to the development of life sciences and medicine over the last centuries. However, it is much more difficult to say with a reasonable degree of certainty what would have been achieved if animals had not been used. It seems likely that animals will continue to play a central role in biological and biomedical research in the foreseeable future, much as they do today. Current practice can be summarized by saying that experimental animals are used for three main purposes: to develop pharmaceutical and other medical products, to advance fundamental research in the life sciences, and to test the safety of potentially toxic products and substances. A rough idea of the relative numbers of animals used in these ways (in Western countries at least) is conveyed by recent data indicating that, of the experimental animals used in the European Union in 2005, more than 60% of animals were used in 21
84550.indb 21
11/11/10 6:24:01 PM
22
Handbook of Laboratory Animal Science
biomedical research and development and fundamental biology. Production and quality control of products represented 15% of the total number and toxicological and other safety evaluation 8% of the total number of animals used. The remaining animal uses include teaching, disease diagnoses, and other purposes. Much of this experimental activity requires that live animals be used, and this is the primary source of ethical concern. Some animals are housed so that they have limited freedom. Some are subjected to distressing or painful interventions. Nearly all of them are killed. The overwhelming majority are mammals with highly developed nervous systems. Unlike human subjects, however, they cannot consent to their own participation. Nor, generally, will they benefit from that participation, even if their descendants might. While few of us would question the desirability of discovering new ways to prevent, alleviate, or cure human and animal diseases, these realities surely confront animal researchers with a question: Are we, as human beings, morally justified in using animals as tools for biomedical research? This question can be addressed at different levels of generality. In this chapter, we begin by addressing the most general question: What are our duties to animals? The point of this question is to see use of animals for research as one of many ways in which animals are used or interfered with by humans. What becomes clear when the issue is addressed at this level of generality is that there is not a single ethical view regarding human duties to animals to which all can consent. Rather, different ethical approaches seem to live side by side. To allow for discussions of these approaches in a more structured way, a bit of philosophy is introduced. Three so-called theories of animal ethics— contractarianism, utilitarianism, and animal rights—are introduced with the dual aim of allowing readers to understand different views in the debate and to make up their own minds. In the following sections of the chapter, we focus on what can be done to uphold a high ethical standard when animals are used for experimentation. This part of the discussion is based on the assumption—not universally accepted—that it can sometimes be acceptable to use animals for research. We begin this part of the discussion by examining the moral significance of the species to which an experimental animal belongs. Discussions about the use of animals in research often gravitate toward questions about benefits and costs, and the next two sections reflect this. One discusses benefits, paying particular attention to scientific factors affecting experimental outcomes; the other looks at ways in which costs, in the form of harm to animals, can be minimized. The final section before the conclusion focuses on the ethical evaluation and regulation of animal-based research. We ask who, in practice, is responsible for ensuring that animals are treated ethically by scientists, and we consider whether the regulatory mechanisms typically relied upon today can be improved. What Are Our Duties to Animals? In today’s society, there are obviously many different views about what one is entitled to do to animals. However, these views are often rather superficial. They are rarely thought through. The same person may, when asked, express strong views about the importance of good animal welfare and, at the same time, buy cheap animal products in the supermarket, seemingly without showing concern about the living conditions of the animals whose eggs, milk, or meat he or she is buying. Because of this superficiality—which in many ways is very convenient for getting along with one’s life—it may be a problem just to be led by one’s feelings when discussing what is right and wrong in one’s dealings with animals. Such feelings are often unstable or ambivalent, and the ambivalence encourages double standards. This, however, is both morally objectionable and logically indefensible. Furthermore, it is clear that, at present, in the West, people are engaged in an increasingly serious debate about the rights and wrongs of animal use. However, it seems unlikely that professionals taking part in this debate will be able to communicate effectively if they merely push for their own
84550.indb 22
11/11/10 6:24:01 PM
Ethics of Animal Research
23
intuitively held beliefs. To be able to make themselves understood to people who hold different views, they must be able to understand the nature of their disagreement. To get a deeper understanding of the underlying ethical views, we shall here turn to moral philosophy, which distinguishes a number of types of ethical theory. In principle, any of these might underlie a person’s views about the acceptable use of animals. Here, three prominent theoretical positions will be presented: contractarianism (e.g., Narveson 1983), utilitarianism (e.g., Singer 1993, but also Frey 2001), and the animal rights view (Regan 1983). These have been selected because they have direct and obvious implications for the ongoing debate over the use of animals for research. (There are other views. For a more comprehensive presentation, see Sandøe and Christiansen, 2008.) Contractarianism: One sometimes hears it said that animals are morally insignificant or lack moral status. In the past, this attitude has been defended on the basis that animals do not reason about the world or on the basis that animals do not use language. Contractarians are broadly sympathetic to this kind of thinking. They regard morality as a system of hypothetical contracts that we tacitly enter into with one another. Animals cannot enter into these contracts, or agreements, because they lack the linguistic and intellectual skills to do so. Hence, animals are not bearers of rights and duties. Within the contractarian approach, animals are afforded ethical protection, however. It is just that the protection is indirect. Many people care about animals. They care especially about animals they own, of course, but they also care about animal use in general. They are therefore highly unlikely to go along with the idea that animals can be treated however one chooses to treat them. This means that the hypothetical contracts that underlie morality contain “clauses” requiring certain animals, at least, to be treated in ways that we tend to regard as fitting or acceptable. According to this view, animals can be compared with plants or features of the natural landscape; they have no inherent moral right to be treated in a certain way, but we happen to value them and sometimes rely on them, so they enjoy a borrowed kind of moral status, or secondary moral protection (see Sandøe and Christiansen 2008, Chapter 2). The implications of contractarianism for animal research are as follows. To the extent that people care about the animals used in experiments, the tacit contracts that constitute morality will contain clauses affording some protection to those animals. The scientific community ought to act in ways that people in general would broadly agree to or contract into. Clearly, most people care about cats and dogs more than they care about rats and mice.1 For the contractarian, this means that causing suffering to the former is likely to be more ethically objectionable than causing suffering to the latter. Similarly, nonhuman primates will probably receive more protection than other animals because their plight is of very considerable concern to many people. Ordinarily, of course, “contractors” in all walks of life depend heavily on their ability to provide what people want and avoid doing what people dislike. In the contractarian picture of animal ethics, commonplace observations like these apply in a very direct way to the animal research community. What matters are the feelings and beliefs of fellow humans on whose collaboration one depends to gain a license to operate. In this approach, then, setting ethical limits to the use of animals for research is really about defining a publicly acceptable framework that allows humankind to harvest the potential benefits of animal-based research. It is a key strength of this view that it has a built-in tendency to capture public attitudes to animal experimentation. Contractarianism ensures that animal ethics reflect the way people actually feel about various kinds of animal use. Unfortunately, this connection with human attitudes can also be cited as a weakness of the contractarian position. In a contractarian approach to animal experimentation, the plight of the animals themselves is not really the issue. Do animals matter only insofar as we happen to care about them? What if we stopped caring? Would that make it okay to do whatever one wants to an animal? Unlike contractarianism, the next view to be considered, utilitarianism, will give a negative answer to these two questions.
84550.indb 23
11/11/10 6:24:01 PM
24
Handbook of Laboratory Animal Science
Utilitarianism: According to this view, morality has one basic rule: Always act so as to maximize the well-being of those affected by your actions. In practice, of course, we will still have to follow simpler, everyday moral principles, such as do not lie, keep your promises, look after your children and parents, etc. Many of these moral principles that we apply in everyday life can, according to utilitarianism, be looked on as rules of thumb that enable us to serve the single basic principle to create the greatest possible amount of good. The good to be maximized, well-being, is usually defined in terms of enjoyment and the absence of suffering. It therefore requires sentience.2 Many of the animals in our care are sentient and thus are the sorts of beings that can be given or denied this type of well-being. Many of these animals therefore have moral status. For the utilitarian, all well-being matters in exactly the same way, whether it belongs to a concert pianist or a pregnant sow. In this sense, all sentient creatures, human and nonhuman, deserve equal moral consideration. There will be plenty of situations in which we can improve the wellbeing of animals in our care at little cost to our own welfare. When these situations arise, we have a moral obligation to attend to the animals’ interests because we have a moral obligation to act always in ways that maximize well-being. In this sense, animals make genuine moral demands on us. In the utilitarian approach, then, ethical decisions require us to strike the most favorable balance of benefits and costs for all the sentient individuals affected by what we do. But doing the right thing, according to the utilitarian, is not only a matter of doing what is optimal. It is also essential to do something rather than nothing: If something can be done to increase well-being, we have a duty to do it. This utilitarian duty to act so as always to bring about improvements has important consequences for society. In contemporary Western society, we retain a general tendency to give ourselves priority over animals. A thoroughgoing utilitarian will regard this tendency as essentially wrong. However, the human-centered outlook is obviously well established and, in view of this, it may well be that, for the time being at least, any attempt to ensure that sentient animals are accorded the same status as human beings is bound to fail. This may be especially true when it comes to animals used as tools in research that may potentially save many human lives. It may be that the best a utilitarian can hope to achieve is higher levels of animal welfare within the current system. In the case of laboratory animals, a pragmatic utilitarian might be willing to apply something called the “principle of the three Rs,” which we discuss in detail later.3 This principle requires researchers to replace existing live-animal experiments with alternatives, reduce the number of animals used, and refine methods so as to cause animals less suffering (Russell and Burch 1959). It is not hard to see that less invasive sampling techniques, improved housing systems, and more precise models requiring fewer animals to be used are likely to be viewed as morally attractive developments within the utilitarian perspective. In the ethical debate over animal research, the main conflict is usually between the pursuit of human benefits, on the one hand, and the animals’ interest in avoiding suffering, on the other. Sometimes, however, the utilitarian will want to weigh not just animal interests against human interests, but also the interests of different animals against each other. Obviously, animal experiments can benefit animals as well as humans. In fact, many of the insights underlying modern veterinary medicine have been derived from experiments on animals. When a pet cat is vaccinated against feline leukemia, it benefits from immunological research performed on other cats, although, of course, the primary purpose of the research was the development of treatments for human diseases. In deciding whether an animal experiment is ethically justifiable, it is sometimes necessary, then, to take into account the benefits of the results to animals as well as any hoped-for human gains. Both of these can be set against costs to animals whose interests are sacrificed in the experiment. Utilitarianism, as described previously, suggests that animal interests are best sacrificed when such sacrifice leads to the protection or satisfaction of vital human interests, as happens in much biomedical research. But is that an acceptable view? A more radical utilitarianism might be worth
84550.indb 24
11/11/10 6:24:01 PM
Ethics of Animal Research
25
exploring. Animal experimentation sometimes means sacrificing vital animal interests in continued life and the avoidance of abject suffering. Insisting firmly that human and animal interests deserve equal consideration, the utilitarian philosopher Peter Singer has concluded that the sacrifice of such vital animal interests is acceptable only when the benefits are extraordinarily important: “If a single experiment could cure a disease like leukemia, that experiment would be justifiable. But in actual life the benefits are always much, much more remote, and more often than not they are nonexistent” (1991). It is evident, then, that a wide range of views is represented within the utilitarian approach. Some utilitarian observers accept animal experiments when there are no alternatives and as long as we do our utmost to prevent and alleviate animal suffering. Others, like Singer, set the demand for human benefit higher and would prefer to see nearly all such experiments abolished. What all utilitarians agree on, however, is the methodological precept that ethical decisions in animal research require us to balance the harm we do to laboratory animals against the benefits we derive for humans and other animals. This precept—the notion that we can work out what is ethical by trading off one set of interests against another—is precisely what is denied by advocates of animal rights. Animal rights: In response to the problem just described, some theorists of animal ethics have developed the view that animals have rights. The main point of moral rights is to define boundaries that should not be crossed under any circumstances (unless, of course, the holder of the right waives the right, which is academic if we are talking about animal rights). Which animal rights are rights needs to be clarified, but even without this clarification one can see the appeal of a rights theory. The attribution of rights to animals allows us to insist that some ways of treating animals are totally unacceptable; not unacceptable only if enough people disapprove or if the benefits secured are too small, but unacceptable, period. What rights do animals have, in this view? A radical suggestion, but one that is not without supporters, is that every sentient animal has the right not to be treated merely as a “means to an end.” Sentient animals should not be used as instruments in the pursuit of human goals. In particular, they should not be killed for human purposes. They have a right not to be killed. The implications of this way of looking at matters are dramatic and far reaching. Tom Regan (1989) and other adherents of the animal rights view have argued for wholesale abolition of animal-based research (and most other forms of animal use, including the use of animals for food production). It matters not that an experiment will cause only minor harm to the animals it involves. It matters not that this experiment is of extraordinary importance to humanity at large. The only thing that matters is that every time an animal is used for an experiment, it is treated as a mere means to an end. This being so, animal experimentation should cease. It is possible to imagine a more moderate advocacy of the animal rights approach, however. The right not to be killed is regarded as basic by some proponents of animal rights. But one might be doubtful about this, partly because animals, maybe with the exception of great apes, have a much more limited perspective on the future than we have. What matters to animals is that, here and now, they are well off, whereas we have aspirations and worries that reach across our entire lives. In light of this, one might suggest that animals have something like a right to protection from suffering or certain levels of suffering. It could then be argued that all animals should be protected from suffering of the kind covered by the right—for example, the kind of suffering involving intense or prolonged pain or distress that the animal cannot control. Although this is not the standard animal rights view, it preserves the key idea that there are absolute, non-negotiable limits to what can be done to animals. An important idea related to animal rights is the principle of fairness. The key point here is that what matters is not only the sum of positive and negative consequences, as claimed by the utilitarian approach, but also the distribution of these consequences between individuals. For example, when animals are used in research involving pain, it may be considered fair and therefore better that a larger
84550.indb 25
11/11/10 6:24:01 PM
26
Handbook of Laboratory Animal Science
number of animals suffers a small and bearable amount of pain than that a few animals suffer strong pain—even if the total sum of pain is assumed to be larger in the first case (Tannenbaum 1999). It should be clear that the three perspectives outlined before are not, in any simple way, compatible. Contractarians, for example, will have a permissive attitude to most animal experiments, whereas advocates of animal rights will take a restrictive—often an abolitionist—position. Even moderate forms of the animal rights view will, on some occasions, conflict with the utilitarian approach. To see this, consider an experiment that causes a great deal of suffering to the animals involved, but that is very likely to lead to significant benefits to many humans or animals. Moderate rights advocates will probably want to prohibit the experiment so that that level of suffering is not visited on the animals by us. By contrast, utilitarians may not object to the experiment because they think that, on balance, the benefits will probably outweigh the suffering imposed on the animals. The use of rats as models of arthritis might be a relevant illustration. This model is created using injection of collagen, a substance from bone joints that causes a form of autoimmune arthritis to develop. Attempts have been made to alleviate the pain of the rat with painkillers. However, since all available painkillers also, directly or indirectly, have anti-inflammatory effects, their use may lead to undesirable interference with the research. Thus, it seems reasonable to expect that the rats used to test potential drugs for arthritis may suffer pain similar to that endured by human arthritis patients. This kind of model would be accepted by the contractarian. The brutal truth is that in the potential contract negotiation, the animal has nothing to offer in return for not being experimented on. It could also be accepted from a utilitarian perspective, with the argument that the admittedly rather high cost imposed on the animals is outweighed by potential benefits to arthritis patients. However, from even a moderate rights perspective, the experiment may look unacceptable; even moderate animal rights place a non-negotiable duty on us not to cause the relatively high level of suffering associated with multiple inflammations in joints. Where does this leave matters? It is important to see that the three approaches we are considering do not agree upon the rights and wrongs of animal experimentation. Thus, there is a need for discussion. Even though society will define limits in terms of legislation, each person will have her or his limits for what is considered acceptable regarding animal use, including use of animals for research. And different persons will almost always have different limits. In light of the ethical theories presented previously, it is not only possible to make up one’s own mind about what is right and wrong in our dealings with animals, but it is also possible to understand the views of other people. Such an understanding is an important requisite for a civilized dialogue about animal use. One of the issues that may come up in such a dialogue concerns the choice of animal species to be used in experimentation. We will turn to this issue next. Does Animal Species Matter? Animals of very different species are used in research. The choice of animal depends heavily on the kind of research being done, of course, but it is also affected by the experience and expertise of the researcher, the facilities of the institution, legislation, and sometimes public discussion in the country where the work is being carried out. The significance of these last factors is brought out, for example, by the fact that even for in vivo research in the neurosciences, which requires an animal with a complex enough nervous system to embody mechanisms for learning and memory formation, the available research species range from nematodes to chimpanzees.4 Does the choice of species matter when it comes to ethical evaluation of animal-based research? It does, as we will now try to explain, but just how it matters is determined differently by different ethical theories. Contractarians will be concerned primarily with public sensitivities to the experimental use of different species. Through their impact on the tacit contracts we live under, these differentiated sensitivities give rise to a species-specific ethics of animal use. By contrast, utilitarians
84550.indb 26
11/11/10 6:24:01 PM
Ethics of Animal Research
27
will focus on the issue of whether animals of a particular species are capable of suffering or frustration. The more rigorously it can be demonstrated that animals of a given species are able to suffer or experience frustration, the stronger the case will be for not using them. Similarly, advocates of animal rights will be concerned mainly about psychological sophistication, since rights are more readily ascribed to animals with advanced mental capacities. In ethical discussion of the significance of species, two concepts are especially important: that of sentience and that of the sociozoological scale. We shall therefore organize what we have to say around these concepts. Sentience is the capacity to perceive or feel things. A sentient creature experiences the world around it (e.g., Duncan 2006). It may also experience feelings and emotions. In the words of Thomas Nagel (1974) in his classical analysis of this type of conscious experience, “The fact that an organism has conscious experience at all means, basically, that there is something it is like to be that organism.” Scientific understanding of sentience (both human and animal) is still limited. Neurobiologists have not yet managed to explain it in terms of the material mechanisms of the nervous system. In general, our belief that other individuals are sentient is based on the observation that they are behaviorally and physically similar to us. In other words, if an individual acts in a way that is similar to the way we would act in a certain kind of circumstance and that individual possesses something like a central nervous system, we regard it as probable, on the basis of an argument from analogy, that this individual is sentient. This reasoning is relatively uncontroversial for adult human beings, but when we extend it to nonhuman animals, the issues become more complicated. Here, verbal evidence is unavailable and the behavioral and physical similarities are more limited. Although common sense detects sentience in many species, the scientific case for attributing it obviously needs to be based on systematically assessed evidence. Smith and Boyd (1991) set out a systematic method of assessment of the kind needed here. They provide a checklist of neuroanatomical/physiological and behavioral criteria that can be used to determine whether a nonhuman animal has the capacity for pain, stress, and anxiety. In relation to any of these kinds of experiences, the checklist will include the possession of higher brain centers and evidence of behavioral reaction to potentially nociceptive, anxiogenic, or stressful experiences. Further evidence will accumulate if the behavioral reactions are modulated by drugs with a known anxiolytic or analgesic effect in humans. Evidence will also accrue if peripheral nervous structures (including receptors, signal substances, and hormones) are involved in each type of reaction— especially if there is a connection between these latter structures and the higher brain centers. As an increasing number of these criteria are met, the case for categorizing the animal as sentient builds. When we look at the way taxonomically distinct animals fare under systematic scrutiny of this kind, we see that there are two important lessons to be learned. The first is that all vertebrates meet the criteria for sentience. When Smith and Boyd’s original analysis was published, positive evidence existed only for mammals and birds, but over the last decade, it has been demonstrated that the criteria are also met by fish (Ashley and Sneddon 2008; Braithwaite and Boulcott 2008). The second lesson is that, for many of the invertebrates, we still know too little to be able to say whether sentience can safely be attributed. Thus, Eisemann and colleagues (1984) have presented a list of reasons why it is unlikely that insects are able to feel pain (including their lack of a behavioral response to protect an injured limb), while Lockwood (1987) and Sherwin (2001), relying primarily on behavioral evidence, have argued that we should consider extending the argument of analogy to support the conclusion that insects are sentient beings. Ethically speaking, then, how important is sentience as a factor in the selection of species for animal research? In one way, it might be said to be very significant. Both utilitarians and advocates of animal rights attach significance to it; even the contractarian might have an indirect interest in it, since sentient animals tend to matter more to humans than nonsentient ones. In another way, however, the significance of sentience as a criterion of species selection can be seriously doubted. It is relatively poorly understood and it is attributed to animals largely on the basis of imperfect
84550.indb 27
11/11/10 6:24:02 PM
28
Handbook of Laboratory Animal Science
analogies. Worse, a systematic checklist of scientifically respectable indicators shows that sentience is possessed by all vertebrates and possibly some invertebrates. That is not of much help if one is trying to determine which species to use in a potentially painful in vivo procedure. There is a rather different way to approach the questions of whether and how animal species matters. Throughout human culture, there is clearly a perception of hierarchy among animals—a quasi-moral ordering that gives some species higher status than others. This hierarchy has been labeled “the sociozoological scale” (Arluke and Sanders 1996). The central idea of the scale is that people rate animal species as morally more or less important—and therefore more or less worth protecting—on the basis of a number of factors. These include how useful an animal is, how closely people typically associate with it, and how “cute” it is. They also include how dangerous the animal is capable of being and how “demonic” it is perceived to be. Clearly, the sociozoological scale varies from place to place and time to time; however, today, at least in Western societies, some companion animal species—notably dogs and cats—seem to be at the top of it. Among other animals, large carnivores and nonhuman primates also figure at the top end of the scale. In the middle are large farm animal species such as cattle and pigs. Toward the bottom are pests or vermin such as rats and mice. Fish, viewed by some to be alien, cold, and slimy, also appear to be quite low on the scale. At any rate, among the animals used for research, there is a hierarchy, running from primates at the top to rodents and fish and on down to insects and other invertebrates. The sociozoological scale is, in many ways, based on tradition and unexamined prejudice, and its use as a basis of animal protection can be criticized both scientifically and ethically. From the utilitarian and the animal rights perspective, it is bound to seem morally wrong to discriminate among animals solely on the basis of the scale—an unfairness comparable to racist treatment of humans. In the contractarian view, on the other hand, there is nothing problematic about treating animals in line with the scale and thus giving more protection to primates and dogs than is given to rodents and fish. Indeed, this is a morally attractive policy because, in the contractarian view, animals matter only to the extent that they matter to humans. We conclude this section with a case illustrating the complex way in which the sociozoological scale and sentience, and contractarianism and utilitarianism, sometimes interact with one another. Looking for a vertebrate that is smaller and easier to reproduce and manipulate genetically than the typical laboratory rodent, life science researchers are increasingly turning to zebrafish. From the contractarian/sociozoological perspective, the use of this species in research is relatively unproblematic. Fish look very different from us; plainly, they live in conditions quite unlike those we live in and their plight matters to the average person much less than that of the domestic cat or possibly even the laboratory rat. For the utilitarian, however, the fact that fish are sentient may render their employment in research morally questionable. Certainly, from the utilitarian perspective, we will be obliged to consider the harm that research may do to the fish and make efforts to prevent it. Here, the perceived distance between human beings and fish may be a disadvantage for the fish, since it may make it difficult for a human observer to recognize the fish’s signs of distress—particularly given our relative lack of knowledge of pain and fear behavior in these animals.
How Can Benefits Be Maximized? Broadly speaking, the aim of animal research is to secure benefits—chiefly through the acquisition of new knowledge that provides answers to fundamental questions in biology or improves human and animal health and safety. However, as many animal researchers will ruefully confirm from personal experience, the assumption that benefits will be delivered cannot be taken for granted.
84550.indb 28
11/11/10 6:24:02 PM
Ethics of Animal Research
29
Science is not a predictable “manufacturing” activity; even when we are armed with well-defined questions and correctly designed and carefully executed experiments, it is sometimes impossible to predict whether a research project will improve our understanding of important biological mechanisms or lead to the development of therapeutics. However difficult it is to predict them, assessing benefits is, of course, fundamental for balancing them against the harm the experimental procedures may cause to the animals involved. Assessments of benefits also help promote the most effective use of resources. Here it is important to say that assessing benefit is not a question of distinguishing between applied and fundamental research. Instead, the aim should be to address whether a suggested research project is likely to be able to generate the benefits it aims to. Drawing on data and feedback from European ethics review committees, the Federation of European Laboratory Animal Science Associations (FELASA) working group has recently described a set of key questions that ought to be asked about any research project involving animals (Smith et al. 2007). On the benefit side, these questions include: • • • •
How will the results add to existing knowledge and how will they be used? Are the objectives realistic, original, and timely? How is the work related to previous and ongoing work in the research group and elsewhere? How likely is it that the benefits will be attained, based on: • Choice of animal model and scientific approach? • Experimental design? • Competence of staff? • Appropriate facilities? • Communication of results?
In academic research, the first two questions are typically addressed in the scientific evaluation of funding applications. We will focus here on the last question and, in particular, on issues connected with the choice of animal model, experimental design, and communication. In animal research, the suitability of animal models is often a critical factor determining whether or not the expected scientific and medical benefits are secured. In some areas of research, the choice of animal species to be used is obvious; for example, the agricultural scientist interested in aspects of dairy cow metabolism will develop his research on dairy cows. But in much fundamental biology and biomedical research, animals are used as models: Researchers study animals of one species with the aim of gaining understanding with a wider application or with application to another species (typically, humans). Critical discussion of what characterizes a good animal model is curiously rare in the scientific literature. Most review papers on animal models limit themselves to an overview of the models and the connected discussion of any results that have been generated in studies using them. However, it has been forcefully pointed out that suitable animal models and the appropriate use of them are crucial in improving the success rate of pharmaceutical drug development (i.e., in moving from a promising compound to an approved, marketable drug) (Kola and Landis 2004; Markou et al. 2009). Critical evaluations of animal models address various aspects of validity. The evaluative methods used have been developed most extensively in the field of neurobehavior, following Paul Willner’s (1984) enquiries into depression models. The most reliable measure of how well a model models is, of course, “predictive validity”—that is, how well results obtained using the model predict outcomes in other species of interest. But it will often be many years before such information is available about a model. Thus, in the development of treatments for human disorders, such validity is confirmed only when putatively effective compounds have made it all the way into studies with human volunteers—a process normally taking at least 10 years. Therefore, researchers will look for earlier theoretical indications of model validity. The notion of “construct validity”—recording how similar the underlying mechanisms of the model and the other species of interest are—is useful here.
84550.indb 29
11/11/10 6:24:02 PM
30
Handbook of Laboratory Animal Science
Unavoidably, scientists operate under practical constraints. Most research is, to some extent, dependent on existing technologies. It is shaped by factors such as what models have been used before, what models the researcher has expertise in, whether an animal colony has already been set up at high cost, and so on. A telling example here is provided by genetic models of Huntington’s disease. In a recent study, Heng and co-workers (2008, p. 8) concluded: The practical advantages of the strong R6/2 phenotype [with poorer construct validity] make it unlikely that it will be replaced as the preferred modelâ•›…â•›The milder phenotype and late onset of behavioral abnormalities of transgenic full length and knock-in murine models [with better construct validity] make them difficult to use for preclinical pharmacology.
This kind of decision reflects the reality in which scientists operate. However, the simple, general aim should be to use the best scientific model for the study in question. As gene technology has developed over the last decades, many transgenic models are now available for diseases of a genetic origin. Is it then still relevant to use the older, pharmacologically induced models? This and related questions can be applied to a wide range of research areas. We now move on to look at the way in which experimental planning and design affect research benefits because this is an issue around which considerable and challenging evidence has accumulated over the last couple of years. We will use the example of animal research underlying the development of treatments for stroke in humans. In this field, a number of compounds have shown neuroprotective effects in animal models, but very few have turned out to be effective in clinical trials on humans (van der Worp et al. 2005). This could be explained by the fact that animals are poor stroke models for the human condition and offer low predictive validity. However, this is not the only plausible explanation. Researchers concerned over the limited translation of preclinical research results into effective human stroke treatments have carried out several systematic reviews of the earlier animal experiments. They have found a number of critical shortcomings in experimental planning and design. For example, in many of the animal experiments, the efficacy of the prospective treatment was probably overestimated as a result of design bias. Often, animals were not randomly allocated to treatments and researchers, who were not blinded when they administered treatments (drug or control) or assessed outcomes, may have influenced measurements unconsciously (van der Worp et al. 2005; Crossley et al. 2008). Significant clinical differences were also an unwelcome factor, in that the animals used were generally young and healthy before the experimentally induced stroke, while human stroke patients are often elderly and hypertense (Macleod and Sandercock 2005). The third issue we will discuss is communication. By and large, if research is to be beneficial, it will be important for the results to be made public. Publication in peer-reviewed journals is a prominent feature of modern academic life, particularly in the sciences. As is well known, the performance of today’s researchers is measured largely on the basis of the number of publications they have in influential journals. Thus, at least in academia, there is no doubt that researchers will invest time and effort in the communication of their results. However, it is generally difficult to get studies with negative results (no effect of treatment) published. As a direct consequence of this, publications are likely to reflect only a subset of the research that has been carried out in the field. This has serious ethical consequences. In particular, it affects the number of animals used in research, as is explained in the following passage: The “publication bias” of journals in favor of hypothesis-confirming resultsâ•›…â•›m ight be a reason for the slow progress in the development of new animal models and their validation. Negative results often go unpublished, and poor concepts, hypotheses, and models survive, notwithstanding a vast amount of contradictory data, merely because these data are not made available to the scientific community.â•›… Publication of negative findings from well-conceived and performed studies can help investigators to evaluate and ultimately abandon the development of an invalid and irrelevant animal model and help reduce the unnecessary use of laboratory animals. (Van der Staay 2006, p. 147)
84550.indb 30
11/11/10 6:24:02 PM
Ethics of Animal Research
31
It can be powerfully argued, then, that there is an urgent need to create a coordinated, internationally recognized, searchable database where data on negative experimental findings can be deposited. To this point, we have focused exclusively on biomedical research, where the main purpose is to understand disease mechanisms and develop treatments. Where fundamental biological research is concerned, the practical benefits of using animals tend to be harder to predict because applications of any results are further away from the research itself. By and large, it is not a useful exercise to ask whether an applied biomedical study is more likely to deliver benefit than a fundamental study of biological mechanisms. The advance of science and technology requires both varieties of research to be pursued. How Can Harm Be Minimized? In discussing scientific benefits, we pointed out that a balance of benefits and harms is struck in ethically acceptable research. We now turn to the other side of this equation: the harm factor. Ethical concerns about compromised animal welfare are rarely addressed effectively via a demonstration of human benefit alone. Efforts to reduce the harm done to animals during the research are generally important as well. Indeed, in ethical terms, harm reduction may be the more urgent concern. Certain levels of pain and suffering imposed on animals in the name of science are regarded by most people today as quite literally intolerable—a belief readily understood in terms of moderate animal rights, of course. But if this is right, there will be certain kinds of experimental procedures in which high levels of pain for animals are caused, for which no amount of reassurance about benefits will serve as justification. The only way to deal with these ethical concerns will be by mitigating the animal harm. The three Rs principle (Russell and Burch 1959) addresses harm reduction through three approaches: the replacement of animal research with alternative animal-free methods, the reduction of the number of animals used, and the refinement of methods to minimize the distress caused to animals used in research. In what follows, we shall examine each R in turn, seeking to bring out the main consequences for animal research ethics. The replacement principle is particularly important from the ethical viewpoint, since it is one of the few ethical precepts over which there is broad consensus. It is, for example, the only element of the three Rs that animal rights advocates endorse. The idea behind it is simple: If it is possible to obtain scientific benefits without using live animals, we should do so. Scientists reading this chapter will be aware that experimental procedures that do not involve live animals, such as in vitro (e.g., cell lines), ex vivo (e.g., tissue culture), and in silico (e.g., bioinformatics) methods, are widely used in research laboratories already. In fact, these methods and the techniques that employ animal research subjects often work in a complementary fashion. However, strictly speaking, replacement requires existing procedures using animals to be abandoned in favor of animal-free methods, and this kind of readjustment is rarely straightforward. Typically, when scientists apply for ethical approval for their proposed studies, they are asked to explain why they cannot be carried out without animals. Usually, the answer given is that the study requires the complexity of a complete living organism and is therefore not suitable for in vitro approaches. Is it possible to challenge that answer? It may be difficult to do so today, but science is a rapidly developing endeavor; as new methods appear, the possibilities change. Procedures involving animals may come to be replaced by novel in vitro and in silico methods or by carefully designed studies with human volunteers or innovative uses of existing patient data. In biomedical research, new techniques are approaching or even entering a sphere of research activity traditionally dominated by animal models. In the early research phase, in vitro methods play an important role in characterizing potentially effective compounds prior to preclinical research on animals (e.g., Markou et al. 2009). At a later stage, the pharmacokinetics of candidate drugs are
84550.indb 31
11/11/10 6:24:02 PM
32
Handbook of Laboratory Animal Science
normally studied in animals. However, microdosing techniques, in which the uptake and metabolism of a drug are studied in human volunteers given doses so low that they have no biologically significant effect, are being explored (Rowland 2006).5 Innovative use of human data permitting animal replacement can be illustrated with two cases. Conventionally, the genetic effects and effects of the intrauterine environment on fetal origins of chronic disease are researched with embryo transfer or cross-fostering of animals. However, human medically assisted reproduction also results in parent–offspring pairs with contrasting combinations of genetic and environmental similarity and dissimilarity. Exploiting this, a research team has recently used fertility clinic records to build a database of information that can be used to study the effects of maternally provided prenatal environment directly on human data (Thapar et al. 2007). In a second case, a workshop brought together scientists and promoters of alternatives to animal-based pain studies. It concluded by listing studies using human volunteers that would not only decrease the number of animals used, but also produce more useful data by allowing the establishment of direct links between the human subjective pain experience and the biological parameters under study (Langley et al. 2008). While the previous examples concern research, overall development of replacement methods has focused on routine testing, the production of biological material, and teaching. A steadily (albeit slowly) increasing number of alternative test methods have gained regulatory acceptance (ECVAM 2008) and the highly invasive in vivo production of monoclonal antibodies using the ascites method in mice can, in most cases, now be avoided (Hendriksen 1998). A number of teaching tools, ranging from videos to interactive software to highly sophisticated mannequins, allow living and euthanized animals to be replaced at various levels of teaching (Interniche 2008). Through a combination of novel teaching tools and carefully guided practice on patients in the veterinary hospital, it is even possible to complete veterinary training without killing animals or performing invasive treatments on them, solely for training purposes (Knight 2007). Let us turn to the reduction principle. In animal studies that involve harm, the use of fewer animals will normally cut (as it were) collective animal suffering. That is the primary ethical motivation for reduction.6 However, reduction has other benefits. For one thing, it is good resource management; laboratory animals and their housing and care are costly and research resources are limited. Reduction considerations may also go hand in hand with good (i.e., efficient) experimental design with proper attention paid to standardization and the control of variation (Nevalainen 2004) and the choice of administration routes with a high degree of control (Svendsen 2005). It is important to appreciate, on the other hand, that reduction may have an adverse impact. It has been pointed out that using too few animals to produce meaningful results is as unethical as using more animals than necessary (Nevalainen 2004). Interestingly, a number of systematic reviews indicate that many animal studies use too few animals to provide reliable data (Sena et al. 2007). Such studies cause harm without benefits and involve poor use of resources. Another problem—essentially a conflict between reduction and refinement—arises because lowering the total number of animals used will sometimes place a greater burden on each animal that continues to be used. If a given quantity of plasma can be obtained by bleeding the same animal several times instead of bleeding several animals once, it is unclear, to say the least, whether we would be making the world a better place by doing the former (Hansen et al. 1999). It is certainly not eccentric to suppose that, if a burden must be borne, it is best shared. This was presented earlier as the principle of fairness to the individual animal (Tannenbaum 1999) and is also a reason to think critically about the possibility of reducing animal numbers by reuse of animals. Of course, this does not mean that reuse of animals is not worth considering—as long as the accumulated burden on one animal is not larger than what is acceptable within a single procedure. The notion of “animal numbers” is also less clear than might initially be supposed. What are we to reduce? The total number of animals used? Or the number of animals used relative to scientific output? After a period of steady decline, figures on laboratory animal use have risen over the last few
84550.indb 32
11/11/10 6:24:02 PM
Ethics of Animal Research
33
years (Hudson 2007). However, growth in investment in biomedical research may mean that the number of animals now being used relative to the amount of scientific activity taking place is falling.7 Again, should we be concentrating on the number of animals being used or the number of animals suffering? Not all animal research induces pain or suffering. The majority of approved procedures in animals are classified as minimally invasive. In many experimental regimes, animals are euthanized before they are exposed to invasive treatment or develop signs of disease. Utilitarians will want to focus exclusively on the number of animals that suffer. Proponents of animal rights may prefer to talk about reductions in total numbers, as might contractarians, if public attitudes were to become fiercely animal liberationist. Once it has been shown that a research aim cannot be pursued without animal use and once the animal numbers have been cut as much as possible, the refinement principle urges us to minimize any pain or distress that will be caused by amending experimental procedures. Few people would challenge this principle—at least, as long as conformity with it poses no threat to scientific results and does not require exorbitant funding. The only thing we need note is that limits on refinement may be required by the fact (noted earlier) that an inverse relationship sometimes exists between refinement and reduction. Experiments can be refined in various ways. We mention only a few. The most direct strategy is to adapt experimental procedures so that they cause less pain or distress. In addition to this, the housing and day-to-day care of experimental animals can often be improved. Environmental enrichment—that is, the provision of resources that enable animals to interact with and control features of their environment and to engage in motivated behaviors—normally improves animal well-being (e.g., Olsson and Dahlborn 2002; Wolfer et al. 2004). In some experiments, appropriate anesthesia and analgesia can play a vital role in pain management (e.g., Morton 2007). When animals are experimentally required to develop progressively severe conditions, as happens in degenerative disease models, an important refinement can be achieved by humane end points. In this sort of case, the technician or researcher uses clinical signs as end point parameters rather than awaiting the animal’s spontaneous death. Interestingly, housing adaptations and humane end points are also scientific considerations: Severely affected animals not offered the refinements are likely to die from secondary causes (e.g., dehydration or malnutrition in rodents unable to feed and drink from the cage top), rather than the disease under study. Survival or mortality when the cause of death is unknown or only indirectly related to the disease is not a high-quality variable to measure. As demonstrated by Scott and colleagues (2008), failure to consider non-disease-related mortality in a neurodegenerative disease model may even account for false treatment effects. How Can Standards Be Maintained? In previous sections of the chapter, we have tried to describe and explain the theoretical basis of a range of ethical norms and standards applying to laboratory animal use. In this final section, we turn to look at how these standards are maintained. The maintenance of standards in society is invariably achieved through a combination of “hard” regulation and “soft” promotion. People are encouraged to act in ways that society deems acceptable both by rules (sometimes backed by sanctions) and by policies that promote a positive attitude to the values underlying those rules. Though we focus mainly on regulation here, we do not underestimate the importance of the soft promotion of responsible attitudes to animal research. Nobody imagines for a moment that sexism, for example, can be eradicated from society through legislation and regulations alone. We know that policies reinforcing antisexist culture and values are also required. Similarly, it is impossible to ensure that animal-based research is ethical simply by imposing rules and regulations. Ultimately, the aim must be an animal research community identifying with the values that underpin the rules—to create and maintain a culture, within animal experimentation, of ethical responsibility.
84550.indb 33
11/11/10 6:24:03 PM
34
Handbook of Laboratory Animal Science
With that important proviso, we turn now to regulation. Who is responsible for ensuring that animals are treated ethically in the laboratory, and how well do they do their jobs? The first part of this question is much easier to answer than the second. Most animal-based research is funded, directly or indirectly, by public money. This means that the public, or society as a whole, must be counted among a research institution’s stakeholders. Society uses a number of mechanisms to guarantee that research on animals is carried out in an acceptable way. The most obvious of these is legislation, which, in terms of enforcement, is a powerful tool. But the legislative process is sluggish, while science and technology develop rapidly; this means that laws must be broadly drafted if they are not promptly to go out of date. As a consequence, the real decision making on treatment of animals in research projects is usually delegated to an ethics or animal care and use committee. Committees can act in a more flexible way. They are also able to enter into dialogue with scientists proposing experimental projects and, in that way, to challenge scientists to develop their research in line with evolving best practice. Ethics committees and other similar bodies are often the only formal regulatory bodies tasked with looking in detail at research projects. A complete and transparent review process is dependent on committee composition and dynamics; it should represent all important stakeholders in the discussion equally. There seem to be at least three main stakeholders: • Researchers and industry (usually represented by scientists) that have an interest in being able to perform their proposed studies • Animals (usually represented by veterinarians and animal caretakers) that have an interest in being protected from harm • Society (represented by lay members as well as interest groups, such as patient organizations and animal protection organizations)
The involvement of a wide range of parties with various approaches to the ethical issues raised by animal research will help reduce the risk of committees becoming biased toward researchers—a risk that has been noted in at least one study (Schuppli and Fraser 2007). Not every aspect of animal research can be under the ethics committee’s control, and the ultimate responsibility for the way in which animals are used rests with individual researchers. This is true, not just in moral terms, but also practically, because many decisions with ethical implications are made in the course of ongoing research. It can therefore be argued that critical discussion and self-regulation within the scientific community are hugely important. Those performing animalbased research must ask themselves whether their work prompts ethical concerns. The earlier sections of this chapter are intended to assist with that kind of inquiry. Increasingly, ethics and the three Rs are considered in the assessment of funding applications. In the review of manuscripts submitted for publication, by contrast, it seems that most journals continue to require merely a statement affirming that the research complies with official recommendations, relevant legislation, and/or an ethics committee’s decision. It would be hard to deny that scientific journals could make better use of their position in the research process to raise the ethical standards of animal use. Refusals to publish papers based on the ethics of methodology would send a very strong signal to scientists. A policy of encouraging or requiring the authors of papers containing animal-based research to describe any ethical problems raised by their work would also be beneficial. It is noticeable that information on the adverse effects of experimental methods on animals is rarely reported in scientific papers (e.g., Olsson, Hansen, and Sandoe 2008). Finally, society has a legitimate interest in the activities of animal researchers. This is not only because much research involves public money, but also because sentient animals are a type of being deemed worthy of consideration and protection. This means that scientists using animals are, at some level, accountable to society; they must seek to explain their work, and they must seek equally
84550.indb 34
11/11/10 6:24:03 PM
Ethics of Animal Research
35
to listen to public concerns. Engagement of this kind is always in the scientist’s best interests. As we remarked in the previous edition of this chapter, in the twenty-first century, transparency and accountability are watchwords applied in most areas of collective human endeavor. Thus, faced with questions about their work, the worst thing animal researchers can do is to try to shut the inquirer out (Olsson et al. 2002). Notes
1. See the discussion of what has been labeled the “sociozoological scale” in the section “Does Animal Species Matter?” Contractarian theory automatically aligns animal ethics with this scale—for example, by affording greater moral protection to pets than to wild animals. 2. We discuss sentience in more detail in the section “Does Animal Species Matter?” 3. See the section “Minimizing Harm.” Note that some advocates of contractarianism and animal rights might well agree with the utilitarian about the desirability of applying the principle of the three Rs. 4. Of course, for single experiments the choice of species is often much more limited. 5. This application of microdosing is yet to be fully validated. 6. Even in animal studies that do not involve harm, reduction may still be thought valuable: An animal rights advocate, for example, might welcome the fact that fewer animals are being used as means to an end. 7. The figures visible to the public portray the absolute numbers of animals used—an issue of great relevance to those involved in public science outreach activities.
References Arluke, A., and C. R. Sanders. 1996. Regarding animals. Philadelphia: Temple University Press. Ashley, P. J., and L. U. Sneddon. 2008. Pain and fear in fish. In Fish welfare, ed. E. J. Branson. New York: Wiley–Blackwell. Braithwaite, V. A., and P. Boulcott. 2008. Can fish suffer? In Fish welfare, ed. E. J. Branson. New York: Wiley–Blackwell. Crossley, N. A., E. Sena, J. Goehler, J. Horn, B. van der Worp, P. M. W. Bath, M. Macleod, and U. Dirnagl. 2008. Empirical evidence of bias in the design of experimental stroke studies—A metaepidemiologic approach. Stroke 39 (3): 929–934. Duncan, I. J. H. 2006. The changing concept of animal sentience. Applied Animal Behavior Science 100 (1–2): 11–19. ECVAM (European Centre for the Validation of Alternative Methods). Accessed November, 26, 2008. Eisemann, C. H., W. K. Jorgensen, D. J. Merritt, M. J. Rice, B. W. Cribb, P. D. Webb, and M. P. Zalucki. 1984. Do insects feel pain? A biological view. Experientia 40 (2): 164–167. Frey, R. G. 2001. Justifying animal experimentation: The starting point. In Why animal experimentation matters, ed. E. Frankel Paul and J. Paul, 197–214. Piscataway, NJ: Transaction Publishers. Hansen, A. K., P. Sandøe, O. Svendsen, B. Forsman, and P. Thomsen. 1999. The need to refine the notion of reduction. Paper presented at the Humane End Points in Animal Experiments for Biomedical Research 1999. Hendriksen, C. F. M. 1998. A call for a European prohibition of monoclonal antibody production by the ascites procedure in laboratory animals. Atla-Alternatives to Laboratory Animals 26 (4): 523–540. Heng, M. Y., P. J. Detloff, and R. L. Albin. 2008. Rodent genetic models of Huntington disease. Neurobiology of Disease 32 (1): 1–9. Hudson, M. 2007. Why do the numbers of laboratory animal procedures conducted continue to rise? An analysis of the home office statistics of scientific procedures on living animals: Great Britain 2005. AtlaAlternatives to Laboratory Animals 35:177–187. Interniche (International Network for Humane Education). Accessed November, 26, 2008. Knight, A. 2007. The effectiveness of humane teaching methods in veterinary education. Altex-Alternativen zu Tierexperimenten 24 (2): 91–109. Kola, I., and J. Landis. 2004. Can the pharmaceutical industry reduce attrition rates? Nature Reviews Drug Discovery 3 (8): 711–715.
84550.indb 35
11/11/10 6:24:03 PM
36
Handbook of Laboratory Animal Science
Langley, C. K., Q. Aziz, C. Bountra, N. Gordon, P. Hawkins, A. Jones, G. Langley, T. Nurmikko, and I. Tracey. 2008. Volunteer studies in pain research—Opportunities and challenges to replace animal experiments—The report and recommendations of a focus on alternatives workshop. Neuroimage 42 (2): 467–473. Lockwood, J. A. 1987. The moral standing of insects and the ethics of extinction. Florida Entomologist 70 (1): 70–89. Macleod, M., and P. Sandercock. 2005. Systematic reviews improve clinical research design—Can they help improve animal experimental work? RDS News Winter: 8–10. Markou, A., C. Chiamulera, M. A. Geyer, M. Tricklebank, and T. Steckler. 2009. Removing obstacles in neuroscience drug discovery: The future path for animal models. Neuropsychopharmacology 34:74–89. Matthews, R. A. 2008. Medical progress depends on animal models—Doesn’t it? Journal of the Royal Society of Medicine 101 (2): 95–98. Morton, D. B. 2007. Experimental procedures: General principles and recommendations. In The welfare of laboratory animals, ed. E. Kaliste. Dordrecht, the Netherlands: Springer. Nagel, T. 1974. What is it like to be a bat? Philosophical Review 83 (4): 435–450. Narveson, J. 1983. Animal rights revisited. In Ethics and animals, ed. H. B. Miller and W. H Williams. Totowa, NJ: Humana Press. Nevalainen, T. 2004. Training for reduction in laboratory animal use. Atla-Alternatives to Laboratory Animals 32 (Supplement 2): 65–67. Olsson, I. A. S., and K. Dahlborn. 2002. Improving housing conditions for laboratory mice: A review of environmental enrichment. Laboratory Animals 36 (3): 243–270. Olsson, I. A. S., A. K. Hansen, and P. Sandoe. 2008. Animal welfare and the refinement of neuroscience research methods—A case study of Huntington’s disease models. Laboratory Animals 42 (3): 277–283. Olsson, I. A. S., P. Robinson, K. Pritchett, and P. Sandøe. 2002. Animal research ethics. In Handbook of laboratory animal science, 2nd ed., ed. G. van Hoosier and J. Hau. Boca Raton, FL: CRC Press. Regan, T. 1983. The case for animal rights. Berkeley: University of California Press. ———. 1989. The case for animal rights. In Animal rights and human obligations, ed. T Regan and P. Singer. Englewood Cliffs, NJ: Prentice Hall. Research Defense Society. http://www.rds-online.org.uk/pages/home.asp?i_ToolbarID=8&i_PageID=94 (accessed November, 26, 2008). Rowland, M. 2006. Microdosing and the 3Rs. NC3Rs 5:1–7. Russell, W., and R. Burch. 1959. The principles of humane experimental technique, 2nd ed. London: Methuen. Sandøe, P., and S. Christiansen. 2008. Ethics of animal use. Oxford, England: Blackwell. Schuppli, C. A., and D. Fraser. 2007. Factors influencing the effectiveness of research ethics committees. Journal of Medical Ethics 33 (5): 294–301. Scott, S., J. E. Kranz, J. Cole, J. M. Lincecum, K. Thompson, N. Kelly, A. Bostrom, J. Theodoss, B. M. Al-Nakhala, F. G. Vieira, J. Ramasubbu, and J. A. Heywood. 2008. Design, power, and interpretation of studies in the standard murine model of ALS. Amyotroph Lateral Sclerosis 9 (1): 4–15. Sena, E., H. B. van der Worp, D. Howells, and M. Macleod. 2007. How can we improve the pre-clinical development of drugs for stroke? Trends in Neurosciences 30 (9): 433–439. Sherwin, C. M. 2001. Can invertebrates suffer? Or, how robust is argument-by-analogy? Animal Welfare 10:S103–S18. Singer, P. 1991. Animal liberation. 2nd ed. London, England: Thorsons. Singer, P. 1993. Practical ethics. Cambridge, England: Cambridge University Press. Smith, J. A., and K. M. Boyd. 1991. Lives in the balance: The ethics of using animals in biomedical research. Oxford, England: Oxford University Press. Smith, J. A., F. A. R. van den Broek, J. C. Martorell, H. Hackbarth, O. Ruksenas, W. Zeller, and FELASA Working Group. 2007. Principles and practice in ethical review of animal experiments across Europe: Summary of the report of a FELASA Working Group on Ethical Evaluation of Animal Experiments. Laboratory Animals 41 (2): 143–160. Svendsen, O. 2005. Ethics and animal welfare related to in vivo pharmacology and toxicology in laboratory animals. Basic & Clinical Pharmacology & Toxicology 97 (4): 197–199. Tannenbaum, J. 1999. Ethics and pain research in animals. ILAR Journal 40 (3): 97–110.
84550.indb 36
11/11/10 6:24:03 PM
Ethics of Animal Research
37
Thapar, A., G. Harold, F. Rice, X. Ge, J. Boivin, D. Hay, M. van den Bree, and A. Lewis. 2007. Do intrauterine or genetic influences explain the fetal origins of chronic disease? A novel experimental method for disentangling effects. BMC Medical Research Methodology 7:25. van der Staay, F. J. 2006. Animal models of behavioral dysfunctions: Basic concepts and classifications, and an evaluation strategy. Brain Research Reviews 52 (1): 131–159. van der Worp, H. B., P. de Haan, E. Morrema, and C. J. Kalkman. 2005. Methodological quality of animal studies on neuroprotection in focal cerebral ischaemia. Journal of Neurology 252 (9): 1108–1114. Willner, P. 1984. The validity of animal models of depression. Psychopharmacology (Berlin) 83 (1): 1–16. Wolfer, D. P., O. Litvin, S. Morf, R. M. Nitsch, H. P. Lipp, and H. Wurbel. 2004. Cage enrichment and mouse behavior—Test responses by laboratory mice are unperturbed by more entertaining housing. Nature 432 (7019): 821–822.
84550.indb 37
11/11/10 6:24:03 PM
84550.indb 38
11/11/10 6:24:03 PM
Chapter 3
An Overview of Global Legislation, Regulation, and Policies
Kathryn Bayne, Bryan R. Howard, Tsutomu Miki Kurosawa, and Maria Eugenia Aguilar Nájera Contents Introduction.......................................................................................................................................40 Europe...............................................................................................................................................40 Changing Political Climate..........................................................................................................40 National Legislation.....................................................................................................................40 Europe-wide Regulation.............................................................................................................. 41 Current Developments................................................................................................................. 42 The New Directive....................................................................................................................... 43 Consequences...............................................................................................................................44 North America.................................................................................................................................. 45 United States of America............................................................................................................. 45 Cornerstones of an Animal Care and Use Program................................................................ 45 Federal Oversight of Animal Research, Testing, and Teaching.............................................. 48 Other Laws, Regulations, and Policies.................................................................................... 51 Canada.......................................................................................................................................... 52 Latin America (LA).......................................................................................................................... 53 Colombia......................................................................................................................................54 Costa Rica....................................................................................................................................54 Argentina......................................................................................................................................54 Uruguay........................................................................................................................................ 55 Mexico.......................................................................................................................................... 55 Cuba............................................................................................................................................. 56 Venezuela..................................................................................................................................... 56 Chile............................................................................................................................................. 56 Brazil............................................................................................................................................ 57 Peru and Guatemala..................................................................................................................... 57 Ecuador........................................................................................................................................ 57 Panama......................................................................................................................................... 57 Paraguay....................................................................................................................................... 57 Puerto Rico.................................................................................................................................. 57 39
84550.indb 39
11/11/10 6:24:04 PM
40
Handbook of Laboratory Animal Science
Nicaragua..................................................................................................................................... 57 Asia ................................................................................................................................................. 57 Japan............................................................................................................................................. 58 Korea............................................................................................................................................ 59 People’s Republic of China..........................................................................................................60 Taiwan (Chinese Taipei)..............................................................................................................60 Philippines................................................................................................................................... 61 Thailand....................................................................................................................................... 62 Acknowledgments............................................................................................................................. 62 References......................................................................................................................................... 63 Introduction With the steady increase in international collaborations in animal research, testing, and teaching, there is a concomitant increase in a need to be familiar with the standards of oversight in different countries. This task has been made difficult by the lack of a single, concise compilation of the laws, rules, regulations, policies, and guidelines in use around the world, and by their constant evolution toward a more restrictive stance. The goal of this chapter is to provide an introduction to the governance of animal use in a variety of countries, with more detail provided for certain parts of the world than others principally due to the greater complexity of standards to be met in those countries. Europe Changing Political Climate At the time of this writing (fall 2009), considerable change is taking place within the legislative climate in Europe. The number of European member states has increased from 15 in 1995 to 27, including several that formerly had close social and economic ties with the USSR; concerns for animal well-being and the regulation of the scientific uses of animals were previously a low priority for many of these countries. Technological developments have resulted in the availability of new animal models—particularly genetically altered animals, which are now increasingly used as models of human disease—and in preparation of gene knockouts and gene knock-ins as a means of better understanding biological processes, particularly as they relate to humans. Developments in scientific equipment make it possible to conduct long-term studies on unrestrained animals, continuously administer substances using implanted minipumps, and obtain information about important biological mechanisms using noninvasive imaging or telemetry equipment, thereby avoiding the need always to kill animals and process tissues postmortem. Changes are similarly taking place in the way in which animals are cared for; generally, laboratory animals are free of the principal clinical and subclinical infections common in the 1990s and are caged in ventilated racks or similar housing under conditions of much improved hygiene. It is debatable how much our knowledge has increased concerning the biological needs of the animals we care for, but most people are now more aware of perceived good practice. Against this backdrop, existing Europe-wide regulation is being reexamined with a view to bringing it up to date and ensuring wider compliance. National Legislation Historically, European nations had implemented laws that prohibited the ill treatment of animals. Among the earliest of these were Martin’s laws (1822) in the United Kingdom, which prohibited cruelty to sheep, cattle, and horses, and were expanded in 1835 to protect dogs and bulls and
84550.indb 40
11/11/10 6:24:04 PM
An Overview of Global Legislation, Regulation, and Policies
41
to outlaw cruel sports. More wide ranging legislation was later introduced into Britain, under the Protection of Animals Act of 1911. In France, the Grammont law of 1850 prohibited the ill treatment of animals, but only when this took place in public; it was repealed in 1959 by Decree No. 59-1051, which removed the requirement for witnesses to be present and hence greatly expanded its scope. The Norwegian Animal Welfare Act of 1935, revised in 1974 (and several times subsequently), required that animals (mammals, birds, toads, frogs, salamanders [newts], reptiles, fish, and crustaceans) must be treated well and account taken of their instinctive behavior and natural needs, to prevent causing them unnecessary suffering. In Switzerland, which is not a member of the European Union, the Federal Act on Animal Protection (1978) set out measures to ensure animal protection and welfare. The measures outlined the responsibilities of animal owners in order to help prevent cruelty to animals and provided guidelines to ensure animal welfare. Both this act and the Animal Protection Ordinance (1981) included measures to regulate the use of animals in experiments, including specifying the need to train staff involved in the conduct of such experiments, authorization of procedures, arrangements for inspection of facilities, etc. Switzerland ratified the European Convention ETS 123 (Council of Europe 1986) in November 1993 and implemented it in the following year. Switzerland also tends to implement the requirements of Directive 86/609 (European Commission 1986), although it is under no obligation to do so. Much of the earlier animal protection legislation remains in place today. In order to create a legislative framework that permits the conduct of animal experimentation, enabling legislation must be implemented that sets aside laws that safeguard animal welfare. This immediately raises substantial ethical issues for people who place a high regard on existing animal-welfare legislation. Some believe that it is wrong to set aside that legislation under any circumstances, although the majority recognizes that some exemptions may be necessary. In particular, governments attempt to create an economic climate that favors advances in understanding human and animal biology and medical applications of this knowledge—for example, by developing new treatments for diseases, such as effective medicines, new surgical approaches, and new therapeutic programs. In the UK, such enabling legislation was included in the Cruelty to Animals Act (1876), introduced in response to concerns among the public and the scientific community about the conduct of animal experiments in physiology, which at the time were often conducted with little or no anesthesia. Anesthetic techniques of the time were crude, and there was little understanding of the impact of procedures that we recognize today as painful would have on sentient animals. Indeed, for many, the sentience of animals was still questioned. The Cruelty to Animals Act for the first time specifically authorized carrying out experiments on living animals, but placed conditions on the way in which they were done, the reasons for which they were done, and the qualifications of persons responsible for performing them; however, there was little uniformity in the provisions introduced. Europe-wide Regulation In 1986, two pan-national regulatory developments within Europe attempted to introduce a level of uniformity to the ways in which experimental animals are kept and used. These were the convention for the protection of vertebrate animals used for experimental and other scientific purposes (ETS 123) (Council of Europe 1986) and the directive on the approximation of laws, regulations, and administrative provisions of the member states regarding the protection of animals used for experimental and other scientific purposes (86/609/EEC) (European Council 1986a). The Council of Europe is an intergovernmental body of 47 member states, established in 1949, whose objective is to promote human rights and democracy in Europe; one of the ways it does this is by issuing conventions. European conventions must be approved by the Committee of Ministers (its decision-making body) and constitute an open treaty that can be “signed” by member states of the council. They have no statutory force, but subject to sufficient support, parties who sign (express an interest) and then ratify the treaty are legally bound to implement its provisions.
84550.indb 41
11/11/10 6:24:04 PM
42
Handbook of Laboratory Animal Science
In the same year (1986), the European Commission introduced a directive (86/609) that further regulated the way in which animals could be kept and used (European Council 1986a). The European Commission is a politically independent institution that oversees the interests of the European Union as a whole and has powers to make proposals for legislation, policies, and programs of action and for implementing the decisions of Parliament and the council. A complex legislative process is involved, but it has powers (through the European Parliament) to introduce directives and regulations. A directive requires all members of the European Union to introduce, into their national law, measures that either match or exceed the conditions specified, whereas regulations become effective without further action at a national level. Nations that fail to implement legislation to enforce directives satisfactorily may be subject to European Court proceedings and subjected to fines. The directive of 1986, for “the protection of vertebrate animals used for experimental and other scientific purposes (86/609/EEC)” was adopted by the European Council. The intention of this directive was to ensure that when animals are used for scientific purposes, all member states would introduce similar measures, thereby ensuring uniform application of the principles of the three Rs (reduce, refine, replace) of Russell and Burch (1959). This was seen as a means of maintaining a free and open market and avoiding distorting competition or introduction of unfair trade barriers. The directive regulated the use of vertebrate animals from specified stages of development for any experimental or other scientific purpose that may cause them pain, suffering, distress, or lasting harm. It set out a series of minimum regulatory controls for the acquisition and care of the animals and for staff. It defined requirements for registration and control of establishments where animals are bred, kept, and used, and it required that some species, such as nonhuman primates, dogs, and cats, must be individually identified and their life-long records maintained. Statistical data were to be collected for EU-wide compilation and publication by the European Commission. The controls specifically excluded certain actions, such as humane killing or using the least painful method to allow an animal to be identified. The powers of the European Economic Community under the Treaty of Rome allowed it only to regulate animal experimentation conducted for commercial purposes; consequently, when the directive was approved, a resolution was also approved in which member states agreed to apply similar controls to all types of animal experimentation (European Council 1986b). Although one intention of Directive 86/609/EEC had been to harmonize the regulation of animal experimentation in the EU, some member states chose to enact further reaching legislation whereas others applied only minimum rules. Additionally, the legal provisions made for enforcement differed in different countries. For example, procedures for granting of project authorization, authorization of reuse of animals, and arrangements for national inspections vary widely in different member states. Similarly, requirements for the collection and publication of annual statistics are implemented in different ways by different states; for example, some reporting requirements include numbers of animals killed for use of tissues whereas others exclude these. In part, these differences arise from the different agencies that states have identified as enforcement bodies (competent authorities). In the UK, this is the Home Office; in France, the Ministry of Agriculture; in the Netherlands, the Ministry of Health, Welfare and Sport; and, in Italy, the Ministry of Health. Current Developments Appendix A of the Council of Europe Convention ETS 123 lays down details of the ways in which experimental animals may be kept: husbandry provisions, cage sizes, environmental conditions, etc. The convention incorporates a requirement for the parties to review and, where necessary, update the requirements of Appendix A. Over a 5-year period from the turn of the century, a substantial review of recommendations was carried out; as a result, a number of changes were introduced in 2006. Specifically, the changes related to the way in which laboratory animals are required to be housed (Council of Europe 2006).
84550.indb 42
11/11/10 6:24:04 PM
An Overview of Global Legislation, Regulation, and Policies
43
Previously, in 1998, the European Commission had ratified the Council of Europe Convention ETS 123 (subject to exclusion of the requirement to communicate certain statistical data) and incorporated the provisions into Directive 86/609 (European Council 2003). Because of this, the terms of the convention are now binding on all member states, rather than only those who chose to sign the convention in their own right. In November 2008, the European Commission published a draft proposal for a new directive to replace that enacted in 1986. The text of the proposal had been many years in preparation and involved widespread discussion among interested parties, including the scientific community and animal protection organizations. As part of the process of revising the directive, an Internet consultation was carried out between June and August 2006 to determine public opinion in EU countries about the scientific uses of animals. Three quarters of respondents believed that the level of welfare and protection of experimental animals within the EU was either poor or very poor and only 40% agreed with the use of animals to develop medical treatments for disease. This was a self-selected survey susceptible to influences by campaigning organizations, but it suggested the considerable public concern that undoubtedly influenced the European Commission in the measures it has recommended. At the time of this writing, the proposal has been presented to both the European Parliament (composed of members elected by each nation) and the European Council (which consists of representatives from the member states). Considerable negotiation is under way to achieve final agreement of all three bodies—the commission, council, and parliament—and a number of amendments have already been proposed. The New Directive Although the final text of the new directive has yet to be finalized, a number of common themes have emerged that are very likely to be included. Like the previous directive, the new one will specify circumstances under which animals may be used for experimental purposes. It will require that animals be used for experimental purposes only if there is no suitable alternative and then only under circumstances that ensure that the minimum number is used. The species and methodology chosen should be the ones least likely to result in pain, suffering, distress, or lasting harm, and the experiment should be likely to yield a satisfactory scientific result. Restrictions will be placed on the reuse of animals. Breeders, suppliers, and users of laboratory animals must all be authorized by the authority responsible for implementing the national legislation (the “competent authority”) and such authorization shall be renewable after a fixed period. Member states will be expected to establish inspection procedures to ensure that the terms the directive are adhered to, and the commission itself is likely to retain power to carry out its own inspections when it deems this necessary. One impact of the new directive will be a focus on scientific projects as the principal point of control. Each project will need to be submitted for authorization to the competent authority. The application must include not only an explanation of the way in which the experiment has been designed, the reason for selection of the model, and a description of the procedures to be carried out, but also a nontechnical summary (this may not be mandatory in all cases). The applicant must also provide evidence that the three Rs have been implemented fully, including the use of humane end points where appropriate. Some tests still occasionally used in toxicity testing, such as the LD50, in which death is used as a final end point, will be prohibited unless there really is no alternative. Applications will be evaluated from a point of view of their scientific justification and evidence for implementation of the three Rs. It is likely that the severity of procedures will need to be classified according to a scheme established by the European Commission; this will comprise four categories: nonrecovery, mild, moderate, and severe. Procedures likely to cause severe and persistent pain, suffering, or distress that cannot be alleviated are likely to be prohibited.
84550.indb 43
11/11/10 6:24:04 PM
44
Handbook of Laboratory Animal Science
In deciding whether or not to approve project applications, the competent authority must consider the expertise of those involved in designing the experiments and ensuring the three Rs are implemented and require reassurance that animal husbandry standards are satisfactory and that there is adequate veterinary care. The competent authority will be required to decide whether or not to authorize the project within a fixed time period (currently 40 working days). In some circumstances, member states may be allowed to introduce simplified administrative procedures for projects that do not involve the use of nonhuman primates and are classified as moderate, mild, or nonrecovery. A system of retrospective project assessment may be required to establish whether or not the objectives of the project were achieved, whether the harms inflicted on animals accorded with those anticipated beforehand, and whether all aspects of the three Rs were adequately implemented. All projects involving the use of nonhuman primates or procedures judged to be severe are likely to be subject to retrospective assessment. Another major change from the 1986 directive is a move toward more uniform and formal application of the three Rs. One aspect of this is likely to be a requirement for facilities at which animal experimentation takes place to establish a committee responsible for ensuring implementation of the three Rs at a local level. These have been called “animal welfare bodies” or “ethical review bodies,” but their remit will be to ensure that animal welfare receives a high priority at all stages of an animal’s life, from the time it is born or arrives at the establishment until the time it is killed. It is also anticipated that each member state will be required to set up a national committee to advise the relevant competent authority and other interested parties about matters relating to the acquisition, breeding, accommodation, care, and use of animals so as to ensure best practices. In line with the requirements of the revised European convention, standards of care and accommodation of animals will be expected to ensure that any restrictions on the extent to which animals can satisfy their physiological and ethological needs are minimized and that the environment and condition of animals be checked each day. There may also be a requirement for staff involved in the care and use of animals to be trained adequately. Animals experiencing avoidable pain, suffering, distress, or lasting harm must be identified promptly and palliative measures introduced as far as possible. Reflecting public concerns, restrictions are likely to be placed on the breeding and use of nonhuman primates. There will probably be a move to prohibit the capture of wild primates for experimental purposes and also to end their use for breeding so that, after an appropriate transition period, only the offspring of nonhuman primates that have been bred in captivity will be used in procedures. Almost certainly, the use of nonhuman primates will be subject to additional constraints, including stronger scientific justification, and there is likely to be a complete ban on the use of great apes except in very special circumstances. At the time of writing (September 2010) the new Directive has just completed its second reading by the European Parliament and awaits minor revisions of a technical nature, translation, and publication. A period of time will be allowed for member states to implement requirements of the Directive into their national legislation, and it is unlikely that this will be achieved before early 2013. Probably, all member states will find it necessary to make changes to their national legislation; in many cases, these changes may be substantial. Many European countries have decided to avoid making adjustments to their legislation until the requirements the new directive have been published. Despite this, as a condition of entry, new entrants to the European Union are required to comply with the existing directive (86/609) and are therefore obliged to implement provisional legislation, which presumably will have to be changed when the new directive is introduced. Consequences Provisions of the European directive must be implemented in national law and, consequently, it is argued that a detailed and finely structured directive will result in broadly similar legislative arrangements within individual countries. Creation of a level economic playing field also provides
84550.indb 44
11/11/10 6:24:04 PM
An Overview of Global Legislation, Regulation, and Policies
45
opportunities for the more uniform application of general principles, including application of the three Rs, identifying which types of experiments may be permitted or are excluded, ensuring that studies are carried out by competent persons, and ensuring that animals are appropriately sourced, transported, cared for, and used. Establishment of a relatively uniform legislative framework makes it easier for those engaged in commercial or academic investigations that involve laboratory animal science to plan and implement transnational activities, thus facilitating the free movement of people. Individuals should find it easy to adapt to the legislative requirements of different countries. It remains to be seen whether these outcomes will be achieved. Because of the broad level of consensus necessary for introduction of a directive, there is wide public consultation, which provides an opportunity for the views of all stakeholders to be considered. The bigger the pool of ideas is, the more likely it is that best practice will be identified. Collection of a broad, shared experience from a wide range of experts and the general public ensures that European citizens feel that they can contribute to law making. This can have drawbacks, however, because it provides considerable opportunity for organized lobby groups to exert focused pressure on the legislative program within Europe. Thus, resulting legislation may be distorted by attempts to accommodate extreme viewpoints rather than creating law that is broadly satisfactory to the majority of the European population. In consequence, the legislation eventually drafted may prove to be less flexible than originally intended. It can be difficult to take account of differing national cultures and interests—for example, the preponderance of studies involving fish, which have great economic importance in Norway, rather than more conventional warm-blooded vertebrates. Proposals for reporting numbers of animals used and strictures on how they are used may not always be drafted in a way that makes implementation straightforward. One further disadvantage of centralized legislative procedures is that some citizens may feel that the decision-making process is so remote that they cannot contribute effectively to the development of law. It is interesting that in the survey conducted by the European Union preparatory to developing the replacement for Directive 86/609, most citizens polled felt that standards of care and use of laboratory animals within their own country were higher than the average for the European Union.
North America United States of America In the United States, oversight of animal care and use for research, testing, and teaching is achieved by numerous laws, regulations, policies, and guidelines from two principal government organizations: the U.S. Department of Agriculture (USDA) and the U.S. Public Health Service (PHS). Other guidance may be derived from scientific panels (e.g., through the Institute for Laboratory Animal Research [ILAR]) and be endorsed by the government as required standards. Institutions that meet certain standards may participate in the voluntary accreditation program offered by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC International), which provides external peer review of an institution’s program and further quality assurance. Cornerstones of an Animal Care and Use Program An institution that uses animals for research, education, or testing purposes must determine which of the many federal and state regulations, policies, and guidelines to follow. Although an institution may be obligated to follow a variety of standards, some consistent elements distinguish successful animal care and use programs. These are discussed next.
84550.indb 45
11/11/10 6:24:04 PM
46
Handbook of Laboratory Animal Science
Institutional Animal Care and Use Committee The group of individuals comprising the institutional animal care and use committee (IACUC) represents institutional and public interests and has the responsibility for oversight and evaluation of the entire animal care and use program and facilities. Because committee members act on behalf of the institution, their role is pivotal to engendering a humane and progressive animal care and use program. The successful program is overseen by a committee that is engaged and knowledgeable and receives strong administrative support. Because the committee is responsible for investigating reports of concern regarding animal welfare, its functions must be well known throughout the institution and there must be ready (and confidential) access to the committee membership. Training The importance of adequate training for all those involved in the animal care and use program is underscored by the emphasis it receives in the USDA’s Animal Welfare Regulations (USDA 1991), PHS Policy on Humane Care and Use of Laboratory Animals (PHS Policy, OLAW 2002), and the Guide for the Care and Use of Laboratory Animals (National Research Council 1996). The Animal Welfare Regulations and the PHS Policy require institutions to ensure that people caring for or using animals are qualified to do so. The Animal Welfare Regulations stipulate several key topics be included in the institution’s training program: • Humane methods of animal maintenance and experimentation, including the basic needs of each species of animal, proper handling and care for the various species of animals used by the institution, proper preprocedural and postprocedural care of animals, and aseptic surgical methods and procedures • The concept, availability, and use of research or testing methods that limit the use of animals or minimize animal distress • Proper use of anesthetics, analgesics, and tranquilizers for any species of animal at the institution • Methods to report any deficiencies in animal care and treatment • Use of the services at the National Agricultural Library, such as appropriate methods of animal care and use, alternatives to the use of live animals in research, prevention of unintended and unnecessary duplication of research involving animals, and information regarding the intent and requirements of the Animal Welfare Act
The Guide urges adequate training be provided to members serving on the IACUC so that they can appropriately discharge their responsibilities. In addition to the IACUC members, the Guide recommends that the professional and technical personnel caring for animals be trained, as well as investigators, research technicians, trainees (including students), and visiting scientists. The Guide also endorses training in occupational health and safety, in procedures specific to an employee’s job, and in procedures specific to research (e.g., anesthesia, surgery, euthanasia, recognition of the signs of pain and/or distress, etc.). Occupational Health and Safety Although not mandated by the Animal Welfare Regulations, the Guide and thus the PHS Policy require that an occupational health and safety program be in place that is specific to the animal care and use program. The details of the occupational health and safety program will vary among institutions but will be predicated on the experimental and nonexperimental hazards identified at each institution and an assessment of the risks posed to personnel (by either job classification or the health of the individual) by these hazards. The Guide emphasizes that health professionals (doctors or nurses, as appropriate) should be involved in the design and implementation of the program. Participation by individuals involved in the animal care and use program should be based on “the
84550.indb 46
11/11/10 6:24:05 PM
An Overview of Global Legislation, Regulation, and Policies
47
hazards posed by the animals and materials used; on the exposure intensity, duration, and frequency; on the susceptibility of the personnel; and on the history of occupational illness and injury in the particular workplace.” Several federal standards and regulations have been published that must be incorporated into the occupational health and safety program, depending on the species and hazardous agents in use: for example, Biosafety in Microbiological and Biomedical Laboratories (Centers for Disease Control and Prevention/National Institutes of Health 2007), Occupational Safety and Health Administration’s Blood-borne Pathogen Standards (Occupational Safety and Health Administration 2001), and Recombinant DNA Guidelines (National Institutes of Health 2009). Adequate Veterinary Care The Animal Welfare Regulations and the PHS Policy stipulate that the veterinarian must have the authority to oversee several key components of the animal care and use program, including animal procurement and transportation; quarantine, stabilization, and separation of animals; surveillance, diagnosis, treatment, and control of disease; surgery; the selection of analgesic and anesthetic agents; method of euthanasia; animal husbandry and nutrition; sanitation practices; zoonosis control; and hazard containment. The veterinarian must be qualified through either experience or training in laboratory animal medicine or in the species being used. The veterinarian brings a specific perspective to the deliberations of the IACUC and is a voting member. The Animal Welfare Regulations describe the program of adequate veterinary care as including the following: • Availability of appropriate facilities, personnel, equipment, and services • Use of appropriate methods to prevent, control, diagnose, and treat diseases and injuries, inclusive of the availability of emergency, weekend, and holiday care • Daily observation of all animals to assess their health and well-being • Guidance to researchers regarding handling, immobilization, anesthesia, analgesia, tranquilization, and euthanasia • Nutrition • Pest and parasite control • Adequate preprocedural and postprocedural care in accordance with current professional standards (see also APHIS Tech Note March 1999, Animal Care Policy #3, and APHIS Form 7002) (Animal Plant Health Inspection Service 1992, 1999)
The Report of the American College of Laboratory Animal Medicine on Adequate Veterinary Care in Research, Testing and Teaching (American College of Laboratory Animal Medicine 1996) describes a program of adequate veterinary care as including (1) disease detection and surveillance, prevention, diagnosis, treatment and resolution; (2) guidance on anesthetics, analgesics, tranquilizer drugs, and methods of euthanasia; (3) review and approval of all preoperative, surgical, and postoperative procedures; (4) promotion and monitoring of an animal’s well-being before, during, and after its use; and (5) involvement in the review and approval of all animal care and use at the institution. This report is used by AAALAC International as a reference standard in its assessments of animal care and use programs. Resources The infrastructure of the animal care and use program, including facilities, equipment, number and qualifications of personnel, and genetic and health status of the animals used, has a significant influence on the quality of the program. The Animal Welfare Regulations specify the size of the primary enclosures in which animals are to be kept. The Guide describes several environmental variables—such as temperature, ventilation, illumination, sanitation standard, and cage size, as
84550.indb 47
11/11/10 6:24:05 PM
48
Handbook of Laboratory Animal Science
well as the components of the physical plant—that can facilitate the research, testing, or teaching goals of the institution and states that “a well-planned, well-designed, well-constructed, and properly maintained facility is an important element of good animal care and use” (National Research Council 1996). AAALAC International has identified that physical plant deficiencies (specifically, the operation of the heating, ventilation, and air conditioning systems) rank as the third most common concern (following the scope of the occupational health and safety program and the functioning of the IACUC) requiring correction before a full accreditation status can be granted. In this manner, the impact of the facility on the safety and well-being of the animals is underscored. Federal Oversight of Animal Research, Testing, and Teaching Federal laws for the humane treatment of animals have been in place since 1873, when Congress passed a law governing the treatment of livestock during shipment for export. The law was called the 28 Hour Law after the maximum length of time animals could be transported before receiving food, water, and rest (Anderson 2002). This law was later repealed and a new 28 Hour Law was passed in 1906 that is still in effect today. However, the first federal law to protect nonfarm animals was not passed until 1966: the Laboratory Animal Welfare Act, administered by the Animal and Plant Health Inspection Service (APHIS) of the USDA. At the time, this law was primarily directed at dog and cat dealers and required that individuals or corporations that bought or sold dogs or cats for laboratory activities be licensed and adhere to certain minimum standards for the care of animals and that users of cats or dogs for research register with the USDA and meet minimum standards for animal care. For animal users, the law applied only to animals held prior to or after the laboratory activity. Interestingly, the New York Anticruelty Bill of 1866 addressed the use of animals in research and predated federal interest in this subject (Rozmiarek 2007). The Laboratory Animal Welfare Act of 1966 was amended in 1970, 1976, 1985, 1990, 2002, and 2008 to broaden its coverage. Public Law 91-579, the Animal Welfare Act of 1970, increased the species of animals covered under the law to include all warm-blooded animals and increased the scope of applicability of the law to include the time animals were held in the facility. Specifically exempted were horses not used in research and agricultural animals used in food and fiber research, retail pet stores, state and county fairs, rodeos, purebred cat and dog shows, and agricultural exhibitions. Public Law 94-279, the Animal Welfare Act Amendments of 1976, included common commercial carriers, such as airlines, under the law, which subsequently led to standards being developed for shipping containers and conditions of shipment. Public Law 99-198, the Improved Standards for Laboratory Animals Act, added several new provisions to the law, including minimization of animal pain and distress and consideration of alternatives to painful procedures, consultation with a doctor of veterinary medicine for any practice that could cause pain to animals, limitation on conducting more than one major survival surgery on an animal, establishment of an IACUC to provide oversight of the animal care and use program and facilities, provision of specific training to personnel, provision of exercise to dogs, and a stipulation to promote the psychological well-being of nonhuman primates. The 1990 amendment to the Animal Welfare Act, Public Law 101-624 Food, Agriculture, Conservation, and Trade Act of 1990, Section 2503, Protection of Pets, established a holding period for dogs and cats at shelters and other holding facilities prior to sale to dealers. The law also requires dealers to provide written certification to the recipient regarding the background of each animal. Of increasing debate has been the exclusion of rats and mice from the Animal Welfare Act. The 1970 amendment to the Animal Welfare Act stated that an animal was defined as “any live or dead dog, cat, monkey (nonhuman primate animal), guinea pig, hamster, rabbit, or other such warm-blooded animal as the Secretary may determine is being used, or is intended for use, for research, testing, experimentation, or exhibition purposes, or as a pet.” In this way, the Secretary
84550.indb 48
11/11/10 6:24:05 PM
An Overview of Global Legislation, Regulation, and Policies
49
of the Department of Agriculture was provided the authority to determine which animals would be covered by the act. In 1977, the USDA promulgated regulations that specifically excluded rats, mice, and birds used for research from the definition of “animal.” In 2002, Senator Jesse Helms added an amendment to the AWA in the Farm Bill, signed by President George W. Bush on May 13, 2002, that redefined the term “animal” in the law to match the current definition in the regulations. This change means that the definition of “animal” in the Animal Welfare Act excludes “birds, mice of the genus Mus, and rats of the genus Rattus, bred for use in research” from the definition of “animal.” By changing this term, the USDA does not have the authority to regulate animals excluded by the new definition. However, the USDA general counsel has determined that the uses of these animals for other purposes are now covered by the law. Since the 1966 Act, Congress has vested the USDA with both promulgation and enforcement authority. The USDA is required to conduct unannounced annual inspections of research facilities, with follow-up inspections until any cited deficiency has been corrected. Exempt from this provision are federal research facilities. Researchers, intermediate handlers, and common carriers are required to register with the USDA, while animal dealers and exhibitors must be licensed. Research facilities and U.S. government agencies are required to purchase animals only from licensed sources, unless the source is exempted from obtaining a license. Failure to comply with regulatory requirements, despite formal notification of an item or items of noncompliance and an opportunity to effect a correction, can result in fines levied on the facility, suspension of authority to operate, and even permanent revocation of the facility’s license to operate. Thus, the enforcement arm of the USDA’s oversight responsibility is strong and has been used over the years to improve animal welfare at dealers, exhibits, and research facilities. The other federal agency charged with oversight of research animal care and use is the PHS. The PHS Policy was implemented in 1973 and was revised in 1979 and 1986. Today, the PHS authority is derived from Public Law 99-158, the Health Research Extension Act of 1985, Section 495, Animals in Research. Under this Act, institutions conducting animal research using PHS funding, such as through the National Institutes of Health, must comply with the PHS Policy. The policy requires submission by the funding recipient (referred to as an “awardee institution”) of an Animal Welfare Assurance Statement, which must be approved by the PHS’s Office of Laboratory Animal Welfare (OLAW), National Institutes of Health, which commits the institution to follow the U.S. Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing, Research, and Training (Interagency Research Animal Committee 1985) (see Table€ 3.1) and the Guide (National Research Council 1996). Because the PHS Policy’s (OLAW 2002) standards for animal care and use are based on the Guide, the PHS Policy covers all vertebrate animals used in research, testing, or teaching. In addition to stating a commitment to animal welfare, the Assurance Statement must designate clear lines of authority and responsibility for institutional oversight of the work, inclusive of a designated “Institutional Official,” who is ultimately responsible for the animal care and use program; identify a qualified veterinarian involved in the program; provide a description of the occupational health and safety program for relevant personnel in the program; provide a description of mandated training; and provide a description of the facility. In the Assurance Statement, the institution must indicate whether the animal care and use program is reviewed by a third party, such as the Association for Assessment and Accreditation of Laboratory Animal Care International, or the program and facilities are reviewed solely by internal systems of the institution. Institutions in this latter category must provide a copy of their most recent semiannual report with the assurance. The Assurance is renegotiated with OLAW every 5 years. OLAW can approve, disapprove, restrict, or withdraw approval of the Assurance. PHS awarding agencies, such as the NIH, may not make an award for an activity involving live vertebrate animals unless the prospective awardee institution and all other institutions participating
84550.indb 49
11/11/10 6:24:05 PM
50
Handbook of Laboratory Animal Science
Table€3.1╅U.S. Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing, Research, and Training (IRAC 1985) The development of knowledge necessary for the improvement of the health and well-being of humans as well as other animals requires in vivo experimentation with a wide variety of animal species. Whenever U.S. Government agencies develop requirements for testing, research, or training procedures involving the use of vertebrate animals, the following principles shall be considered; and whenever these agencies actually perform or sponsor such procedures, the responsible Institutional Official shall ensure that these principles are adhered to: I. The transportation, care, and use of animals should be in accordance with the Animal Welfare Act (7 U.S.C. 2131 et seq.) and other applicable Federal laws, guidelines, and policies. II. Procedures involving animals should be designed and performed with due consideration of their relevance to human or animal health, the advancement of knowledge, or the good of society. III. The animals selected for a procedure should be of an appropriate species and quality and the minimum number required to obtain valid results. Methods such as mathematical models, computer simulation, and in vitro biological systems should be considered. IV. Proper use of animals, including the avoidance or minimization of discomfort, distress, and pain when consistent with sound scientific practices, is imperative. Unless the contrary is established, investigators should consider that procedures that cause pain or distress in human beings may cause pain or distress in other animals. V. Procedures with animals that may cause more than momentary or slight pain or distress should be performed with appropriate sedation, analgesia, or anesthesia. Surgical or other painful procedures should not be performed on unanesthetized animals paralyzed by chemical agents. VI. Animals that would otherwise suffer severe or chronic pain or distress that cannot be relieved should be painlessly killed at the end of the procedure or, if appropriate, during the procedure. VII. The living conditions of animals should be appropriate for their species and contribute to their health and comfort. Normally, the housing, feeding, and care of all animals used for biomedical purposes must be directed by a veterinarian or other scientist trained and experienced in the proper care, handling, and use of the species being maintained or studied. In any case, veterinary care shall be provided as indicated. VIII. Investigators and other personnel shall be appropriately qualified and experienced for conducting procedures on living animals. Adequate arrangements shall be made for their in-service training, including the proper and humane care and use of laboratory animals. IX. Where exceptions are required in relation to the provisions of these Principles, the decisions should not rest with the investigators directly concerned but should be made, with due regard to Principle II, by an appropriate review group such as an institutional animal care and use committee. Such exceptions should not be made solely for the purposes of teaching or demonstration.
in the animal activity have an approved Assurance with OLAW and provide verification that the IACUC has reviewed and approved those sections of the grant application that involve the use of animals. Applications from organizations with approved Assurances must address five specific points pertaining to the use of animals: • A detailed description of the proposed work, including species, strain, sex, age, and number of animals to be used in the proposed work • A justification of the use of animals, species, and number of animals • Information on the veterinary care for the animals • A description of the procedures for ensuring that discomfort, distress, pain, and injury will be minimized • A description of the method of euthanasia and the reason for the selection of that method, including a justification for any method that does not conform with the American Veterinary Medical Association’s (AVMA) euthanasia guidelines
Awardee institutions that do not comply with the standards of the Guide, the USDA Animal Welfare Regulations (USDA 1991), and other standards referenced in the PHS Policy (e.g., the AVMA’s Euthanasia Guidelines; AVMA 2007) may have their Assurance Statement restricted,
84550.indb 50
11/11/10 6:24:05 PM
An Overview of Global Legislation, Regulation, and Policies
51
Table€3.2╅Composition and Functions of the Institutional Animal Care and Use Committee USDA (Minimum of Three Members)
PHS (Minimum of Five Members)
Chairman Doctor of veterinary medicine Not affiliated with the institution
Doctor of veterinary medicine Practicing animal research scientist Nonscientist Not affiliated with the institution
1. Review at least once every 6 months the institution’s program for humane care and use of animals. 2. Inspect at least once every 6 months all of the institution’s animal facilities (including satellite facilities/ animal study areas). 3. Prepare reports of the IACUC evaluations conducted and submit the reports to the institutional official. 4. Review and investigate concerns involving the care and use of animals at the institution. 5. Make recommendations to the institutional official regarding any aspect of the institution’s animal program, facilities, or personnel training. 6. Review and approve, require modifications in (to secure approval), or withhold approval of those components of activities related to the care and use of animals. 7. Review and approve, require modifications in (to secure approval), or withhold approval of proposed significant changes in ongoing activities regarding the care and use of animals in ongoing activities. 8. Be authorized to suspend an activity involving animals.
which in turn can limit access to PHS funding for research. Sustained noncompliance with the PHS Policy can result in withdrawing the approval of the Assurance and cessation of all PHS funding for animal-based activities. The awardee institution must also submit an annual report. Institutions must report any change in category status from that noted in the Assurance Statement. Institutions indicate the dates of their IACUC semiannual program reviews and facility inspections and provide copies of any “minority views” filed by IACUC members with the annual report. The role of the IACUC in providing local oversight of animal care and use is a key element of the PHS Policy. Although the required composition of the IACUC for the PHS differs slightly from USDA requirements, due to the memorandum of understanding concerning laboratory animal welfare among APHIS/USDA, the Food and Drug Administration (FDA), and the NIH that sets forth procedures for cooperation among the three agencies in their oversight of animal care and use programs, the general functions and responsibilities of the IACUC are similar (see Table€3.2). OLAW conducts site visits of awardee institutions “for cause” and “not for cause.” In addition, an ongoing significant mission of OLAW is the educational outreach it performs in collaboration with awardee institutions. Jointly sponsored workshops focus on information of value to Institutional Officials and IACUCs to provide appropriate oversight of animal care and use. OLAW also provides guidance through articles in journals, commentary on other articles, NIH guide notices, and a listserv. Other Laws, Regulations, and Policies In 1978, the FDA promulgated regulations for the conduct of animal-based research of new or existing pharmaceutical agents, food additives, or other chemicals. These regulations, known as the good laboratory practice (GLP) regulations (Code of Federal Regulations 1998), specify appropriate diagnosis, treatment, and control of disease in animals used in this work. The Environmental Protection Agency (EPA) has issued companion regulations for conducting research pertaining to health effects, environmental effects, and chemical fate testing in a separate set of GLP regulations (Code of Federal Regulations 1997). GLP regulations of both the FDA and EPA rely heavily on adequate and detailed record keeping. Records must include standard operating procedures,
84550.indb 51
11/11/10 6:24:06 PM
52
Handbook of Laboratory Animal Science
animal identification, food and water analysis, documentation that any pesticides or chemicals used near the animals do not interfere with the study, and documentation of any disease and treatments that animals experience. On-site inspections are conducted to ensure compliance with the GLP standards. The Department of Defense (DoD) developed a “Policy on Experimental Animals” in 1961 to ensure that all research at DoD facilities involving animals was conducted in accord with certain principles of animal care (Department of Defense 1995). Later versions of this policy included overseas sites. Subsequently, a joint regulation, entitled “The Use of Animals in DoD Program,” from the Army, Navy, Air Force, Defense Nuclear Agency, and Uniformed Services University required all DoD facilities to “seek accreditation by AAALAC” and to establish local institutional animal care and use committees. State laws to protect animals have a long history, with the first anticruelty law passed in 1641 in the Massachusetts Bay Colony to prevent riding or driving farm animals beyond established limits (U.S. Congress Office of Technology Assessment 1986). All 50 states and the District of Columbia have enacted anticruelty laws. The overarching goals of these laws are to protect animals from cruel treatment, require that animals have access to suitable food and water and shelter from extreme weather. Some state laws define “animal” and some do not. The state laws encompass a diversity of approaches to providing protection to animals. Some states have additional provisions for animals used in research, and many states prohibit the sale of pound animals into the research stream. In general, criminal penalties are imposed for offenses. On occasion, state anticruelty laws have been used against research facilities. In recent years, state and federal laws have been used by private citizens or citizen groups claiming “standing to sue” on behalf of animals. The issue of “standing” has undergone a long litigation process and a chronology of court decisions on this issue has been compiled by the National Association for Biomedical Research (1999). Because animal research can involve a variety of different species, several other federal acts, laws, and treaties have bearing on animal use. These include the U.S. Endangered Species Act, which restricts the research conducted on these animals to research that would directly benefit the species under investigation; the Marine Mammal Protection Act, which provides authority for scientific research on marine mammals by special permit; the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES), which requires signator countries to obtain a permit for the import or export of certain species; the Lacey Act, which governs import, export, and interstate commerce of foreign wildlife; and the Migratory Bird Treaty Act, which makes it unlawful to take or possess any protected bird except by permit. Canada The Canadian constitution precludes federal legislation pertaining to the use of animals in research, testing, or education because such use is under provincial jurisdiction. Six provinces have established legislation regarding animal research, five of which reference the Canadian Council on Animal Care (CCAC) guidelines and policies. In addition, although there is no federal requirement to participate in the CCAC assessment program, the two principal funding agencies require grantee institutions to have a Certificate of Good Animal Practice• and to comply with CCAC guidelines and policies for continued funding. Contractors performing work for the federal government are required to adhere to CCAC guidelines, as specified in the Public Works and Government Services Canada Standard Acquisition Clauses and Conditions Manual, Section 5, Subsection A, Clause A9015C: Experimental Animals. The CCAC, founded in 1968, places responsibility for humane animal care and use with the animal care committee (ACC) at each institution. The ACCs are granted specific authority and provided with terms of reference under which they operate (e.g., membership, authority, responsibilities, and functioning). The CCAC’s mission is “to act in the interests of the people of Canada to ensure
84550.indb 52
11/11/10 6:24:06 PM
An Overview of Global Legislation, Regulation, and Policies
53
through programs of education, assessment and persuasion that the use of animals, where necessary, for research, teaching and testing employs optimal physical and psychological care according to acceptable scientific standards, and to promote an increased level of knowledge, awareness and sensitivity to relevant ethical principles.” Thus, the CCAC has two principal functions: (1) developing guidelines and policies to govern experimental animal care and use, and (2) monitoring compliance with those guidelines and policies. The CCAC is an independent organization and receives funding from the Medical Research Council (MRC) and the Natural Sciences and Engineering Research Council (NSERC). The CCAC establishes guidelines for its certified institutions to follow; these are currently contained in the two-volume Guide to the Care and Use of Experimental Animals (Canadian Council on Animal Care 1984, 2000). Adjunct guidelines address topics such as animal use protocol review, transgenic animals, selecting appropriate endpoints, and developing an animal user training program. The CCAC also has established several policies, including the ethics of animal research, review of scientific merit, social and behavioral requirements of experimental animals, acceptable immunological procedures, and categories of invasiveness. On-site assessments using panels of experts from the animal care and use community and a representative nominated by the Canadian Federation of Humane Societies are conducted triennially. An institution is deemed to be in compliance if the CCAC report prepared by the assessment panel and approved by the assessment committee (a standing committee composed of at least four council members) contains only regular, minor, and/or commendatory recommendations, and the institution submits an implementation report for any regular recommendations that is judged to be satisfactory. Institutions found to be in compliance or conditional compliance will receive a CCAC Certificate of Good Animal Practice. If the CCAC report contains major and/or serious recommendations whose correction does not require verification by an on-site reassessment, but rather can be verified through documentation, and the institution provides to the CCAC an implementation report that is judged to be satisfactory, then compliance is maintained. An assessment report containing major or serious recommendations may place the institution in a status of conditional compliance, probation, or noncompliance. All relevant funding agencies and government ministries and departments are notified of an institution’s noncompliance with CCAC guidelines (Canadian Council on Animal Care 2000). Sustained noncompliance with CCAC guidelines and policies can ultimately result in withdrawal of all animalbased research funding to the institution. Latin America (LA) As is true in the rest of the world, LA has a wide range of public and private institutes, centers, and institutions that carry out experimental procedures that involve the use of animals. With the purpose of meeting international requirements and to ensure appropriate treatment of these animals, most of LA countries have included in their legislation some reference regarding animal care, use, and welfare. In some cases, this has entailed taking legislation applied in other countries, such as the United States and Canada, as a model; in other instances, the country has or has created its own legislation in which aspects from other countries’ legislation are included. In many countries, standards regarding laboratory animal care and use are, at present, considered under more general laws for the protection and care of animals, which also describe guidelines for the treatment of domestic, companion, decoration, and exotic species. Some countries, such as Argentina, Chile, and Brazil, have bills specifically designed for laboratory animals that are under review for approval by the local authorities. Relative to this, it is important to mention that, since 1999, Mexico has had a federal regulation (NOM-062-ZOO-1999, Technical Specifications for the Production, Care and Use of Laboratory
84550.indb 53
11/11/10 6:24:06 PM
54
Handbook of Laboratory Animal Science
Animals) and was the first country in LA to legislate specifically on the subject. Argentina has three laws or regulations issued by independent organizations (one by the National Drug and Food Administration and the other two by the National Service of Agrifood Health and Quality of the Argentina Republic (Servicio Nacional de Sanidad y Calidad Agroalimentaria [SENASA]) that complement one another and regulate the use of laboratory animals in quality control procedures, teaching, and investigation, as well as animal facility operations. In a similar way, Costa Rica has an animal welfare law (republic law) in which a whole chapter is dedicated to experimental animals. This law has a special appendix (Science and Technology Ministry Decree 26668) that goes into detail of some aspects regarding laboratory animal production, care, and use. The laws, regulations, and guidelines that regulate the use of experimental animals in LA countries, as well as the date on which and purpose for which they were created, are summarized in the following sections. Colombia In 1989, the 84 Law of December 27, 1989, adopted the National By-Law of Animal Protection, which includes a special chapter dedicated to the use of experimental animals with investigation purposes (sixth chapter). This law requires or recommends the creation of an ethics committee (Art. 26) and the application of the three Rs. The law established economic sanctions that may include disqualification of the offender for a maximum of 5 years if a federal employee is involved. This is a national law and not only refers to the use of animals with experimental purposes, but also covers all animal species that are directly or indirectly related to men and those that are part of the local fauna. Costa Rica Since December 1994, Costa Rica has had an animal welfare law. In 1992, this country presented a bill intended to regulate the animal rights proclaimed in the Universal Animal Rights Declaration that is now filed. In 1993, the Gazette 242, December 20 1993, published a bill, Animal Welfare and Ethology Law that has practically no relation to laboratory animal use. The animal welfare law of 1992 includes some chapters regarding experimental animals and establishes some considerations related to animal experimentation. It has some general ethical principles related to the Three Rs of Russell and Burch (1959). As in Colombia, this is a general law, though it has an attached decree issued by the Science and Technology Ministry, which includes specific points that address the production, care, and use of experimental animals; these are considered mandatory standards in this country. Argentina On December 20, 1996, the National Administration of Drugs, Food and Medical Technology approved the Animal Facility Regulation for laboratories that manufacture medical products and/ or conduct tests for third parties (ANMAT disposition no. 6344; published in the official bulletin 20-01-97). This regulation is a requirement for companies that manufacture medical products, carry out control tests, and/or work as third-party contractors; it includes standards related to facilities, environmental conditions, animal welfare, hygiene, reinvestments, and waste management in these activities. However, even if it is considered mandatory, the regulation does not include other establishments that produce and maintain animals for experimental purposes. This regulation does not mention the type of sanction that would be applied to an institution in a noncompliance situation. In a similar way, in 2002 SENASA issued the 617/02 resolution, Requirements, Conditions and Procedures for the Technical Equipment of Laboratories with Production and Maintenance
84550.indb 54
11/11/10 6:24:06 PM
An Overview of Global Legislation, Regulation, and Policies
55
Animal Facilities, and Experimental Areas, and in 2003 the 76/03 resolution, Creation, within the Management of Laboratories and Technical Service Control of the National Health and Food Processing Areas, of the Permanent Assessment Council in Animal Facility Management. Argentina also has a bill formulated by its local laboratory animal association (AACyTAL), which was published in its 15/16 bulletin and is now waiting to be revised and approved by the local authorities. The aim of this bill is to guarantee the protection of animals used in experimental procedures and other scientific purposes, taking into consideration the three Rs and, as in the Mexican federal regulation, the control and maintenance of records of the facilities dedicated to these types of activities, regardless of their purpose (animal production; research, testing, or teaching; and/ or the combination of any of these). Recently, this bill was sent to the senate and parliament to be analyzed and approved. There is still no resolution regarding this request. Uruguay Before September 15, 2009, the production, care, and use of animals for experimental purposes were regulated to some extent by an Animal Experimentation Honorary Commission that had, since 1996, organized and given training courses in laboratory animal science to technicians, researchers, teachers, and students. After September 15, 2009, Uruguay implemented specific federal regulation for the production, care, and use of laboratory animals (Use of Animals in Experimental, Teaching and Scientific Investigation Activities) that, as with the federal Brazilian law, refers only to species from phylum Chordata, subphylum Vertebrata. This law required the creation of a National Experimentation Commission (Comisión Nacional de Experimentación, CNEA), and ethical committees for the use of animals for experimental purposes, establishing administrative and economic penalties for those institutions that fail to comply with it. Uruguay also has a general decree (Decree 82000-200) that establishes responsibilities toward animal welfare. This decree has important similarities to the anticruelty laws of other countries in this region. The only public university of the country, the University of the Republic, has an ordinance regarding the use of animals for experimental and/or teaching purposes. Mexico To standardize and regulate the production, care, and use of laboratory animals within its territory, in 1999 Mexico published a federal regulation, Norma Official Mexicana NOM-062-ZOO1999, Technical Specifications for the Production, Care and Use of Laboratory Animals, which is current, but undergoing a second revision that will include some animal species not considered when the law was created, as well as some ethical principles related to the care and use of these animals. This law is based mainly on the Guide for the Care and Use of Laboratory Animals (National Research Council 1996). The current version applies to animal facilities and/or other establishments that produce and/or maintain rodents (rats, mice, guinea pigs, hamsters, and gerbils), rabbits, carnivores (dogs and cats), nonhuman primates, and swine. Besides addressing issues related to the production, care, and use of the previously mentioned species, installation design, and environmental considerations, it states that each facility must have a veterinarian, an IACUC, and an occupational health program. Moreover, it requires each facility that produces and/or maintains animals for experimental purposes to register with the corresponding authorities and to send an annual activity report that addresses specific information requested. Oversight of this law is the responsibility of the Agriculture, Livestock and Rural Development, Fisheries and Food Secretariat and state governments. Compliance is the responsibility of the General Directorate of Animal Health and the delegations of the secretary of Agriculture, Livestock and Rural Development, Fisheries and Food Secretariat. The law does not mention the type of
84550.indb 55
11/11/10 6:24:06 PM
56
Handbook of Laboratory Animal Science
sanction that would be applied to facilities that fail to comply with its content, leaving these aspects to the application of other laws, which are also mentioned in the document as references. Besides this federal law, Mexico has local and/or institutional laws and regulations such as the Animal Protection Law, which includes general standards regarding laboratory animal care and use. In a similar way, in Chapter 3, Title 7, of the General Heath Law, Articles 121–126 refer to general aspects regarding the use of animals for research and experimentation purposes related to human health, stating clearly that all experimental protocols should be designed to minimize or prevent animal suffering, that animals must be euthanized using appropriate means, that all animal facilities should have facilities and spaces according to the requirements of each of the animal species housed, and that the production areas must be under the management of qualified personnel. Under this law, the manager or director of the institution is responsible for the security measures pertaining to the care and use of laboratory animals and for establishing an effective occupational health program. Cuba The National Center for Production of Laboratory Animals (CENPALAB) published the Practical Code for the Use of Laboratory Animals in 1992. In this same year, the Academy of Sciences of Cuba created and approved the Cuban Professional Workers Ethical Code, which is mandatory for all researchers in the country. This code considers legal sanctions defined and determined by different Cuban organizations. Resolution 110, which establishes the creation of ethics committees in institutions of the Cuban national health system, was developed in 1997 and approved in 2000 by the Vadi Resolution No. 4/00. Chapters V and VII of this resolution refer specifically to the use of animals in research. In 2001, a group of Cuban scientists presented a bill of animal welfare, which was later incorporated into the animal welfare law as a result of an agreement taken under the National Veterinary Sciences Plenum. In 2004, in Regulation 39/04 (BPS 2004), the creation of the IACUC was approved and, during April 2007, the Guide for Determination of Humane Final End Point in Animals Used in Biomedical Research was submitted for approval to the plenum in the International Veterinary Sciences Meeting at La Habana. Venezuela On June 21, 1999, the Sciences and Technology Ministry of Venezuela published the Code of Bioethics and Biosecurity, which established guidelines for the use of live animals in research based on bioethical principles such as responsibility, justice, autonomy, and caution. The first and second chapters of this code mention the requirements that must be fulfilled for the use of animals for experimental purposes. These include the application of the three Rs and issues related to animal welfare, personnel training, reduction of pain, veterinary supervision, and humane euthanasia; it refers to ICLAS international guidelines for aspects regarding animal facility design. Chile A bill intended to regulate the use of animals for experimental purposes has been introduced in Chile. This document has been in the congress for several years, waiting to be analyzed and approved. The only law in the country regarding animal treatment is quite general and includes only welfare issues. Nevertheless, most universities and research institutes have ethics committees that regulate the use of animals for experimental purposes, ensuring compliance with international specifications. The National Council of Research, Science and Technology, which is responsible for funding research projects in the country, requires that all projects that use animals be approved by the ethics committee of the institution and then reviewed by its own bioethics commission. Moreover, in some universities, research protocols must be evaluated by a bioethics and biosecurity commission. Due
84550.indb 56
11/11/10 6:24:06 PM
An Overview of Global Legislation, Regulation, and Policies
57
to the lack of specific legislation, Chile bases its requirements on the FELASA recommendations, the CCAC guidelines, and the Guide for the Care and Use of Laboratory Animals. Brazil PL 1.153, 1995, was approved by the congress in October 2008 as no. 11794 and enforced by Decree No. 6899 in July 2009 (i.e., Law 11794/2008, Decree 6899/2009). This bill considers species of the phylum Chordata, subphylum Vertebrata, except man. The decree obliges the Ministry of Science and Technology to create CONCEA, the National Council on the Control of Animal Experiments. CONCEA will develop the guidelines and potentially act as an appellate body if the local institutional committee cannot solve a particular question or problem. Each institution conducting animal research must register with CONCEA and must have an ethics committee (Comisiones de Ética en el Uso des Animales, Ceua). Peru and Guatemala These two countries have no legislation. The institutes and universities of Peru base their internal guidelines on the Guide for the Care and Use of Laboratory Animals. Guatemala has only the internal guideline of the Central Animal Facility of the University of San Carlos. Ecuador Although Ecuador has no specific laboratory animal legislation, there is an animal protection foundation with legal attributes that oversees issues related to laboratory animal use. Panama Panama has a bill, PL 20, which, under its fifth chapter, mentions some general aspects related to the use of animals in research. Paraguay In Paraguay, there is only a general anticruelty guideline that refers to farm animals and regulates aspects related to compliance with international requirements for exportation of animals for food purposes. Puerto Rico An animal protection law was created in Puerto Rico in 2004. Chapter V of this law refers to general aspects of animal welfare related to animals used for experimental purposes. Nicaragua Nicaragua has a special law for animal protection and a bill, 121/000123, which establishes guidelines for the production, transport, experimentation, and euthanasia of all animal species. Asia The importance of laboratory animal science is becoming more recognized in Asian countries. Several national laboratory animal science associations have been established in the region. One of the oldest national associations in Asia is the Japanese Association of Laboratory Animal Science
84550.indb 57
11/11/10 6:24:07 PM
58
Handbook of Laboratory Animal Science
(JALAS), founded in 1951. Biomedical scientists have worked to establish a national legislation system in their respective countries to reflect the public’s concern for research-animal welfare. One progressive legislative system in Asia is the amendment of Law for the Humane Treatment and Management of Animals (Law No. 105, 1973) in Japan. In other Asian countries, as in Japan, laws, regulations, and guidelines pertaining to laboratory animals follow with scientists’ advice. The most recent improvement of the legislative system pertaining to laboratory animals is based on the advancement of biomedical science, the internationalization of Asian countries, and public interest in animal welfare in Asia. In 2003, the Asian Federation of Laboratory Animal Science Associations (AFLAS) was established. The exchange of information this forum presents, including legislation in laboratory animal science, increased among the member countries of AFLAS. The initial establishment of a legislative system for laboratory animals was influenced by Western countries, but more recent amendments of the legislation have been influenced by neighboring Asian countries as well as Western countries. In a recent revision of laws, regulations, and guidelines in Asia, the three Rs have been emphasized (e.g., in the Japanese Law for the Humane Treatment and Management of Animals, revised in 2005, and the Korean Animal Protection Law, revised in 2007) (Korea Animal Protection Law 2007). The standard of laboratory animal welfare in Asia is influenced by AAALAC International. Revisions of laws, regulations, or guidelines in some Asian countries, such as the Korean Animal Protection Law (enforced in 2008) and guidelines noticed in 2006 by various ministries in Japan, encourage assessment by an independent body. Japan The Law for the Protection and Management of Animals was amended in 1973. The most recent revision was made in 2005 and the name was changed to the Law for the Humane Treatment and Management of Animals (Japan 2009). Before the revision, “refinement” was the only “R” of the three Rs that was directly included. The principles of “reduction” and “replacement” were subsequently added to the law. More specifically for laboratory animals, the document, “Standards Relating to the Care and Management of Experimental Animals” (Notice No. 6 of the Prime Minister’s Office 1980), was revised in 2006 by the Ministry of Environment. At this time, the name changed to “Standards Relating to the Care and Management of Laboratory Animals and Relief of Pain” (Notice No. 88 of the Ministry of Environment, April 28, 2006, Science Council of Japan 2009). The Ministry of Education, Culture, Sports, Science and Technology and the Ministry of Health, Labor and Welfare developed “Fundamental Guidelines for Proper Conduct of Animal Experiments and Related Activities in Academic Research Institutions under the Jurisdiction of the Ministry of Education, Culture, Sports, Science and Technology” and “Basic Policies for the Conduct of Animal Experimentation in the Ministry of Health, Labor and Welfare.” The ministry of Agriculture and Fisheries published the “Guideline for the Proper Conduct of Animal Experimentation.” These guidelines require the establishment of institutional regulation for animal experiments. The president of a research institution is the responsible person for the proper conduct of animal experimentation. The president is required to designate institutional laboratory animal care committee members to review animal use protocols and to oversee management of the laboratory animal care program. The guidelines also require disclosure pertaining to information about animal experiments. The Science Council of Japan prepared the “Guidelines for Proper Conduct of Animal Experiments” to serve as a reference material or a model when research institutions compile their own regulations for animal experimentation in accordance with the preceding fundamental guidelines and basic policies in 2006. The guidelines resemble the ILAR Guide (National Research Council 1996).
84550.indb 58
11/11/10 6:24:07 PM
An Overview of Global Legislation, Regulation, and Policies
59
It was thought that laboratory animal housing and holding should be regulated by law in Japan. However, there was strong opinion that animal experiments conducted as a component of academic activities should not be over-regulated by law. In this way, Japan favors the establishment of a system based on Japanese customs. The “Guidelines on Methods of Sacrificing Animals” (Notice No. 40 of the Prime Minister’s Office, July 4, 1995) were revised in 2007. The methods of euthanasia of laboratory animals should follow the recommendations in this guideline, with emphasis on minimizing pain and distress during the procedure. There are many other laws, regulations, and guidelines related to animal experimentation in Japan. For example, the guidelines require safety management for animal experimentation. Genetic engineering experiments; animal experiments using radioactive materials or radiation; experiments using poisons, deleterious substances, or psychotropic drugs; and animal experiments using pathogenic agents or hazardous chemicals must be conducted in strict compliance with related laws and ordinances. Animal carcasses and laboratory waste must be disposed of appropriately, using the methods stipulated in in-house regulations. Laws and ordinances related to waste material regulated by law must be followed. In particular, violations of the Law Concerning the Conservation and Sustainable Use of Biological Diversity through Regulations on the Use of Living Modified Organisms are reported. The usage of most transgenic and knockout mice must comply with this law. Recently, bioterrorism has gained public interest. In response, the Law Concerning the Prevention of Infectious Diseases and Medical Care for Patients with Infectious Diseases has been revised several times to restrict the importation of animals and the usage of infectious agents. To prevent the potential spread of infectious diseases in humans when importing living mammals and birds and the carcasses of rodents and lagomorphs, the import of animals is controlled by this law, the Enforcement Regulations of the Law Concerning the Prevention of Infectious Diseases and Medical Care for Patients of Infectious Diseases, and other regulations. The system of notification of import also applies to rodents to be used as laboratory animals. The importation and exportation of laboratory animals including gene-modified rodents to and from Japan must comply with this law. The importation of primates is restricted by the Invasive Alien Species Law and Law Concerning the Prevention of Infectious Diseases and Medical Care for Patients with Infectious Diseases. The importation of farm animals is regulated by the Domestic Animal Infectious Disease Control Law. The importation of dogs and cats and some other species of animals, including laboratory animals, is regulated by the Rabies Prevention Law. A stringent quarantine program, vaccinations, and individual identification with microchips are applied to these species of animals. Korea The Korean Animal Protection Law was amended in 1991. This law was not stringently implemented in Korea. However, as a reflection of public concern about animal welfare, the Korean Ministry for Food, Agriculture, Forestry and Fisheries amended the animal protection law in 2007 and the law was enforced in 2008. The law consists of 26 articles, and articles 13 and 14 are directly related to animal experiments. Article 13, “Experiments with Animals,” states the general requirements for the conduct of animal experimentation, including the consideration of the Three Rs alternatives. As a refinement, animal experiments should be conducted with the least pain and distress, using sound veterinary practices, including analgesics, sedatives, and anesthetics. The law requires euthanasia with the least pain and distress after animal experimentation. Article 14, “Establishment of Animal Experimentation Ethics Committee,” states that the president of the research institution must establish the committee. The committee should comprise one chairman and 3–15 members; at least
84550.indb 59
11/11/10 6:24:07 PM
60
Handbook of Laboratory Animal Science
one-third of the members must be nonaffiliated members. The four specialists in various areas are defined by the law, such as a registered veterinarian, animal welfare specialist, lawyer, and professor at higher education institutions. The composition of the Korean Animal Protection Law is very similar to the Japanese Law for the Humane Treatment and Management of Animals, but the Korean law is more stringent for animal experiments. The establishment of the ethics committee is suggested by various guidelines in Japan, but not Japanese law. However, the Korean law clearly states the establishment of an ethics committee and the members of the committee are defined by the law, including a veterinarian and nonaffiliated members. The Korean law is similar to the U.S. Animal Welfare Act, but does not exclude mice, rats, and birds as laboratory animals. The Korean Animal Protection Law may be one of the most stringent laws in terms of animal experimentation. However, another law, the Laboratory Animal Law by the Korean Food and Drug Administration, has been newly enforced. Thus, there are now two laws to regulate laboratory animals in Korea. People’s Republic of China The Regulation for Administration of Laboratory Animals was amended in 1988. Under this regulation, the National Standards for Laboratory Animals were published in 1997. After the publication of these standards, more practical laws related to laboratory animals were amended—namely, Administration of Laboratory Animal Facility Law in 1998 and Laboratory Animal Permission Law in 2001. For an overview of the framework of regulations pertaining to laboratory animal use in China, see Kong and Qin (2010). In 2001, following the latest progress of international laboratory animal science, the State Technology Supervision Administration issued a new edition of national standards for laboratory animals. In the revised edition, minimum living space of animals was added to the requirements of environment and housing facilities. The national standard was intended to meet the requirements of animal ecology and welfare. The guideline for humane treatment of laboratory animals was issued and implemented in 2006 by the Ministry of Science and Technology. The guideline was the first state-policy-related document that specially directed administrators and technicians on how to pay attention to the welfare of laboratory animals. This guideline clearly stipulates the task and responsibility of the administration committee of producing and applying institutions. It also actively initiates the Three Rs principles—reduction, replacement, and refinement. According to the guideline, the institution intending to use laboratory animals has to obtain the administration’s permission. The national level of laboratory animal issues is the charge of the Ministry of Science and Technology, but the Provincial Bureau of Science and Technology is in charge of this issue locally. The provincial bureau establishes the animal research committee, which approves laboratory animal facilities and laboratory animal specialists and technicians. The larger cities in China, such as Beijing and Shanghai, have established their own municipal offices for laboratory animal science, including the Beijing Administration Office of Laboratory Animals and the Office of Shanghai Administrative Committee for Laboratory Animals and Regulations. In these cities, biomedical science and industries are advancing very rapidly, and the quality of laboratory animals and laboratory animal welfare are regulated as successfully as in most Western countries. Taiwan (Chinese Taipei) The Animal Protection Law was promulgated in 1998 (Taiwan Animal Protection Law 1998). Chapter III of this law specifically addresses scientific application of animal experiments from Article 15 to Article 18. Chapter II, Article 12, of the law precludes the killing of animals. It says
84550.indb 60
11/11/10 6:24:07 PM
An Overview of Global Legislation, Regulation, and Policies
61
that an animal shall not be allowed to be killed at will with the exception of several instances, including “for the purpose of scientific [experimentation].” Euthanasia of laboratory animals is specifically addressed in Article 17: “After a scientific experiment, the conditions of the experimental animals shall be examined immediately. If parts of their limbs or organs have been lost, or they continue to suffer the pain that affects their living quality, they shall be put to death in a least painful way.” The reuse of laboratory animals is prohibited. The replacement of animal experimentation is not clearly stated but reduction and refinement are addressed in Article 15. A management system at research institutions to supervise the scientific application of animal experiments is also recommended. The law requires that the competent authority regulate the source, application, and management of experimental animals and set up an ethics committee of animal experimenters to supervise and manage the scientific experiments. The ethics committee is required to have at least a veterinarian and a representative of an animal protection group from the private sector. The establishment of ethics committees is required by the Regulations for Establishing the Experimental Animal Ethic Committee of the Council of Agriculture. Article 2 of the regulation defines the function of the experimental animal ethics committee, including supervision and management of the scientific application of animal experiments, formulation of the rules, ways and measures for animal protection, and overseeing the management team. The committee is required to meet every 3 months. The establishment of the management group is more specifically described in the Regulation for Establishing the Management Group of Animal Experiments. In Article 2 of the regulation, the composition of the management group is defined in detail, including a doctor of veterinary medicine and certified trained specialists. Article 3 of the regulation defines five missions for the management group including review of the scientific application of animal experiments, suggestions for the improvement of animal experiments, suggestions for laboratory animal facilities, supervision of animal procurement, and supervision of annual reports of animal experimentation. The protection of animals is controlled by the Enforcement Rules of Animal Protection. The rules define the institution that performs the scientific application of animal experiments, including: • • • • • • •
Schools above the level of college Animal drug manufacturer Medicine manufacturer Biological drug manufacturer Hospital Research institution Other scientific applications of animal experiments designated by the central competent authority; however, the conduct of animal experiments in educational institutions below the level of junior high school is restricted in Article 17 of the law
The laboratory animal legislation in Chinese Taipei is well defined in various laws, regulations, and rules. These documents are translated into English. In terms of disclosure of laboratory animal welfare legislation, Taiwan is one of the most progressive among the Asian countries. Philippines Republic Act No. 8485, the Animal Welfare Act, was amended in 1998. Section 6 says that the killing of any animal is unlawful, with exceptions that include authorized research and experiments. The method of killing animals is defined as a humane procedure, which means the use of the most scientific methods available may be determined and proved by the committee. The committee on animal welfare attached to the Department of Agriculture is described in Section 5 of the Act.
84550.indb 61
11/11/10 6:24:07 PM
62
Handbook of Laboratory Animal Science
However, there are not any specific statements for laboratory animals in the Animal Welfare Act. The Philippine Association for Laboratory Animal Science (PALAS) was established in 1986. PALAS acted as a main scientific body to establish the Animal Welfare Act in 1998 and its corresponding rules in terms of laboratory animals. The PALAS Code of Practice for the Care and Use of Laboratory Animals covers the practical aspects of laboratory animal welfare issues in the Philippines. Thailand The Ethical Principles and Guidelines for the Use of Animals by the National Research Council of Thailand were published in 1999. This guideline consists of two parts: 1) ethical principles and 2) guidelines for the use of animals and monitoring of the ethical guidelines for the use of animals. The former consists of five chapters: • Animal users are to be aware of the value of the life of animals. • Animal users are to be aware of the accuracy of the research outcome using the minimal number of animals. • The use of wild animals must not violate laws or policies for wildlife conservation. • Animal users need to be aware that animals are living beings just as humans are living beings. • Animal users must keep detailed data and records of animal experiments.
These chapters are followed by practical guidelines. Monitoring of the ethical guidelines for the use of animals consists of two chapters, including an institutional level and a national level. The establishment of at least one committee is advised to manage and be accountable for the use of animals. The committee members should be diversified and comprise upper level administrative members of the institution, researchers, and lay people. The chapter defines the responsibilities of the institutional committee. At the national level, the National Research Council of Thailand is required to appoint a committee to be responsible for and to promote the ethical use of animals in research; the chapter includes language that defines the committee’s authority. Acknowledgments The authors wish to thank the following individuals for providing information regarding the laws, regulations, and guidelines pertaining to laboratory animal care and use in their respective countries: Dr. Silvia Ortiz Barreto, Brazil Dr. B. Taylor Bennett, United States Dr. Silvia Herrera Bernuy, Peru Dr. Cecilia Carbone, AACyTAL, Argentina Dr. Melvin Dennis, United States Dr. Gilly Griffin, CCAC, Canada Dr. Joel Majerowicz, Brazil Dr. Lázara Martínez, SCCAL, Cuba Dr. Gabriela Méndez, Chile Dr. Manuel Moya, AVECAL, Venezuela Dr. Ekaterina Rivera, Brazil Dr. Adela Rosenkranz, Argentina Dr. Juan Gonzalo Restrepo Salazar, Colombia Dr. Liliana Pazos Sanou, ACCMAL, Costa Rica Dr. Amarillis Saravia, Guatemala Dr. Alba Savaterry, Uruguay
84550.indb 62
11/11/10 6:24:07 PM
An Overview of Global Legislation, Regulation, and Policies
63
References American College of Laboratory Animal Medicine. 1996. Report of the American College of Laboratory Animal Medicine on adequate veterinary care in research, testing and teaching. Cary, NC. AVMA (American Veterinary Medical Association). 2007. AVMA guidelines on euthanasia. Schaumburg, IL. http://www.avma.org/issues/animal_welfare/euthanasia.pdf (accessed April 27, 2008). Anderson, L. 2002. Laws, regulations, and policies affecting the use of laboratory animals. In Laboratory animal medicine, ed. J. Fox, L. Anderson, F. Loew, and F. Quimby, 19–33. New York: Academic Press. Animal Plant Health Inspection Service. 1992. Form 7002. Program of veterinary care for research facilities or exhibitors/dealers. ———. 1999. Ensuring adequate veterinary care: Roles and responsibilities of facility owners and attending veterinarians. Tech note, March 1999. Canadian Council on Animal Care. 1984. Guide to the care and use of experimental animals, vol. 2. Ottawa, Canada. http://www.ccac.ca/en/CCAC_Programs/Guidelines_Policies/GUIDES/ENGLISH/toc_v1.htm (accessed April 27, 2008). ———. 2000. CCAC policy on compliance and noncompliance. http://www.ccac.ca/en/CCAC_Programs/ Guidelines_Policies/POLICIES/COMPLI.htm (accessed April 27, 2008). Centers for Disease Control and Prevention/National Institutes of Health. 2007. Biosafety in microbiological and biomedical laboratories, 5th ed. Washington, D.C.: U.S. Government Printing Office. Code of Federal Regulations. 1997. Title 40: Protection of the environment; Chapter 1: Environmental Protection Agency; Subchapter E: Pesticide programs; Part 160: Good laboratory practice standard. Washington, D.C.: Office of the Federal Register. ———. 1998. Title 21: Food and drugs; Chapter 1: Food and Drug Administration, Department of Health and Human Services; subchapter A: General; Part 58: Good laboratory practice for nonclinical laboratory studies. Washington, D.C.: Office of the Federal Register. Council of Europe. 1986. Convention for the protection of vertebrate animals used for experimental and other scientific purposes (ETS 123). Strasbourg: Council of Europe. ———. 2006. Appendix A of the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (ETS no. 123); Guidelines for Accommodation and Care of Animals (Article 5 of the Convention). Approved by the Multilateral Consultation. Department of Defense. 1995. The use of laboratory animals in DoD programs, Directive number 3216.1. http://www.army.mil/usapa/epubs/pdf/r40_33.pdf (accessed April 27, 2008). European Commission. 1986. Directive for the protection of vertebrate animals used for experimental and other scientific purposes (86/609/EEC). Official Journal of the European Commission L 358:1–29. European Council. 1986a. Council Directive 86/609/EEC of 24 November 1986 on the approximation of laws, regulations and administrative provisions of the member states regarding the protection of animals used for experimental and other scientific purposes. Official Journal of the European Commission L 358: 1–29. Brussels. ———. 1986b. Resolution of the representatives of the governments of the member states of the European Communities, meeting within the council of 24 November 1986 regarding the protection of animals used for experimental and other scientific purposes. Official Journal of the European Commission L 358:18.12.1986, p. 1. ———. 2003. Directive 2003/65/EC of the European Parliament and of the Council of 22 July 2003 amending Council Directive 86/609/EEC on the approximation of laws, regulations and administrative provisions of the member states regarding the protection of animals used for experimental and other scientific purposes. http://eur-lex.europa.eu/smartapi/cgi/sga_doc?smartapi!celexplus!prod!DocNumber&lg=en&ty pe_doc=Directive&an_doc=2003&nu_doc=65 (accessed August 14, 2009). Interagency Research Animal Committee. 1985. U.S. government principles for the utilization and care of vertebrate animals used in testing, research, and training. Japan Law for the Humane Treatment and Management of Animals. http://www.cas.go.jp/jp/seisaku/hourei/ data/AWMA.pdf (accessed December 29, 2009). Kong, Q., and Q. Qin. 2010. Analysis of current laboratory animal science policies and administration in China. ILAR e-Journal 51:E1-E10. http://dels.nas.edu/ilar_n/ilarjournal/51_1/PDFs/v51(e1)Kong.pdf (accessed March 5, 2010).
84550.indb 63
11/11/10 6:24:08 PM
64
Handbook of Laboratory Animal Science
Korea Animal Protection Law. 2007. http://www.koreananimals.org/animals/apl/2007apl.htm (accessed December 29, 2009). National Association for Biomedical Research. 1999. Animal Legal Defense Fund (ALDF) et al. v. Glickman et al. and NABR. U.S. District Court for the District of Columbia, civil action no. 96-408 (CRR), U.S. Court of Appeals No. 97-5009 consolidated with 97-5031 and 97-5074. Summary as of June 8, 1999. Washington, D.C. National Institutes of Health. 2009. NIH guidelines for research involving recombinant DNA molecules. Bethesda, MD. http://oba.od.nih.gov/oba/rac/guidelines_02/NIH_Gdlnes_lnk_2002z.pdf (accessed December 29, 2009). National Research Council. 1996. Guide for the care and use of laboratory animals. Washington, D.C.: National Academy Press. Occupational Safety and Health Administration, Department of Labor. 2001. Occupational exposure to bloodborne pathogens; needlestick and other sharps. 29 Code of Federal Regulations 66:5317–5325. OLAW (Office of Laboratory Animal Welfare), National Institutes of Health. 2002. Public Health Service policy on humane care and use of laboratory animals. Bethesda, MD. Philippines Animal Welfare Act of 1998. http://www.angelfire.com/ok2/animalwelfare/welfareact.html (accessed December 29, 2009). Rozmiarek, H. 2007. Origins of the IACUC. In The IACUC handbook, 2nd ed., ed. J. Silverman, M. Suckow, and S. Murthy, 1–9. Boca Raton, FL: CRC Press LLC. Russell, W., and R. Burch. 1959. The principles of humane experimental technique, 2nd ed. London: Methuen. Science Council of Japan. 2009. Guidelines for Proper Conduct of Animal Experiments June 1, 2006, by Science Council of Japan http://www.scj.go.jp/ja/info/kohyo/pdf/kohyo-20-k16-2e.pdf (accessed December 29, 2009). Taiwan (Chinese—Taipei) Animal Protection Law. 1998. http://www.animallaw.info/nonus/statutes/sttaapl1998. htm (accessed December 29, 2009). U.S. Congress Office of Technology Assessment. 1986. State regulation of animal use. In Alternatives to animal use in research, testing, and education, 305–331. Washington, D.C.: U.S. Government Printing Office. USDA (U.S. Department of Agriculture). 1991. Code of federal regulations, Title 9, Part 3, animal welfare; standards; final rule. Federal Register 56 (32): 1–109.
84550.indb 64
11/11/10 6:24:08 PM
Chapter 4
Assessment of Animal Care and Use Programs and Facilities
Javier Guillen, Letty V. Medina, and James R. Swearengen Contents Introduction....................................................................................................................................... 65 Background....................................................................................................................................... 67 History of ISO.............................................................................................................................. 67 History of AAALAC International.............................................................................................. 67 History of GLP Regulations......................................................................................................... 68 System Descriptions.......................................................................................................................... 69 ISO............................................................................................................................................... 69 AAALAC International............................................................................................................... 71 GLP Regulations.......................................................................................................................... 73 Discussion......................................................................................................................................... 75 Summary........................................................................................................................................... 76 References......................................................................................................................................... 77 Bibliography...................................................................................................................................... 79 Introduction “Good animal care and good science go hand in hand.” This caption on a poster developed by the United States National Institutes of Health (NIH) Office of Animal Care and Use was intended to remind NIH employees and visitors that sound science relies on high-quality animal care and use practices. Many years before this poster was produced, Drs. Russell and Burch sent much the same message to the scientific community when they wrote in 1959, Alternatives to animal tests are a key part of humane science, which can be captured in the concepts of refinement, reduction, and replacement, referred to as the 3Rs. The 3Rs are based equally on ethical consideration of animals in the laboratory setting and the recognition that when the researcher in experimental design and implementation appropriately applies these principles, they result in a situation that is likely to produce more robust scientific results. (Russell and Burch 1959)
This recognition that consistent, high-quality animal care and use programs are vital in producing research results of corresponding high quality has led to the development of several programs 65
84550.indb 65
11/11/10 6:24:08 PM
66
Handbook of Laboratory Animal Science
that we will refer to in this chapter as “assessment systems.” The three such systems to be presented here are accreditation by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC International), good laboratory practice (GLP) regulations, and certification/registration by the International Organization for Standardization (ISO). In addition to describing these three systems, this chapter will provide information on the historical bases for their development and the ways in which the differences in origins influence their procedures and ultimate aims. We will also attempt to show that, despite these differences, all three can be integrated at the institutional level, with each providing distinct benefits. Why is consistent, high-quality animal care important as it relates to biomedical research, teaching, and testing? There are at least three different motivators that underlie the need for quality: legal necessity, an ethical imperative, and the validity of scientific results. For most scientists and institutional staff who support the activities of research involving animals, the baseline for an animal care and use program is established in laws or other governmental requirements. Minimum requirements for animal welfare in science are delineated in the United States by the Animal Welfare Regulations (Code of Federal Regulations 1985) and the Public Health Service (PHS) policy (Public Health Service Policy on Human Care and Use of Laboratory Animals 1996), and in European countries by Directive 86/609/EEC and the Council of Europe ETS 123 with its recently revised Appendix A (Council Directive on the Approximation of Laws, Regulations, and Administrative Provisions of the Member States Regarding the Protection of Animals Used for Experimental and Other Scientific Purposes, Directive 86/609/EEC 1986 and Appendix A of the European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes [ETS No 123] 2006). Other countries have animal welfare laws of varying stringency, ranging from very general anticruelty statutes to the prohibition of using some animal species, like great apes, in research. Additional legal requirements are imposed if the experimental data generated through animal research (as well as in vitro methods) are to be submitted to a governmental authority, such as the U.S. Food and Drug Administration (FDA) or the European Medicines Agency (EMEA), for premarketing approval. However, meeting the minimum legal requirements for animal research is increasingly being viewed by both the public and the scientific community as merely a foundation on which to build programs that optimize animal welfare. To those who hold this view, using animals in research brings with it an ethical imperative to go beyond the strict letter of the law to strive for the highest standards of animal care and use. Indeed, scientists themselves recognize that using animals to further scientific knowledge is a privilege that they are ethically obliged as individuals not only to acknowledge, but also to factor constantly into their research decisions. Validation that institutions and individuals are striving to achieve high standards in this specific area is one function of external assessment groups such as AAALAC International. Scientific inquiry is guided by the scientific method, one principle of which is that a hypothesis must withstand the scrutiny of other scientists. This scrutiny involves not only the validity of the hypothesis itself, but the methods originally used to test the hypothesis. When scientific methods include animal experiments, a number of variables are introduced that can significantly affect the reproducibility of data and hence the ability to validate a study. These variables include the genetic “purity” of the animal if rodent inbred strains are used; the animal’s health and immune status, including latent infections; potential animal stressors, including those involved with routine maintenance as well as experimental manipulations; and the animal’s macro- and microenvironments. Minimizing these and other animal-related variables is a goal that is advanced by applying consistent, uniform high standards of procurement, husbandry, veterinary care, and experimental use. Systems such as GLP and ISO provide methods and measures to ensure the requisite consistency and, along with participation in AAALAC International’s accreditation program, demonstrate that high standards in these broader areas are being met and maintained.
84550.indb 66
11/11/10 6:24:08 PM
Assessment of Animal Care and Use Programs and Facilities
67
Background Each of the three systems to be discussed here occupies an individual niche in promoting and assuring quality animal care and use practices. To understand the functions of each system and the nature of their niches, it is useful to examine how each was conceived and developed. History of ISO In Jack Latimer’s book, Friendship among Equals (1997), one of the founders of ISO, Willy Kuert, writes: ISO was born from the union of two organizations. One was the ISA (International Federation of the National Standardizing Associations), established in New York in 1926, and administered from Switzerland. The other was the UNSCC (United Nations Standards Coordinating Committee), established only in 1944, and administered in London. The conference of national standardizing organizations which established ISO took place in London from 14 to 26 October, 1946.
The result of this founding meeting of delegates from 25 countries was the creation of a new international organization “the object of which would be to facilitate the international coordination and unification of industrial standards”—in other words, to ease the movement of goods (or services) across international boundaries. The new organization, ISO, which began to function officially on February 23, 1947, has evolved to become a worldwide federation of national standards bodies from more than 150 countries (International Organization for Standardization 2010a). Hundreds of ISO standards exist that define the requirements for everything from foods to film speed. In fact, many of us will recognize the film speed measurement called ISO 100, ISO 200, and so on. The aim of ISO was to promote the development of standardization and related activities in the world with a view toward facilitating the international exchange of goods and services and to developing cooperation in the spheres of intellectual, scientific, technological, and economic activity. Further, ISO defines its standards as documented agreements containing technical specifications or other precise criteria to be used consistently as rules, guidelines, or definitions of characteristics, in order to ensure that materials, products, processes, and services are fit for their purpose. ISO standards cover a huge area of subjects. The subject of most interest and applicability for our purposes is that of “quality management systems,” which is covered in the ISO 9000 family of standards (International Organization for Standardization 2010e). History of AAALAC International The origin of AAALAC International can also be traced to the World War II period. The postwar boom in science, including the dramatic growth in public funding of research in the United States by the NIH, brought with it a commensurate increase in animal experimentation. Thus, in the late 1940s, five veterinarians involved in managing laboratory animal facilities in major institutions in Chicago, Illinois, first met to discuss and share information about the care of laboratory animals. Before the end of the decade, a movement emerged from this group to form a national organization to address the issues facing the growing field of laboratory animal science (Miller and Clark 1999). The first meeting of AAALAC International’s progenitor, the Animal Care Panel (ACP), took place in 1950. At that meeting, Carl Schlotthauer of the Mayo Foundation emphasized the need to “establish some uniformity in animal handling” (Miller and Clark 1999). Although ACP members initially recognized the need for standards, certification, and accreditation, the organization struggled with how to implement such programs during the period between 1950 and 1960. As
84550.indb 67
11/11/10 6:24:08 PM
68
Handbook of Laboratory Animal Science
early as 1951, the ACP had a Committee on Animal Care Standards and a Committee on Regulations for the Care of the Dog, but it was not until 1958 that the committee finally decided that the ACP should undertake accreditation functions for laboratory animal maintenance and care. Finally, in the fall of 1964, a subcommittee of the ACP issued an in-depth report, Accreditation of Laboratory Animal Facilities, recommending establishment of an autonomous entity (not part of the ACP) to conduct an accreditation program. The American Association for Accreditation of Laboratory Animal Care was established as a nonprofit corporation (public charity) in April 1965. Due to the advent of globalization and the need for an internationally recognized standard of animal care and use, the board of trustees decided in 1996 to change the name to the Association for Assessment and Accreditation of Laboratory Animal Care International and expand its role throughout the world. AAALAC International now accredits over 800 programs in 32 countries. History of GLP Regulations Good laboratory practice regulations differ from ISO International Standards and AAALAC International’s accreditation program in several ways. First, GLP regulations are legal requirements developed by the U.S. Food and Drug Administration (FDA), as opposed to ISO standards and AAALAC International’s accreditation program, which are managed by voluntary, nongovernmental organizations. Individual animal programs may voluntarily adopt ISO standards and undergo accreditation by AAALAC International, but they are mandated to follow the GLP regulations when performing some types of studies. Second, GLP regulations have their roots in governmental actions to prevent harm to the general public, whereas ISO standards and AAALAC International’s accreditation program were developed by groups of individuals who sought to standardize and enhance quality management systems, such as an animal care and use program, to better their professional specialties. The U.S. FDA was created with the enactment of the Food and Drug Act of 1906. Its principal mission, then and now, is to protect consumers from harmful foods and drugs, which prior to that time were frequently misbranded and often impure. The FDA’s powers were expanded in 1938 in response to an incident in 1937 in which an incorrectly labeled elixir, sulfanilamide, containing diethylene glycol killed 100 people in 2 months (Cook 2010). Again, in 1962, a tragedy involving widespread, severe birth defects caused by the European-approved drug thalidomide led to additional FDA powers to require manufacturers to prove effectiveness and provide postapproval reports. It was criminal activities, however, that brought about the GLP regulations. Audits authorized by the 1962 law showed that 618 of 867 studies were invalid because of numerous discrepancies between study procedures and data (Baldeshwiler 2003). As a result, the FDA prosecuted four individuals. These toxicology testing laboratory managers were found guilty and sentenced to prison. The subsequent FDA decision to regulate laboratory testing resulted in GLP regulations that were proposed in 1976, finalized in 1978 (U.S. Food and Drug Administration 1978), and took effect in 1979. The GLP regulations were amended in a final rule published on September 4, 1987, to clarify, delete, or amend several provisions to reduce the regulatory burden on testing facilities (U.S. Food and Drug Administration 1987). During this period, concerns about the safety of food and drugs were not confined to the United States. At the international level, the Organization for Economic Cooperation and Development (OECD) was working at the same time to develop guidance in this area. The OECD comprises 30 countries from Asia, Europe, North America, and the Pacific Rim; its goals are to “build strong economies in its member countries, improve efficiency, hone market systems, expand free trade and contribute to development in industrialized as well as developing countries” (Organization for Economic Cooperation and Development 1998). In addition to the 30 members, OECD has active relationships with more than 100 additional countries and many nongovernmental organizations (NGOs), giving it truly global influence. An integral part of its 1981 “Council Decision on the Mutual Acceptance of Data in the
84550.indb 68
11/11/10 6:24:08 PM
Assessment of Animal Care and Use Programs and Facilities
69
Assessment of Chemicals” was establishment of the OECD “Principles of GLP,” which were revised in 1997 (Organization for Economic Cooperation and Development 1998). To further ensure harmonization of the procedures related to GLP compliance monitoring in preclinical safety studies, the OECD has developed guidance documents that strive to ensure that studies are carried out according to the “Principles of GLP.” These include a 1989 council decision, “Recommendation on Compliance with Good Laboratory Practice,” which requires the establishment of national compliance monitoring programs based on laboratory inspections and study audits and recommends the use of the “Guide for Compliance Monitoring Procedures for Good Laboratory Practice” and “Guidance for the Conduct of Laboratory Inspections and Study Audits.” Both of these guidance documents were revised in 1995 (Organization for Economic Cooperation and Development 1998). System Descriptions ISO In scientific research in general, results should be accurate and reproducible and obtained in a clear and transparent way. This demands consistency and high quality during all phases of an animal experiment, such as those shown in Table€4.1. With its emphases on high-quality product development and customer satisfaction, the ISO process can be a powerful management tool for an organization with an animal care and use program. The ISO standard most commonly applied to organizations with animal care and use programs is ISO 9000, the standard for quality systems. According to the official ISO Web site, “The ISO 9000 family of standards represents an international consensus on good quality management practices” (International Organization for Standardization 2010b). The quality management system standards of the ISO 9000 series are based on eight quality management principles (International Organization for Standardization 2010c): customer focus leadership Table€4.1â•…Phases of an Animal Experiment to Which ISO Is Applicable A.╇ Preliminaries 1.╇ Check financial and scientific solidity 2.╇Design outline of study protocol 3.╇ Check legal and ethical aspects B.╇ Plan, define, and verify requirements 1.╇ “Test substance” 2.╇Animals (e.g., housing, feed, destination at end of study) 3.╇Facilities (e.g., equipment, operating rooms, biohazards) 4.╇Staff (e.g., number, techniques, trained) 5.╇ Paperwork (e.g., study protocol, instruction, registration) C.╇ Perform study 1. ╇Animals: order, receive, verify that specifications are met, house 2.╇ Perform techniques needed (e.g., administration of test substance) 3.╇Obtain data and samples (e.g., collection of blood, feces, urine, body weight, observations) D.╇Analyze samples and data E.╇Write report, paper F.╇Archive
84550.indb 69
11/11/10 6:24:09 PM
70
Handbook of Laboratory Animal Science
involvement of people process approach system approach to management continual improvement factual approach to decision making mutually beneficial supplier relationships
ISO 9000 currently includes three quality standards: ISO 9000, ISO 9001, and ISO 9004. After each ISO number, there is typically a colon followed by the year of the most recent update. ISO standards are updated approximately every 5 years (International Organization for Standardization 2010d): • ISO 9000 presents the fundamentals and vocabulary of quality management systems. • ISO 9001 presents the requirements for a quality management system. • ISO 9004 presents guidelines for performance improvements that go beyond the basic requirements.
All of these are process standards, rather than product standards (International Organization for Standardization 2010d). For organizations with animal care and use programs, ISO 9001 is most applicable because “[it] is a useful basis for organizations to be able to demonstrate that they are managing their business so as to achieve consistent, high quality goods and services” (International Organization for Standardization 2010c). It is also a standard against which requirements of a system can be certified by an external organization. In the animal research context, researchers are the “costumers” of the products, which can range from providing and maintaining appropriate animals to performing a complete study (e.g., contract research organizations). ISO 9001 has five sections: (1) quality management system, (2) management responsibility, (3) resource management, (4) product realization, and (5) measurement, analysis, and improvement (International Organization for Standardization 2010d). The core philosophy of the standard is defined by taking a slightly closer look at each of the five parts. In addition to general requirements, the quality management system section of ISO 9001 requires that a company document describe what the system is and how it works (e.g., quality manual, control system for documents and records). The management responsibility section very clearly and specifically assigns the responsibility for creation, implementation, documentation, and improvement of the quality system squarely to an organization’s senior management (e.g., commitment to quality, listening to customers, planning for quality). This is one of the most significant parts of ISO 9001. In addition, a regular, documented review of the quality program, with an eye toward continuous improvement, is required. While all of senior management is identified as responsible, a single person (e.g., director of animal laboratories, study director) is identified as the management representative, who is appointed by top management to ensure that the quality management system for the animal care and use program is established, implemented, maintained, and improved. Resource management emphasizes that human resources must be competent and available in sufficient numbers to assure quality work. This part also requires that the facilities, equipment, supporting services, and training programs are sufficient to assure product quality (e.g., healthy animals, humane care, appropriate use). An appropriate work environment is also required. Product realization focuses on the quality of the product and the need to listen carefully to the customer to assure that requirements are well understood. How design decisions are made, reviewed, validated, and controlled is emphasized. The purchase of anything used to produce data (e.g., healthy animals or good animal care) is included in this part. Vendor audits must be conducted to help assure that purchased materials meet specifications. The last requirement of this section is for a company to take total responsibility for the quality of its goods and services.
84550.indb 70
11/11/10 6:24:09 PM
Assessment of Animal Care and Use Programs and Facilities
71
The fifth section is termed measurement, analysis, and improvement. Decisions about processes and goods and services must be made from review of adequate data obtained from rigorous measÂ� urements and audits. The data are then analyzed to ensure continual improvement of the quality management system. This part also contains the need for a strong corrective and preventive action program. An example of continuous improvement for a laboratory animal program should help in understanding this. If adult rabbits housed for long-term studies have a high incidence of foot lesions from metal flooring, moisture, and obesity, providing a solid, dry resting surface and veterinary care will correct the current issue. Ordering plastic floored caging with better urine deflection and initiating a limited feeding regimen to prevent obesity will prevent rabbit foot problems in the future. Becoming registered with ISO 9001 is a voluntary process. An organization begins by creating a quality system, a quality manual, and all the processes and systems that assure the quality of its goods and services. One important distinction is that management responsibility in ISO 9001 extends to suppliers. In other words, everything purchased is subject to the standard, which in the case of animal care and use programs would include animals, feed, bedding, caging, etc. An ISO inspection is a very rigorous process, entailing such elements as a visit to the facility’s boiler room to determine if the building maintenance staff has adequate training and processes and is following specific written procedures. AAALAC International As noted in this chapter’s “Background” section, AAALAC International is a nonprofit corporation. Governance of the corporation is through a board of trustees composed of representatives from scientific, professional, and nonprofit organizations involved with or otherwise interested in the humane care and use of laboratory animals, including the American Association for Laboratory Animal Science, Federation of European Laboratory Animal Science Associations, International Council on Laboratory Animal Science, International Association for Gnotobiology, Federation of American Societies for Experimental Biology, etc. A complete list of these organizations can be found on the AAALAC International Web site (Association for Assessment and Accreditation of Laboratory Animal Care International 2010a). Through this structure, the direction and policies followed by AAALAC International are established and driven by the communities to which AAALAC International programs are directed. The founders of AAALAC International recognized that the involvement of end-users of an accreditation program would be beneficial in several ways. First, involving laboratory animal specialists and other scientists in the process would ensure that AAALAC International programs would be based on scientific principles, with the standards and procedures employed more likely to be based on empirical scientific data and professional judgment. Second (and related to the first reason), those who would be subject to the accreditation process would be far more likely to be receptive to standards and procedures they had a role in developing. For practically the entire 45 years of its existence, AAALAC International has relied on the Guide for the Care and Use of Laboratory Animals (National Research Council, Institute for Laboratory Animal Resources 1996) as the principal standard against which animal programs are evaluated. This continues to be true in the United States and in countries that do not have at least equivalent standards. For countries with existing laws, regulations, or other standards dealing with research animal care and use, the Guide serves as an adjunct in areas not covered by national standards. In addition to the Guide, AAALAC International employs a number of resources (referred to as “reference resources”) in specific areas where these documents provide more detailed information than the Guide (e.g., for agricultural animals, occupational health and safety, euthanasia, and numerous other topics) (Association for Assessment and Accreditation of Laboratory Animal Care International 2010b). It is important to note that national requirements always serve as the baseline for AAALAC International accreditation and that organizations must be in full compliance
84550.indb 71
11/11/10 6:24:09 PM
72
Handbook of Laboratory Animal Science
Table€4.2 ╅AAALAC International Program Description Outline A.╇Institutional policies and responsibilities 1.╇Monitoring the care and use of animals 2.╇Veterinary care 3.╇ Personnel qualifications and training 4.╇Occupational health and safety of personnel B.╇Animal environment, housing, and management 1.╇ Physical environment 2.╇Behavioral management 3.╇ Husbandry 4.╇ Population management C.╇Veterinary medical care 1.╇Animal procurement and transportation 2.╇ Preventive medicine 3.╇Surgery 4.╇ Pain, distress, analgesia, and anesthesia 5.╇Euthanasia 6.╇Drug storage and control D.╇ Physical plant
with their own national laws, regulations, and policies before they can hope to achieve AAALAC International accreditation. Similarly to the ISO registration process, the AAALAC International accreditation process begins with a form of internal self-assessment. For accreditation, this involves a description of an organization’s animal care and use program through completion of a standard outline provided by AAALAC International that closely follows the sections of the Guide (Table€4.2). The AAALAC International Council on Accreditation is deciding how to incorporate the changes included in the 2010 update of the Guide (National Research Council, Institute for Laboratory Animal Resources 2010). Upon acceptance of the program description, an on-site evaluation by AAALAC International representatives is scheduled. The site visit team comprises at least one member of AAALAC International’s Council on Accreditation, along with one or more additional individuals from a group of AAALAC International ad hoc consultants and specialists. The Council on Accreditation is composed of experts active in their fields, with knowledge and experience in research animal care, use, and/or oversight, who are elected by their peers. To meet the demand of international growth, the Council on Accreditation was expanded in 2004 and 2009 with two new sections established in Europe and the Pacific Rim, respectively. Both sections comprised members from countries included in the designated geographical areas. The council is now composed of three North American sections, one European section, and one Pacific Rim section. Ad hoc consultants and specialists are also elected by the Council; they are chosen based on their knowledge, experience, and area of expertise so as to allow AAALAC International to tailor site visit teams to meet any special circumstances at the organizations seeking accreditation. The site visit is the first step of the independent, third-party peer review process that AAALAC International provides. The on-site evaluation begins with review of the organization’s program description with appropriate staff and additional verification that actual practices correspond to their descriptions. Review of records, interviews with personnel, and evaluations of the animal facility, support areas, and researcher laboratories are also conducted to assess the degree to which AAALAC International standards are being met. The site visit concludes with an exit briefing at which the site visitors provide a summary of their observations and preliminary findings and the
84550.indb 72
11/11/10 6:24:09 PM
Assessment of Animal Care and Use Programs and Facilities
73
organization is provided with an opportunity to correct misperceptions and ensure that there are no errors in fact in the site visitors’ preliminary findings. Following the site visit, the team prepares a draft report, which is distributed to several other Council members for review. This second level of peer review precedes a meeting of the full Council (which meets three times a year) at which the Council member who led the site visit presents his or her report, along with a recommendation for accreditation status. Reviewers’ comments are discussed and the council votes on final accreditation status, completing the third level of peer review. Organizations are informed of their accreditation status through a letter report that includes commendations, findings, and recommendations. For new applicants, possible status includes “full accreditation,” “conditional full accreditation,” “provisional status,” or “withhold accreditation.” Accredited organizations are visited every 3 years and may receive “continued full accreditation” “conditional full accreditation,” “deferred accreditation,” or “probationary accreditation” accreditation; under extreme circumstances, an organization may receive notice that accreditation is revoked. Any report of status other than full accreditation will include a description of findings that must be corrected before full accreditation can be restored (mandatory items) and any additional suggestions for improvement. Full accreditation reports may also include suggestions for improvement. To maintain accreditation, annual reports are required, and organizations must complete updated program descriptions and be visited every 3 years. GLP Regulations As might be expected from regulatory agencies such as FDA, OECD, and the U.S. Environmental Protection Agency (which has its own set of GLP regulations very similar to the FDA’s), GLP requirements for nonclinical laboratory studies are extensive and very detailed (U.S. Food and Drug Administration 1978). Compliance with GLP requirements is accomplished and verified through both internal and external procedures and inspections. Internally, quality assurance (QA) units are required to perform a variety of specified duties. Externally, an authorized regulatory agency official can inspect a laboratory’s facility, records, operating procedures, oversight processes, and specimens at reasonable times and in a reasonable manner. In the United States, the FDA may request copies of the institutional animal care and use committee (IACUC) meeting minutes and operating procedures to ensure that animal care and use oversight is appropriate for the GLP study animals. The subparts of GLP regulations categorize requirements under general provisions, organization and personnel, facilities, equipment, testing facilities operation, test and control articles, protocol for and conduct of a nonclinical laboratory study, records and reports, and disqualification of testing facilities. Subparts and special issues are described in more detail in the OECD’s “Guidance Documents” (Organization for Economic Cooperation and Development 1998). At organizations that conduct animal studies in support of applications for research or marketing permits from regulatory agencies, GLP compliance activities will be required in most of these areas. A frequently heard maxim that may have its roots in GLP compliance is that “if you didn’t document it, you didn’t do it.” Documentation is, in fact, a hallmark of GLP compliance. There are anecdotal reports of pharmaceutical companies delivering entire truckloads of documents to the FDA when requesting marketing approval for a new drug. However, now that electronic regulatory submissions are allowed, a majority of new drug applications are submitted to the FDA electronically. A key element of the documentation process is the standard operating procedure (SOP), and the GLP regulations require that SOPs be established for, but not limited to, 12 specific nonclinical laboratory study methods (Table€ 4.3). A majority of these required SOPs involve animal-related aspects of studies. GLP standard operating procedures must state exactly the manner in which the procedure is to be performed each and every time. Regulations also require that historical records of SOPs, including all revisions, be maintained.
84550.indb 73
11/11/10 6:24:09 PM
74
Handbook of Laboratory Animal Science
Table€4.3╅Areas Requiring SOPs under GLPs Animal room preparation Animal care Receipt, identification, storage, handling, mixing, and method of sampling of the test and control articles Test system observations Laboratory tests Handling of animals found moribund or dead during study Necropsy of animals or postmortem examination of animals Collection and identification of specimens Histopathology Data handling, storage, and retrieval Maintenance and calibration of equipment Transfer, proper placement, and identification of animals
The GLP regulations include several sections that are specific to animals including animal care facilities, animal supply facilities, and animal care. They outline requirements for GLP research facility design to ensure separation of test systems, adequate space, and specialized rooms for conducting nonclinical laboratory studies. Also outlined are the requirements for animal housing, feeding, handling, and care, with an emphasis on identification of animals and elimination of contaminants or variables that might affect the conduct of the study (U.S. Food and Drug Administration 1978). Many of these GLP requirements are complementary to the Animal Welfare Regulations and are considered standard practices in a well-run animal care and use program (Code of Federal Regulations 1985). They include provisions for quarantine of newly arrived animals, health status determinations, identification, separation of species, equipment cleaning and sanitization, feed and water analysis, and bedding changing. Although these are standard practices for many researchers, strict adherence to SOPs and documentation of all key actions differentiate GLP studies from nonGLP studies. Aside from the areas identified previously, other provisions of GLP regulations are also applicable in an animal care and use program. These include the appropriate training of personnel (although the scope and intensity are not specified); adequate numbers of personnel; use of personal protective equipment; adequate and appropriate storage areas; clean, well-maintained laboratory space; and inspected, well-maintained equipment. Maintenance of records, both of animal health and research data, is also a critical component of GLP studies that should be part of any well-run animal care and use program. However, given the quite basic GLP requirements in this area, one cannot assume a priori that full GLP compliance guarantees a high-quality animal care and use program. The rigid nature and complexity of GLP requirements make it inherently difficult to comply with them, especially for the duration of long-term studies (lasting 1 or 2 years). The regulatory agencies at least tacitly acknowledge this through their provisions in GLPs for internal quality assurance units. A QA unit may be “any person or organization element, except the study director, designated by testing facility management” and is responsible for monitoring each study to assure that facilities, personnel, equipment, methods, practices, records, and controls conform with GLP regulations (U.S. Food and Drug Administration 1978). The regulations further require that the QA unit must be entirely separate from and independent of the personnel directing or conducting the study. QA units maintain schedules and protocols and are required to conduct periodic inspections of studies, with reports provided to study directors and facility management. This ongoing internal oversight mechanism plays a significant role in ensuring that the facility, study sponsors, and regulatory agencies meet the strict requirements of GLP regulations.
84550.indb 74
11/11/10 6:24:09 PM
Assessment of Animal Care and Use Programs and Facilities
75
Discussion Attaining and maintaining high standards should be the goal of all organizations that involve the use of animals in research, testing, and teaching. Mechanisms to help achieve this goal can be internal and external. The internal mechanisms are represented by the specialized professionals working closely with the research animals, such as veterinarians, animal care personnel, and researchers, as well as by the institutional animal care and use committees, ethics committees, and oversight bodies. External input can be obtained from inspections by competent authorities and from the systems described previously. Government inspections are not performed in many places and are highly variable in practical terms. Therefore, the application of a dedicated international program of assessment is the best tool to assist internal mechanisms in establishing and maintaining high standards of animal care and use. The three systems described earlier can all be applied to help achieve these goals, either individually or in combination (Howard et al. 2004). Many pharmaceutical companies, contract research organizations (especially those conducting toxicology testing), and centralized laboratory institutes (Van Velden-Russcher and van Herck 2001; Van Herck 1999) integrate all three systems into their animal care and use programs. The fact that three distinct assessment systems can be employed simultaneously is not surprising when one examines their origins, procedures, and ultimate aims. Although aspects of the systems differ to greater and lesser degrees, they should be viewed as complementary rather than exclusionary (Jansen 1999; Ritskes-Hoitinga, Kalisie-Korhonen, and Smith 1999). Beginning with an examination of its history, it is clear that the GLP system came into existence in response to serious adverse events: illness, deformities, and even death related to the development and sale of products consumed by humans. Thus, GLP regulations are heavily oriented toward avoiding harm to humans. Given the complex nature of drug development, it is logical that this assessment system involves strict adherence to rather inflexible requirements at multiple points in that process. The principal focus of GLP regulations is the quality and traceability of the drug safety and efficacy data produced. Review of the origin of ISO reveals that organizational founders were primarily concerned with the need for the international harmonization of standards. Standard development continues to be its main orientation, and the list of ISO standards is both extensive and diverse. Although the vast majority of ISO standards are in areas unrelated to animals, the 9000 series dealing with quality management can clearly be applied to the operation of animal care and use programs, especially for multinational corporations wishing to demonstrate consistency among their operational locations. The principal focus of ISO 9000 is on developing a high-quality product and ensuring customer satisfaction with that product. From its inception, AAALAC International has been primarily concerned with how animals are cared for when they are involved in research, testing, and teaching. AAALAC International was founded on the belief that sound science requires that animals being used in its pursuit should be used and cared for in a harmonized and humane manner under high standards. It is not surprising, then, that the AAALAC International accreditation process continues to emphasize animal care and use, with the goal of minimizing the “animal variable” in science. Animals as well as the uses to which they are put vary in myriad ways. This requires that AAALAC International standards and procedures be sufficiently flexible to accommodate these diversities. The principal focus of AAALAC International is linking animal welfare and quality of science. The differing focuses of GLP regulations, ISO standards, and AAALAC International’s accreditation program quite naturally lead to variations in how the three systems accomplish their goals. GLP regulations are designed to minimize the possibility that a new drug or device will cause unintended harm to human patients. This leaves the regulatory agencies (i.e., FDA in the United States and EMEA in Europe) with few options other than to require absolute adherence to means-related
84550.indb 75
11/11/10 6:24:09 PM
76
Handbook of Laboratory Animal Science
standards. Almost no deviation from very prescriptive “engineering” standards is allowed, with verification of adherence to GLP requirements accomplished by strict internal controls and by inspections by representatives of national authorities. These national authorities have ultimate approval power and are the final arbiters of whether or not an organization has complied with the requirements. The main strength of this system, when applied to animal care and use programs, is that it ensures the consistency of studies. On the other hand, its application to the animal program belongs to a broader scheme within an institution, in which the use of animals is just a part of the process. The legal implications of this system make it applicable only at institutions performing specific, regulated studies. The initiative to implement GLPs at an institution will arise from the legal requirements when planning to perform such studies. People responsible for the research animals will follow the instructions of the quality assurance unit. With its focus principally on products and consumer satisfaction, ISO is primarily concerned with processes and procedures. Customer feedback is critical to enhancement, on an ongoing basis, of the processes that will lead to a consistent, high-quality product. When ISO management systems are applied to animal care and use, the products can be healthy animals, as well as the services provided to maximize accurate, reproducible scientific results. Because of its focus on customer satisfaction, ISO is mainly applied at institutions offering products or services. Because it is a voluntary system, the initiative to implement it may come from the desire of the people managing the animal care and use programs to improve their procedures. In most cases, the application is part of the general scheme of the institution (as is the case for GLPs). Another common feature with the GLPs is that, in both cases, the assessment is performed by external evaluators (government inspectors in the case of GLPs) who are not necessarily involved in the laboratory animal science field. Accomplishing the goal of promoting animal welfare internationally requires a level of flexibility generally greater than that associated with ISO and limited under GLPs. Accommodating different species and the circumstances of their use is manifested in the standards employed by AAALAC International and in the procedures followed to assess organizations’ adherence to them. The Guide and reference resources that form the AAALAC International standards are almost exclusively performance or outcome based, providing (sometimes wide) latitude in the way that organizations may choose to meet the performance standards (Bayne and Martin 1998; Bayne and Miller 2000; Guillen 2010). Under these circumstances, the only practical method for determining whether or not an organization has met the desired outcome is through the process of peer review by individuals with proven abilities to apply professional judgment in a consistent manner (Miller 1998). The main strengths of AAALAC International accreditation are that it can be applied to all sorts of settings in which the use of laboratory animals is involved and that the assessment is performed by specialists in the field, making the process more useful for the institution in terms of suggested improvements. Since AAALAC International is a voluntary scheme focused on all the different areas of the animal care and use program, its practical implementation at the institution may mainly depend on the work of the laboratory animal unit, especially if it is supported by senior management, technical staff, an occupational health and safety officer, and the quality assurance unit, if required. Summary Consistent, high-quality animal care and use programs are critical to the research effort at several levels. In this chapter we have presented three systems that affect the quality of such programs: the ISO standardization process, AAALAC International accreditation, and GLP regulations. Most developed countries have some form of national legislation dealing with the humane treatment of animals used for scientific purposes, thereby establishing a required minimum level of what might be termed “quality.” Those planning to submit animal-related data to a national authority for premarketing approval must meet applicable GLP regulatory requirements.
84550.indb 76
11/11/10 6:24:10 PM
Assessment of Animal Care and Use Programs and Facilities
77
A second driving force for quality in this area is an ethical imperative to care for and use animals in such a way that pain, distress, and discomfort are minimized to the greatest extent practicable. This includes recognizing that using animals in the advancement of science to improve human health is a privilege to be aware of at all times. Third, high-quality animal care and use practices are important elements in producing statistically valid and reproducible scientific data. Through the application of uniform high standards in the areas of animal procurement, husbandry, veterinary care, and experimental use, animal-related variables may be minimized. Although they represent three distinct assessment systems, ISO standards, AAALAC International accreditation, and compliance with GLP regulations should be considered complementary rather than an “either/or” situation. Integration of all three systems into institutional animal care and use programs, when appropriate (for instance, when GLP compliance is legally required), serves to enhance the quality of the overall program. Each system focuses on a different aspect of the use of animals in the advancement of science, and each employs a unique approach to verify that institutions are meeting its requirements. A key component of many animal care and use programs is the committee or body charged with oversight responsibilities, known as the ethics committee, institutional animal care and use committee, animal welfare body, or by another name. In addition to serving as another form of internal quality monitoring, these bodies provide natural loci for the integration of the three systems into an overall institutional program. The histories of the three systems presented are instructive in understanding their differences and how they may fit together as a whole. Questionable and even deadly practices in industry led to GLP regulations designed to reduce risks to humans. These regulations took the form of strict compliance with very prescriptive record keeping and other documentation requirements. This emphasis on consistent, well-documented internal procedures and controls is highly congruent with the ISO 9000 system, which grew from its founders’ desire to facilitate uniformity through adherence to international standards. Animal care and use programs operating under ISO 9000 rely on frequent feedback from “customers” (i.e., scientists using animals) to improve continuously the quality of service provided and management of the overall program. Rooted in the desire to improve laboratory animal care, AAALAC International focuses specifically on institutional mechanisms and practices that directly affect animal welfare. Because of great variations in species and even individual animals, evaluating animal welfare is best done through the application of outcome or performance-based standards, rather than through a prescriptive, means-based approach. To accomplish this, AAALAC International involves an independent thirdparty evaluation conducted by expert peer reviewers in on-site evaluations of animal care and use programs, facilities, and the animals themselves. Organizations that consistently apply well-documented, high-quality management practices (such as those promoted by GLP regulations and ISO standards) that are also responsive to the needs of animals will find achieving AAALAC International accreditation a natural next step. Thus, far from being exclusionary, the three assessment systems presented here can work in concert to improve the quality of all aspects of animal care and use programs.
References Association for Assessment and Accreditation of Laboratory Animal Care International. 2010a. About AAALAC, AAALAC’s member organizations (http://www.aaalac.org/about/memberorgs.cfm). ———. 2010b. The accreditation program, AAALAC’s reference resources (http://www.aaalac.org/ accreditation/resources.cfm). Baldeshwiler, A. M. 2003. History of FDA good laboratory practices. Quality Assurance Journal 7 (3): 157–161.
84550.indb 77
11/11/10 6:24:10 PM
78
Handbook of Laboratory Animal Science
Bayne, K. A., and D. P. Martin. 1998. AAALAC International: Using performance standards to evaluate an animal care and use program. Lab Animal 27 (4): 32. Bayne, K. A., and J. G. Miller. 2000. Assessing animal care and use programs internationally. Lab Animal 29 (6): 27. Code of Federal Regulations. 1985 (revised January 2002). Title 9 (animals and animal products), subchapter A (animal welfare). Washington, D.C.: Office of the Federal Register. Cook, J. D. 2010. Good laboratory practice (GLP) versus CLIA. Guest essay, Westgard QC. http://www.westgard.com/guest16.htm (accessed February 22, 2010). Council of Europe. 2006. Appendix A of the European convention for the protection of vertebrate animals used for experimental and other scientific purposes (ETS No 123). European Union. 1986. Council directive on the approximation of laws, regulations, and administrative provisions of the member states regarding the protection of animals used for experimental and other scientific purposes. Directive 86/609/EEC. Guillen, J. 2010. The use of performance standards by AAALAC International to evaluate ethical review in European institutions. Laboratory Animals 39 (2): 49. Howard, B., H. van Herck, J. Guillen, B. Bacon, R. Joffe, and M. Ritskes-Hoitinga. 2004. Report of the FELASA Working Group on evaluation of quality systems for animal units. Laboratory Animals 38 (2): 103. International Organization for Standardization. 2010a. Founding (http://www.iso.org/iso/about/the_iso_story/ iso_story_founding.htm). ———. 2010b. ISO 9000 essentials (http://www.iso.org/iso/iso_catalogue/management_standards/iso_9000_ iso_14000/iso_9000_essentials.htm). ———. 2010c. Quality management principles (http://www.iso.org/iso/iso_catalogue/management_standards/ iso_9000_iso_14000/qmp.htm). ———. 2010d. The ISO 9000 family—core standards (http://www.iso.org/iso/iso_catalogue/management_ standards/iso_9000_iso_14000/iso_9000_selection_and_use/iso_9000_family_core_standards.htm). ———. 2010e. What are standards? (http://www.ISO.CH/iso/en/aboutiso/introduction/index.html). Jansen, C. C. 1999. A quality system for laboratory animal facilities. Scandinavian Journal of Laboratory Animal Science 26 (1): 17. Latimer, J. 1997. Friendship among equals. International Organization for Standardization. 2010. The ISO story. http://www.iso.org/iso/about/the_iso_story/friendship_equals.htm (accessed February 22, 2010). Miller, J. 1998. International harmonization of animal care and use: The proof is in the practice. Lab Animal 27 (5): 28. Miller, J., and J. D. Clark. 1999. The history of the Association for Assessment and Accreditation of Laboratory Animal Care International. In 50 Years of laboratory animal science, ed. C. W. McPherson, and S. F. Mattingly, chap. 6. Memphis, TN: AALAS. National Research Council, Institute for Laboratory Animal Resources. 1996. Guide for the care and use of laboratory animals. Washington, D.C.: National Academy Press. National Research Council, Institute for Laboratory Animal Resources. 2010. Guide for the care and use of laboratory animals. Washington, D.C.: National Academy press. Organization for Economic Cooperation and Development. 1998. OECD principles of good laboratory practice (http://www.oecd.org/document/63/0,3343,en_2649_34381_2346175_1_1_1_1,00.html). Public Health Service Policy on Humane Care and Use of Laboratory Animals. 1996 (revised January 2002). Washington, D.C.: U.S. Department of Health and Human Services. Ritskes-Hoitinga, M., E. Kalisie-Korhonen, and A. Smith. 1999. Evaluation of quality systems for animal units: Report of the Scand-LAS working group. Scandinavian Journal of Laboratory Animal Science 26 (3): 117. Russell, W. M. S., and R. L. Burch. 1959. The principles of humane experimental technique. London: Methuen & Co. Ltd. U.S. Food and Drug Administration. 1978. Nonclinical laboratory studies, good laboratory practice regulations. U.S. Federal Register 43 (247): 59986–60020. ———. Nonclinical laboratory studies, good laboratory practice regulations. U.S. Federal Register 52 (172): 33768–33782 (http://www.fda.gov/ora/compliance_ref/bimo/glp/87fr-glpamendment.pdf). Van Herck, H. 1999. Implementation of a quality management system in the Central Laboratory Animal Institute (GDL) of the Utrecht Universiteit: Why and how? In Science and responsibility (abstracts of the 29th Annual Symposium of the Scandinavian Society for Laboratory Animal Science), 33, 1999. Van Velden-Russcher, J. A., and H. van Herck. 2001. Quality management in the CLAI. Der Tierschutzbeauftragte 1 (01): 48.
84550.indb 78
11/11/10 6:24:10 PM
Assessment of Animal Care and Use Programs and Facilities
79
Bibliography About AAALAC. AAALAC’s member organizations (http://www.aaalac.org/about/memberorgs.cfm). Accreditation program. AAALAC’s reference resources (http://www.aaalac.org/accreditation/resources.cfm). Gordon, H. 1999. The history of the Public Health Service Policy on Humane Care and Use of Laboratory Animals. In 50 Years of laboratory animal science, ed. C. W. McPherson and S. F. Mattingly, chap. 21. Memphis, TN: AALAS. International Organization for Standardization. 2010. ISO 9000. An introduction (http://www.praxiom.com/ iso-intro.htm). ———. Selection and use of the ISO 9000: 2000 Family of standards (http://www.iso.org/iso/iso_catalogue/ management_standards/iso_9000_iso_14000/iso_9000_essentials.htm). ———. ISO 9001:2000—What does it mean in the supply chain? http://www.iso.org/iso/iso_catalogue/ management_standards/iso_9000_iso_14000/more_resources_9000/9001supchain.htm
84550.indb 79
11/11/10 6:24:10 PM
84550.indb 80
11/11/10 6:24:10 PM
Chapter 5
Education and Training
Nicole Duffee, Timo Nevalainen, and Jann Hau Contents Introduction....................................................................................................................................... 82 Europe............................................................................................................................................... 83 Legal Requirements for Training................................................................................................. 83 FELASA Guidelines for Teaching and Training.........................................................................84 FELASA Accreditation of Training Programs............................................................................ 85 United States..................................................................................................................................... 86 Legal Requirements for Training................................................................................................. 86 Research Personnel................................................................................................................. 86 Institutional Animal Care and Use Committee (IACUC)...................................................... 87 Laboratory Animal Veterinarians........................................................................................... 87 Occupational Health and Safety.............................................................................................. 88 Professional Qualifications.......................................................................................................... 88 Common Approaches to Training................................................................................................ 89 Program Trainers......................................................................................................................... 91 Verification of Training—Animal Use Competence................................................................... 91 Record Keeping............................................................................................................................92 Training Resource Organizations..................................................................................................... 93 Association for Assessment and Accreditation of Laboratory Animal Care, International (AAALAC)............................................................................................................. 93 American Association for Laboratory Animal Science (AALAS).............................................. 93 Animal Welfare Information Center (AWIC), U.S. National Agricultural Library.................... 93 Canadian Association for Laboratory Animal Science (CALAS)............................................... 93 Canadian Council for Animal Care (CCAC)............................................................................... 93 Federation of European Laboratory Animal Science Associations (FELASA)..........................94 International Network for Humane Education (InterNICHE).....................................................94 Laboratory Animal Welfare Training Exchange (LAWTE)........................................................94 Laboratory Animal Management Association (LAMA).............................................................94 Norwegian Reference Center for Laboratory Animal Science and Alternatives.........................94 Public Responsibility in Medicine and Research (PRIM&R).....................................................94 Scientists Center for Animal Welfare (SCAW)............................................................................94 Summary........................................................................................................................................... 95 References......................................................................................................................................... 95 81
84550.indb 81
11/11/10 6:24:10 PM
82
Handbook of Laboratory Animal Science
Introduction Individuals who work with laboratory animals must have appropriate skills and qualifications for performing experimental procedures on animals. The protection of research animals is derived from national laws and regulations; the specific requirements differ geographically. Despite differences in the text of the laws, the intent and rationale are the same. The proper training of individuals to handle and restrain animals and to perform experimental procedures is essential for fulfilling legal requirements that laboratory animals be treated humanely when used for research, testing, and education. Insufficient training may result in substantial harm to animals and may cause occupational injuries from animal scratches and bites or infections from zoonotic agents. In research involving hazardous materials, inadequate training may also cause hazardous agent exposure of other personnel, other animals in the facility, and the environment. Competence in animal methodologies is recognized to support the quality of research and testing. Changes in the physiological status of research animals may have an impact on research data as sources of nonexperimental variation. Inadequate handling and poor methodology, which may distress and even injure an animal, may alter the immune system due to stress and may activate inflammation due to injury or infection (see also Chapter 18). All personnel who are involved in the care, treatment, and use of animals should be appropriately qualified to perform their duties. The definition of personnel often broadly includes individuals working in research (scientists, visiting scientists, technicians, and students), animal husbandry, veterinary care, animal facility management, ethics committees, animal facility maintenance, and animal research administration. Although national mandates may not specify training by type of personnel, an inclusive approach to training benefits any animal research program. When personnel are knowledgeable about key animal welfare standards and issues, the quality of the animal care and use program is enhanced, which in turn better supports the institution’s research mission. Individuals about to work with animals in the research program should be qualified based on the nature of their work and contact with the animals. Specialized training is needed for the responsibilities of managing animal research resources and directing or overseeing programs of animal care, veterinary services, and research administration. Husbandry staff and research staff have similar requirements for training in animal behavior, care, and handling. Both also require an orientation to the animal welfare laws, regulations, policies, and guidelines. However, husbandry personnel require more detailed training on animal housing requirements and housekeeping practices for the animal facility environment. It is recognized that many institutions assign their staff to positions that combine husbandry and research functions; in such a case, the training requirements would reflect the range of their duties. Scientists should be qualified on the design of animal experiments, including the concepts of alternatives to animals and on the conduct of an alternatives search (i.e., research and testing methods that minimize the number of animals required to obtain valid results, minimize animal pain and distress, and replace animals with alternative models). They and other research staff should be qualified in the procedures that they will carry out on animals in their experimental studies, such as behavioral observations, venipuncture, or surgery. Their training should include how to care humanely for the animals during experimental manipulations such as supporting and monitoring their physiological function, providing adequate analgesia to minimize pain, and either supporting their recovery or euthanizing them. Training requirements should include safety practices and the use of protective equipment to avoid exposure to hazardous materials often associated with animal research. Maintenance staff may require training on the safe handling of animal facility operating systems and waste materials. Staff having indirect contact with animals should also be offered training on how the animals used in the research program are appropriately handled and treated in compliance
84550.indb 82
11/11/10 6:24:10 PM
Education and Training
83
with the relevant laws. These individuals may have qualms about the care and use of the animals in the institution. They can benefit from learning about the general nature of the regulatory mandates and the institution’s commitment to animal welfare standards. This training may also be helpful to the organization to avert the development of animal rights activism in persons who might have no preconceived bias against animal research, but may be receptive to animal rights notions in an institutional environment where information is not forthcoming about the animal care and use program. Many institutions all over the world are voluntarily accredited in their laboratory animal program by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). The AAALAC provides training on accreditation standards and processes, with the aim of helping institutions develop the capability to achieve and maintain AAALAC accreditation. The AAALAC also conducts program assessments, apart from accreditation, which assist institutions in determining their program strengths and weaknesses in all areas, including training animal research staff. The assessment outcomes are valuable to identify necessary programmatic or facility corrections and to gauge the institution’s readiness to seek accreditation. Differences in the specifics of laws and regulations among countries or regions have led to the development of different training systems for complying with animal research requirements. Though it is recognized that many countries have mandates covering the competence required of personnel to work with laboratory animals, the first two sections of this chapter describe the systems of training in Europe and the United States. The third section identifies organizations that are training resources in animal research. Europe Legal Requirements for Training The European Union (EU) and Council of Europe (CoE) are federations including most, but not all, European countries. Both have statutes to govern and regulate the use of vertebrate animals for scientific purposes1,2 (see also Chapter 3, this volume). In addition to these statutes, each nation has its own legislation. National legislation should be harmonized with and may exceed the requirements of the European regulations, which serve as the minimum standard. The Council of Europe Convention 123 (the European Directive), Article 26, states that “persons who carry out, take part in, or supervise procedures on animals, or take care of animals used in procedures, shall have had appropriate education and training.” In Europe, an experimental procedure conducted on animals is generally understood to include any manipulation involving discomfort greater than a sensation comparable to a needle stick. The CoE Convention ETS 123 contains the provision that parties should hold multilateral consultations to examine the progress of its implementation and the need for revision or extension of any of its provisions on the basis of new facts or developments. During the last decades, three multilateral consultations have been held. The parties in 1993 adopted the following resolution on education and training of persons working with animals: “This resolution presents guidelines for topics to be included in educational and training programs for four categories of persons working with laboratory animals (from animal caretakers to specialists in animal science).” This requirement is reiterated in the European Directive. Article 5 of the European Directive states that a competent person must oversee the well-being and the state of health of the animals. Article 19 stipulates that a veterinarian or other competent person should be charged with advisory duties in relation to the well-being of the animals. The provisions on competence warrant special attention. Laws and regulations are poor tools when not based upon an understanding of what constitutes humane and responsible animal care and use.
84550.indb 83
11/11/10 6:24:11 PM
84
Handbook of Laboratory Animal Science
Therefore, well-directed education and training provide the means for gaining this understanding and for evaluating ethical ramifications. Article 7 of the European Directive states that only a person considered to be competent or under the direct responsibility of such a person should perform animal experiments. This provision is amplified by Article 14, which states that persons conducting, collaborating in, or supervising experiments or the care of laboratory animals should have appropriate education and training. It is essential that the people involved in the design and conduct of experiments should have received an education in a scientific discipline relevant to the experimental work. They also should be capable of handling and taking care of laboratory animals. Each member state must specify how the provision of competence is to be implemented within national legislation. The experience of training approaches in specific European countries has been described.3 As of the date of this publication, the European Directive is being revised, and the commission’s proposal, with amendments by the leading agricultural and rural development committee, has been accepted by the European parliament in its first reading. The current version of the directive draft (Article 20) states that member states shall ensure that the persons to be authorized have the appropriate veterinary or scientific education and training and have evidence of the requisite competence. Only authorized persons can carry out any of the following functions: performance of procedures on animals, including their killing by a humane method supervision of those taking care of animals supervision or design of procedures and projects
These can be understood to correspond to FELASA A, B, and C category competencies (see following discussion). In the EU, education and training issues are allocated to member countries responsible for publishing minimum requirements with regard to education and training, as well as requirements for obtaining, maintaining, and demonstrating requisite competence. In Article 24, the directive states that a designated veterinarian with expertise in laboratory animal medicine must be in each breeding, supplying, and user establishment; this corresponds to FELASA D-category competence. New elements in the directive draft are that all authorizations of persons will be granted for a limited period, not exceeding 5 years, and that authorization of persons is only granted on the basis of evidence of the requisite competence. Furthermore, member states will be required to ensure the mutual recognition of education and training qualifications and authorization to conduct designated procedures. FELASA Guidelines for Teaching and Training The Federation of European Laboratory Animal Science Associations (FELASA) has made recommendations on the educational and training requirements for staff and personnel working with laboratory animals. Many countries have introduced strict regulations regarding competence based upon these FELASA guidelines: Category A is concerned with persons taking care of animals (animal technicians) and encompasses four levels of training.4 Category B addresses persons carrying out animal experiments (research technicians).5 Category C concerns persons responsible for directing animal experiments (scientists).4 Category D is for specialists in laboratory animal science or laboratory higher management (specialists).6
These recommendations, published as FELASA Working Party Reports in the journal Laboratory Animals (UK), are in the form of syllabi for the training of each category of personnel. The recommendations for categories A and D are in-depth, career-type education, while categories B and C are relatively short courses. Because of the multilingual nature of the continent and differences in
84550.indb 84
11/11/10 6:24:11 PM
Education and Training
85
job titles, recommendations for training objectives are handled through descriptions of duties and responsibilities instead of a defined nomenclature for each category. CoE has adopted the competence categories; hence, they can be regarded as the basic competence classification in Europe. Category A training was originally organized to address four levels of staff needs and experience: level 1 for basic laboratory animal care, level 2 for those with at least 2 years of work experience, level 3 for those with an additional 3 years of work experience (5 years total), and level 4 for those in higher management or specialization. Each level is described by typical duties and responsibilities. Individuals are expected to possess a practical competence in order to progress to the next level. However, category A guidelines are being revised by a FELASA working group, and the outcome is expected to contain fewer levels than is the case at the moment. Category B guidelines contain a set of topics and subtopics to be taught during 40 hours of instruction. Practical exercises are emphasized by means of a recommendation that half of the instruction be devoted to hands-on exercises or demonstrations. Category C training has a prerequisite of a full university degree in a biomedical discipline, such as animal biology, medicine, or veterinary medicine. The category C curriculum is double the length of category B: 80 hours or an equivalent. The multilateral consultations of CoE have adopted this curriculum. Category D training has a prerequisite of a degree in biomedical or veterinary sciences, demonstrated competence at category C level, and appropriate experience in the field of laboratory animal science. Since laboratory animal science combines knowledge of the scientific method and animal welfare principles, as well as specific research on laboratory animals, the category D curriculum includes completion of a scientific project to be published in a peer-reviewed journal. In all, the curriculum is expected to take 2 years. A FELASA working group is developing standards on continuing education for category D that are expected soon after the date of this publication. The European College of Laboratory Animal Medicine, established in 2002, provides training and certification of veterinarians in laboratory animal medicine. Candidate veterinarians must be the first or primary author of two original scientific articles. Candidates must have successfully completed training in a laboratory animal medicine program (category D) acceptable to ECLAM or have other appropriate experience. The ECLAM board examination is based on work experience and passing an examination on topics of the FELASA D category, supplemented with specific, veterinary-oriented topics. Upon passing the examination, the candidates are qualified as ECLAM diplomates. All active diplomates of the college are required to provide documentation of continuing education at specified intervals. FELASA Accreditation of Training Programs In 2003, FELASA established an accreditation system for its four categories of teaching programs (A through D) to provide quality assurance in instruction in laboratory animal science and promote further harmonization within Europe.7 The accreditation program is intended to drive a refinement in training for the benefit of continuous improvements in animal husbandry in research and in the conduct of animal experimentation. The review process is confidential and conducted via electronic communication. National liaison experts are involved in accreditation reviews and provide consultations for candidate programs. The accreditation process involves reviews of descriptions of training topics and objectives, the time allocations per topic, training methodologies, student qualifications, the status of the program as an academic/official/legal requirement, and assessment procedures for student and teacher competence. The European Science Foundation (ESF), the umbrella organization of the European research coordinating and funding organizations, issued a statement supporting accreditation of training programs that reads: “Investigators and other personnel involved in the design and performance of animal-based experiments should be adequately educated and trained. ESF member organizations
84550.indb 85
11/11/10 6:24:11 PM
86
Handbook of Laboratory Animal Science
should encourage the development and organization of accredited courses on laboratory animal science, including information on animal alternatives, welfare, and ethics.”8 The FELASA training accreditation program is the only one in existence, as of the date of this publication. United States Legal Requirements for Training Research Personnel The regulatory environment for the care and use of laboratory animals and therefore the basis for the training of personnel to work with these animals is anchored on a series of federal mandates: two laws, one regulation, and a policy. One federal law is the Animal Welfare Act, which is supported by the Code of Federal Regulations, Title 9, Subchapter A—“Animal Welfare.”9 These regulations, commonly known as the Animal Welfare Regulations, are enforced by the U.S. Department of Agriculture (USDA).10 The second federal law, the Health Research Extension Act of 1985, “Animals in Research,” is implemented by the Public Health Service Policy on Humane Care and Use of Laboratory Animals (often referred to as the PHS policy), which is enforced by the Office of Laboratory Animal Welfare (OLAW) of the National Institutes of Health.11,12 Most research institutions in the United States are covered by one of these laws and the corresponding regulations or policy, and many institutions are indeed covered by both. A guideline that addresses personnel training and qualifications is the Guide for the Care and Use of Laboratory Animals from the National Research Council.13 Both the PHS policy and AAALAC accreditation require compliance with the Guide’s standards. The PHS policy refers specification of training requirements to the Animal Welfare Regulations and the Guide for the Care and Use of Laboratory Animals because of the requirement for PHS-assured institutions to comply with these mandates. The Animal Welfare Regulations specify the training requirements for personnel as follows:10
84550.indb 86
(a) It shall be the responsibility of the research facility to ensure that all scientists, research technicians, animal technicians, and other personnel involved in animal care, treatment, and use are qualified to perform their duties. This responsibility shall be fulfilled, in part, through the provision of training and instruction to those personnel. (b) Training and instruction shall be made available, and the qualifications of personnel reviewed, with sufficient frequency to fulfill the research facility’s responsibilities under this section and Sec. 2.31. (c) Training and instruction of personnel must include guidance in at least the following areas: (1) Humane methods of animal maintenance and experimentation, including: (i) The basic needs of each species of animal (ii) Proper handling and care for the various species of animals used by the facility (iii) Proper pre-procedural and post-procedural care of animals (iv) Aseptic surgical methods and procedures (2) The concept, availability, and use of research or testing methods that limit the use of animals or minimize animal distress. (3) Proper use of anesthetics, analgesics, and tranquilizers for any species of animals used by the facility. (4) Methods whereby deficiencies in animal care and treatment are reported, including deficiencies in animal care and treatment reported by any employee of the facility. No facility employee, committee member, or laboratory personnel shall be discriminated against or be subject to any reprisal for reporting violations of any regulation or standards under the Act. (5) Utilization of services (e.g., National Agricultural Library, National Library of Medicine) available to provide information:
11/11/10 6:24:11 PM
Education and Training
(i) (ii) (iii) (iv)
87
On appropriate methods of animal care and use On alternatives to the use of live animals in research That could prevent unintended and unnecessary duplication of research involving animals Regarding the intent and requirements of the Act
Additional training requirements and qualifications for personnel working with agricultural animals are derived from the Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching, published by the Federated Animal Science Societies and adopted as a standard, via policy, by the USDA.14,15 Institutional Animal Care and Use Committee (IACUC) The Animal Welfare Regulations address the qualifications of the IACUC in Sec. 2.31 (a): “The Chief Executive Officer of the research facility shall appoint an Institutional Animal Care and Use Committee (IACUC), qualified through the experience and expertise of its members to assess the research facility’s animal program, facilities, and procedures.”10 A USDA policy requires that IACUC members receive training specific to their role in the animal-research program:16 IACUC members must be qualified to assess the research facility’s animal program, facilities and procedures. The research facility is responsible for ensuring their qualification, and this responsibility is filled in part through the provision of training and instruction. For example, IACUC members should be trained in understanding the Animal Welfare Act, protocol review, and facility inspections.
In regard to training requirements by the Public Health Service, the Health Research Extension Act of 1985 specifies in Sec. 495 (c)(1)(B):11 Scientists, animal technicians, and other personnel involved with animal care, treatment, and use by the applicant have available to them instruction or training in the humane practice of animal maintenance and experimentation, and the concept, availability, and use of research or testing methods that limit the use of animals or limit animal distress.
The reference in this statement to “other personnel” has generally been interpreted to include members of the IACUC. Laboratory Animal Veterinarians The Animal Welfare Regulations in sec. 2.31 (b)(3)(i) require that the IACUC include as a member a “Doctor of Veterinary Medicine, with training or experience in laboratory animal science and medicine, who has direct or delegated program responsibility for activities involving animals at the research facility.”10 This requirement for veterinary qualifications is reiterated in the PHS policy in sec. IV. A. 3.b.(1).12 Many laboratory animal veterinarians in the United States obtain their qualifications via the American College of Laboratory Animal Medicine (ACLAM), founded in 1957. Veterinarian candidates must first complete an ACLAM-approved training program or a minimum of 6 years of relevant, full-time experience in laboratory animal medicine. Training programs for laboratory animal veterinarians in the United States and Canada have been reviewed.17 ACLAM-approved programs offer a combination of didactic training, supervised practice of laboratory animal medicine, and mentored research experience. Candidates must also have published an original scientific article as a first author. Candidates must pass a certification exam to become qualified as ACLAM diplomats, who must fulfill continuing education requirements to maintain board certification.
84550.indb 87
11/11/10 6:24:11 PM
88
Handbook of Laboratory Animal Science
Occupational Health and Safety Personnel should receive training in the hazards used in the animal research program and related safety equipment and practices. Hazards are classified as biological, chemical, radiological, or physical agents. A federal law called the Occupational Safety and Health Act of 1970 requires that a safe working environment be provided to employees.18 Training on the nature of the occupational hazards and safety equipment and practices is integral to compliance with that law. This standard applies to all staff with regard to hazards associated with the animal facility. For specific recommendations on hazard assessment and safety practices in an animal facility, refer to Occupational Health and Safety in Animal Care and Use Programs from the National Research Council.19 Training for work with hazardous biological agents should be guided by the publication Biosafety in Microbiological and Biomedical Laboratories (BMBL), from the Centers for Disease Control and Prevention (CDC) and the National Institutes of Health.20 Regulations from both the CDC and USDA specify requirements for training in the use of select agents and toxins.21–23 Individuals must receive training on biosafety and biosecurity before they are allowed access to areas where select agents are used. This training should focus on the needs of the individual, the type of work, the risks from the exact select agent or toxin being used, facility security, and emergency procedures. Personnel who work in the vicinity of animals or who work with biological samples obtained from animals also require training related to the institutional occupational health and safety program. This includes individuals who perform tasks involving maintenance, transportation, or administration. Such staff may be exposed to allergens, animal wastes and biological samples, or physical and radiological hazards used in research. The level of risk should be assessed for each type of staff to determine whether safety training is necessary and what topics should be included. Professional Qualifications Certification in competence areas by professional organizations in the United States is encouraged by regulatory and accrediting authorities (USDA, OLAW, and AAALAC). The Guide for the Care and Use of Laboratory Animals identifies technician certification as an option for the provision of staff training (Institutional Responsibilities, Personnel Qualifications, and Training, p. 13)13: “There are a number of options for the training of technicians. Nondegree training, with certification programs for laboratory animal technicians and technologists, can be obtained from the American Association for Laboratory Animal Science (AALAS).” Professional societies have developed certification programs to certify the knowledge and competence of animal research staff in key job descriptions: laboratory animal technician certification by the American Association for Laboratory Animal Science (AALAS): assistant laboratory animal technician (ALAT) laboratory animal technician (LAT) laboratory animal technologist (LATG) laboratory animal facility manager certification by AALAS, the Laboratory Animal Management Association (LAMA), and the Institute for Certified Professional Managers (ICPM): certified manager of animal resources (CMAR) surgical technician certification by the Academy of Surgical Research (ASR): surgical research specialist (SRS) surgical research anesthetist (SRA) surgical research technician (SRT)
84550.indb 88
11/11/10 6:24:12 PM
Education and Training
89
IACUC administrator certification by Public Responsibility in Medicine and Research (PRIM&R): certified professional IACUC administrator (CPIA)
In each of these certification programs, educational and work-experience criteria are specified for eligibility, and candidates must pass an examination to attain certification. Certified laboratory animal technicians may maintain an additional credential as registered animal technicians (at all three certification levels) by continuing their education in a voluntary program known as the AALAS technician certification registry. To maintain their credentials, individuals holding any of the ASR specialty certifications, certified animal resource managers, and certified professional IACUC administrators must periodically recertify by complying with continuing education requirements. Many institutions encourage their staffs to attain certification in the previously mentioned specialty areas to support compliance with the training requirements of federal laws and policies. Often, institutions offer financial incentives, such as pay raises, bonuses, and payment of certification fees. Some institutions may even require certifications as a job requirement or a promotion criterion. Although the training for professional qualifications of veterinarians or veterinary technicians is beyond the scope of an institutional training program, the institution can enhance its compliance with U.S. animal welfare mandates by hiring technical staff with degrees and licensing as veterinary technicians. For example, veterinary technicians who have a 2-year degree in veterinary technology and are state licensed are well qualified to fill positions in research, veterinary medicine, and animal care. Institutions should provide opportunities and support for continuing education to personnel who are professionally licensed, certified, or registered so that they may maintain their qualifications. In informal statements on training programs, OLAW refers to AALAS technician certification and state professional licensing as indications of staff qualification:24 [OLAW] strongly recommends that institutions offer their staff access to training leading to certification in animal technology, such as that available from AALAS or a formally designated academic program. Institutions should also know and ensure compliance with any initial and continuing-education State requirements for the licensing of veterinary or animal health technicians.
Likewise, the Guide for the Care and Use of Laboratory Animals stresses the importance of continuing education (p. 16) to maintain staff qualifications:13 “Personnel caring for laboratory animals should also participate regularly in continuing-education activities, should be encouraged to be involved in local and national laboratory animal science meetings, and participate in other relevant professional organizations.” Common Approaches to Training Collectively, the United States mandates on laboratory animal welfare apply a performance standard to the assessment of compliance with all programmatic requirements, including the training of personnel. A performance standard focuses on the outcome of a process. For a training program, the outcome is measured in how well personnel are qualified to carry out the animal-related procedures. The institution can determine how its program is constituted in terms of staff resources and the objectives, methods, and frequency of training. When used to evaluate a training program, performance standards direct attention toward assessing staff expertise in animal care and use procedures as a means of determining the effectiveness of a training program. An effective training program is expected to result in staff competence and therefore in the humane and appropriate treatment of animals. A lack of competence points to a
84550.indb 89
11/11/10 6:24:12 PM
90
Handbook of Laboratory Animal Science
need for improvement in an institution’s training program. Typically, governmental and accrediting inspectors observe and query staff working with animals as a means of evaluating the effectiveness of a training program. The performance standard for personnel training is best expressed in the Guide for the Care and Use of Laboratory Animals (National Research Council), Animal Care and Use Program, Training and Education (p. 15):13 All personnel involved with the care and use of animals must be adequately educated, trained and/or qualified in basic principles of laboratory animal science to help assure high quality science and animal well-being. The number and qualifications of personnel required to conduct and support a Program depend on several factors, including the type and size of institution, the administrative structure for providing adequate animal care, the characteristics of the physical plant, the number and species of animals maintained, and the nature of the research, testing, teaching and production activities. Institutions are responsible for providing appropriate resources to support personnel training (Anderson 2007); however, the IACUC is responsible for providing oversight and for evaluating the effectiveness of the training program (Foshay and Tinkey 2007).
Not only can training be customized for the type of procedures and species used, training can also be adapted to the level of staff experience and competence and the degree of staff turnover. The option to use either formal or on-the-job training provides flexibility for each institution in meeting its needs and utilizing personnel and other resources as best fits the institution. In an informal statement, OLAW addressed the question of how much flexibility an institution has in the development of a training program to satisfy federal requirements:13 Each assured institution is responsible for training its staff to meet the performance requirements cited in paragraph IV.C.1.a–g of the PHS policy, and guidelines have been developed to assist institutions to meet these objectives. OLAW recognizes research programs vary from one institution to another, and are relative to the size and nature of the institution, staffing, numbers of species and individual animals maintained, and the kinds of research conducted. Therefore, the scope and depth of instructional programs and the frequency at which they are offered will also vary. At a minimum, however, the policy requires institutions to ensure that individuals who use or provide care for animals are trained and qualified in the appropriate, species-specific housing methods, husbandry procedures, and handling techniques. The institution must ensure that research staff members performing experimental manipulation, including anesthesia and surgery, are qualified through training or experience to accomplish such procedures humanely and in a scientifically acceptable fashion. They must also receive training or instruction in research and testing methods that minimize the number of animals required to obtain valid results and minimize animal distress. Institutions must also ensure that staff whose work involves hazardous biological, chemical, or physical agents have training or experience to assess potential dangers and select and oversee the implementation of appropriate safeguards.
As an example of the degree of flexibility in institutional training programs, if influxes of new staff are frequent, a program may emphasize entry-level training. In an institution with a low turnover of staff, there is an opportunity to present advanced topics for continuing-education purposes. General topic areas for training in an institutional program have been described for animal care technicians, researchers, and staff involved in maintenance, transportation, and administration.25 Additional topics and best practices in training have been described for agricultural programs.26 Strategies for training programs have been described for research staff.27 A common approach for assessing the training needs of researchers is to use the animal-use protocol as a vehicle to identify research personnel who will have contact with animals and to describe their individual qualifications for performing their animal-related duties. A training coordinator or the attending veterinarian assesses the training needs of the investigator’s staff in the proposed research activity,
84550.indb 90
11/11/10 6:24:12 PM
Education and Training
91
in collaboration with these personnel. Training requirements are completed before the protocol is approved by the IACUC. The postapproval monitoring of protocols is helpful to identify additional training needs related to issues of regulatory compliance or specific procedures.28 Animal care technicians require training in basic facility operations, husbandry procedures, housing environment, and health care procedures for each species. Training should be planned according to the particular job functions performed.29 Institutional training strategies for IACUCs have been described.30,31 The training of the IACUC should encompass review of animal protocols and animal-related activities, review of program policies, and inspections of facilities. Often, the IACUC chair, the IACUC administrator, or the attending veterinarian has the task of orienting and training new IACUC members to their role. It is also desirable that an experienced IACUC member be assigned as a mentor to the new member. Mentoring is especially important for the nonscientist or nonaffiliated member, since such persons require an additional orientation to the terminology and science of animal research. On its Web site, OLAW provides an online tutorial for animal care and use committee members and others interested in learning about the PHS policy.32 OLAW also sponsors seminars and training throughout the United States that address regulatory requirements and topics of interest to IACUCs. Program Trainers In the United States, the legal responsibility for meeting training requirements related to the use of animals in research lies with the institution; thus, this responsibility falls to the IACUC. Since the IACUC has oversight responsibilities over all aspects of the animal care and use program, it takes on the role of assuring that staff training meets the standards imposed by federal mandates. Commonly, a staff member is designated as a training coordinator to ensure that all personnel who work with animals are provided with appropriate training services.33 It is a best practice that the institutional training coordinator oversees all training activities, such as when additional staff provide training services on species and methods for which they have particular expertise. When no program trainers are on staff, as is often the case in small institutions, IACUCs typically rely on the veterinary staff and senior members of the animal care staff for training research personnel on animal care and use procedures. These individuals offer expertise in basic methodologies, and they can recommend other resources that may best address a specialized training need. When expertise is not available within the animal facility, training may be sought from outside sources, such as researchers with no reporting relationship to the animal facility and experts from other institutions. In areas where multiple research institutions are in close proximity, training programs should capitalize on the expertise available at other institutions. Verification of Training—Animal Use Competence In response to the U.S. federal requirement that animal research staff be qualified to work with laboratory animals, many institutions verify the competence of staff in animal-use activities however the skills were obtained—for example, via the institutional training program or training by colleagues. Verification of competence is relatively straightforward with animal care staff, due to the lines of authority within the animal facility. A greater challenge is to verify competence for research staff members who have no reporting relationship with the animal facility unit. Implementing a system for research staff necessitates an institution-wide policy and administrative support. Methods to assess competence in animal use among research staff generally follow two basic approaches. Some institutions utilize a “certification” process whereby designated trainers visit a laboratory and observe the conduct of animal-use procedures. Individual researchers or, in some cases, entire laboratories receive a recognition, or certification, of their competence for
84550.indb 91
11/11/10 6:24:12 PM
92
Handbook of Laboratory Animal Science
specific procedures. This recognition may encompass the authorization to train others in the same procedure. Institutions using this method typically maintain a documentation of certified individuals or laboratories. A second approach couples competence verification with compliance assessment in a postapproval monitoring program.28 Designated compliance staff visit laboratories periodically to assess compliance with protocols and other animal welfare mandates. Such visits typically have the objectives of verifying that animal-use procedures are conducted in accordance with relevant mandates, verifying that drugs and medical materials are current with respect to product expiration dates, providing information on federal mandates and institutional policies, and distributing related literature. Minor deficiencies are handled as opportunities for providing guidance and training. The balance between training and noncompliance citation should emphasize foremost the training. Training and compliance staff should have a cooperative and helpful attitude toward the research staff, commensurate with principles of customer service. Staff members should remain mindful that their primary goal is to provide support to animal research through their role of assuring compliance with federal and institutional mandates. Record Keeping In the United States, there is no specific statement in federal laws, regulations, or policies to maintain documentation of training. That is, documentation of training is not included among the types of facility records that must be maintained by an institution and that can be inspected by federal agencies, according to federal laws or regulations. Nevertheless, federal and accrediting authorities consistently expect to have access to training records during an inspection of a research institution. The USDA has affirmed this expectation in published articles on the subject of training for compliance with the Animal Welfare Act.34,35 Because training programs are federally mandated, a system of documentation is the only practical way for an institution to prove and for inspectors to verify compliance with the training requirements. Record keeping of training activities is therefore a practical necessity for compliance with federal animal welfare mandates. Various approaches for training documentation have been described.36 Training records may be paper based or electronic (e.g., residing in a database, spreadsheet, etc.). If electronic, they may be stored locally on a computer, intranet, or a host server inside or outside the institution. Records may be associated with individuals, departments, or training activities. The training records may be generated, archived, and accessed by a training coordinator, the IACUC, or another administrative unit. The records may be stored centrally or segregated by administrative units. The documentation should demonstrate that the institution’s training program meets the objectives of government mandates (see previous sections). Moreover, federal and accrediting authorities expect that the documentation system will encompass the entire training program, the records will accurately reflect the training activities, the records will be comprehensible, and documentation will be readily accessible on demand. Institutions commonly use learning management software (database systems) as a means to provide staff training and track and document the training activities. Commercial database systems for animal facility management often incorporate a module for training documentation. Such systems may allow some software customization for accommodating features of the institution’s training program. In addition, some database systems offer connectivity of the database with e-mail to facilitate communication with staff on training matters. Licensing or purchasing access to software minimizes the time and expense needed to create a learning management system. Developing and maintaining learning management systems require extensive technical and computing support, which are important factors to consider when planning the resources to create an in-house system.
84550.indb 92
11/11/10 6:24:12 PM
Education and Training
93
Training Resource Organizations A growing number of resources is available for enhancing a training program and developing professional qualifications. Instructional media (videotapes, CD-ROM, DVD-ROM, and Webbased applications) can augment in-class teaching or provide self-directed learning. Models and devices that simulate anatomical structures are valuable aids for teaching skills in animal procedures. Conferences and workshops provide professionals with new information and techniques to incorporate into their institutional training program. The following organizations offer information and resources for training within the animal research field. Association for Assessment and Accreditation of Laboratory Animal Care, International (AAALAC) For the benefit of institutions seeking or supporting AAALAC accreditation, AAALAC (http:// www.aaalac.org) provides training services via on-site workshops on various topics related to its expectations and perspectives on animal care and use programs. American Association for Laboratory Animal Science (AALAS) The AALAS (http://www.aalas.org) is a membership organization of laboratory animal professionals and institutions. AALAS offers manuals, references, and media for animal research training. The AALAS Learning Library is an online learning management system with courses on animal research topics (http://www.learninglibrary.org). The AALAS National Meeting is an annual conference for laboratory animal science professionals and includes a resource center for viewing an assortment of instructional media from many sources. The CompMed, TechLink, and IACUC-Forum listservs provide a forum for discussions on issues and methodologies within the laboratory animal science community. The IACUC Web site (http://www.iacuc.org) is a link archive for information of interest to IACUCs, and it includes links to institutional training programs for a comparison of methodologies and content. Animal Welfare Information Center (AWIC), U.S. National Agricultural Library The AWIC (http://awic.nal.usda.gov) provides literature and searching assistance to the animal research community on the three Rs (i.e., the alternatives to replace, reduce, or refine the use of animals in research).37 AWIC also conducts workshops on meeting the information requirements of the U.S. federal Animal Welfare Act. Canadian Association for Laboratory Animal Science (CALAS) The CALAS (http://www.calas-acsal.org) is a membership association that offers an annual symposium on animal research topics. CALAS distributes videos on technical procedures for common laboratory animal species and on Canadian animal care committees. The association offers a certification program, called the registry, with three levels for laboratory animal technicians. Canadian Council for Animal Care (CCAC) As the national organization responsible for setting and maintaining standards for the care and use of research animals in Canada, the CCAC (http://www.ccac.ca) publishes syllabi with technical descriptions of animal procedures for use as training modules for research personnel.
84550.indb 93
11/11/10 6:24:12 PM
94
Handbook of Laboratory Animal Science
Federation of European Laboratory Animal Science Associations (FELASA) The FELASA (http://www.felasa.eu) holds a triennial educational symposium for laboratory animal science professionals. FELASA also develops guideline documents through its working parties representing the common interests of its constituent European national and regional associations. International Network for Humane Education (InterNICHE) The InterNICHE (http://www.interniche.org) focuses on animal use and alternatives within biological science, medical, and veterinary medical education. The InterNICHE Alternatives Loan System maintains a library of multimedia and physical models for loan to teachers and students in any country. Laboratory Animal Welfare Training Exchange (LAWTE) The LAWTE (http://www.lawte.org) is an organization of trainers and training coordinators in the laboratory animal field. Members use a listserv for discussions about training issues and methods. The Web site contains information on training media, methodologies, and references. A conference on training issues and methodologies is held every 2 years in the United States. Laboratory Animal Management Association (LAMA) The LAMA (http://www.lama-online.org/) is a membership organization that holds an annual conference jointly with the Allied Trades Association (ATA). As one of three partner organizations in the certified manager of animal resources (CMAR) program, LAMA provides training courses to support those preparing for CMAR certification. Norwegian Reference Center for Laboratory Animal Science and Alternatives The Norwegian Reference Center for Laboratory Animal Science and Alternatives (http:// oslovet.veths.no) maintains the Norina database, which is an English-language archive of information on training materials for use in biological science education. This database offers descriptions of training media that can serve as alternatives or supplements to the use of animals in student teaching at all levels of education. The center also offers Textbase, which is a database of current textbooks within the field of laboratory animal science. Public Responsibility in Medicine and Research (PRIM&R) The PRIM&R (http://www.primr.org) is a membership organization of administrative professionals of IACUCs and institutional review boards. PRIM&R offers conferences, educational programs, guidelines, publications, and training media in ethics and compliance with U.S. mandates in the animal research field. PRIM&R has a certification program culminating in a credential: certified professional IACUC administrator (CPIA). Scientists Center for Animal Welfare (SCAW) The SCAW (http://www.scaw.com) is a membership organization of individuals and institutions involved in animal research. SCAW offers conferences, educational programs, guidelines, and publications in ethics and compliance with U.S. mandates in the animal research field.
84550.indb 94
11/11/10 6:24:12 PM
Education and Training
95
Summary Personnel training is key to the humane care and use of animals in research—so much so that it is specifically addressed in the regulatory mandates covering research animals in Europe, the United States, Canada, and many other countries. Considering the diversity of animal species, husbandry requirements, procedures, and types of personnel common in a large research program, no two institutions are exactly alike in their research program. Therefore, every institution must develop a training program that best fits its needs. The many training resources available commercially or for free from laboratory animal science associations and other organizations are helpful as building blocks for institutions to use in assembling their own training program. The creation of staff positions for training responsibilities has helped many institutions achieve an effective training program, thus supporting their compliance with regulatory mandates for animal welfare. FELASA accreditation of training programs is expected to drive further improvements in research animal welfare. An important component in a training program is the opportunity and encouragement provided by the institution for personnel to obtain professional certification in technical, managerial, or administrative specialties within the laboratory animal field. Maintaining certification through continuing education infuses the staff—and thus the institution—with a culture of ongoing improvement in animal welfare, operations, and science to benefit the research program.
References
84550.indb 95
1. European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes, Council of Europe, European Treaty Series (ETS) No. 123, March 18, 1986. 2. European Commission. 1986. Directive for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (86/609/EEC). Official Journal of the European Commission L 358:1. 3. van Zutphen, B. 2007. Education and training for the care and use of laboratory animals: An overview of current practices. ILAR Journal 48 (2): 72–74. 4. Federation of European Laboratory Animal Science Associations, Working Group on Education. 1995. FELASA recommendations for the education and training of persons working with laboratory animals: Category A and C. Lab Anim 29:121. 5. Federation of European Laboratory Animal Science Associations, Working Group on Education of Persons Carrying Out Animal Experiments. 2000. FELASA recommendations for the education and training of persons carrying out animal experiments (category B). Lab Anim 34:229. 6. Federation of European Laboratory Animal Science Associations, Working Group on Education of Specialists. 1999. FELASA recommendations for the education of specialists in laboratory animal science (category D). Lab Anim 33:1–15. 7. Federation of European Laboratory Animal Science, Working Group on Accreditation of Laboratory Animal Science Education and Training. 2002. Lab Anim 36:373–377. 8. European Science Foundation policy briefing. 2001. Use of animals in research, 2nd ed. 9. Animal Welfare Act as amended, USC, title 7, sections 2131 to 2156. Washington, D.C.: U.S. Government Printing Office. 10. Code of Federal Regulations. 1985. Title 9, chapter 1 (animals and animal products), subchapter A (animal welfare). Washington, D.C.: Office of the Federal Register. 11. Health Research Extension Act of 1985. U.S. Public Law 99–158, section 495, animals in research. Washington, D.C.: U.S. Government Printing Office. 12. Public Health Service. 1996. Public Health Service policy on humane care and use of laboratory animals, Washington, D.C.: U.S. Department of Health and Human Services, 28 (PL 99–158, Health Research Extension Act, 1985).
11/11/10 6:24:13 PM
96
Handbook of Laboratory Animal Science
13. Institute of Laboratory Animal Research, National Research Council. 2010. Guide for the care and use of laboratory animals. Internet [September 2010] Washington, D.C.: National Academy Press. http://www.nap.edu/catalog.php?record_id=12910. 14. Guide for the care and use of agricultural animals in agricultural research and training. 1999. Savoy, IL: Federation of Animal Science Societies. 15. Department of Agriculture, Animal and Plant Health Inspection Service. 2000. Animal welfare: Farm animals used for nonagricultural purposes. Federal Register 65 (23). 16. USDA APHIS. 2006. Animal resource guide, policy #15, IACUC membership. 17. Colby, L. A., P. V. Turner, and M. A. Vasbinder. 2007. Training strategies for laboratory animal veterinarians: Challenges and opportunities. ILAR Journal 48 (2): 143–155. 18. Occupational Safety and Health Act of 1970, as amended, USC, Title 29, Chapter 15, Section 651. Washington, D.C.: U.S. Government Printing Office. 19. Committee on Occupational Safety and Health in Animal Research Facilities, Institute of Laboratory Animal Research, Commission on Life Sciences, National Research Council. 1997. Occupational health and safety in animal care and use programs. Washington, D.C.: National Academy Press. 20. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Institutes of Health. 2007. Biosafety in microbiological and biomedical laboratories, 5th ed. (http://www.cdc.gov/OD/ohs/biosfty/bmbl5/bmbl5toc.htm). 21. CFR [Code of Federal Regulations]. 2005. Title 7 (agriculture), part 331 (possession, use, and transfer of select agents and toxins, final rule). Washington, D.C.: Office of the Federal Register. 22. CFR [Code of Federal Regulations]. 2005. Title 42 (public health), parts 72 and 73 (possession, use, and transfer of select agents and toxins, final rule). Washington, D.C.: Office of the Federal Register. 23. Gonder, J. C. 2005. Select agent regulations. ILAR Journal 46 (1): 4–7. 24. Potkay, S., N. L. Garnett, J. G. Miller, C. L. Pond, and D. J. Doyle. 1995. Frequently asked questions about the public health service policy on humane care and use of laboratory animals. Lab Animal (U.S.) 24:24. 25. Committee on Educational Programs in Laboratory Animal Science, Institute of Laboratory Animal Research, Commission on Life Sciences, National Research Council. 1991. Education and training in the care and use of laboratory animals: A guide for developing institutional programs. Washington, D.C.: National Academy Press. 26. Underwood, W. J. 2005. Training for best practices for agricultural programs. Laboratory Animals 34 (8): 29–32. 27. Conarello, S. L., and M. J. Shepherd. 2007. Training strategies for research investigators and technicians. ILAR Journal 48 (2): 120–130. 28. Smelser, J. F., S. L. Gardella, and B. L. Austin. Protocol audits for postapproval monitoring of animal use protocols. Lab Anim 34 (10): 23–27. 29. Pritt, S., and N. E. Duffee. 2007. Training strategies for animal care technicians and veterinary technical staff. ILAR Journal 48 (2): 109–119. 30. Haywood, J. R., M. Greene, M. L. James, and K. Bayne. 2005. Engaging the IACUC through comprehensive training. Lab Animal (U.S.). 34 (10): 33–37. 31. Greene, M. E., M. E. Pitts, and M. L. James. 2007. Training strategies for IACUC members and the institutional official. ILAR Journal 48 (2): 131–142. 32. Office of Laboratory Animal Welfare. PHS policy on humane care and use of laboratory animals tutorial (http://grants1.nih.gov/grants/olaw/tutorial/index.htm). 33. Kennedy, B. W. 2002. Creating a training coordinator position. Lab Animal (U.S.). 31 (6): 34–38. 34. Slauter, J. E. 2000. Evaluation of a training program: A USDA perspective. Lab Animal (U.S.) 29:25. 35. Slauter, J. E. 1999. When the USDA veterinary medical officer looks at your training program. Animal Welfare Information Center Bulletin 10 (1–2). 36. Pritt, S., P. Samalonis, L. Bindley, and A. Schade. 2004. Creating a comprehensive training documentation program. Lab Animal (U.S.) 33 (4): 38–41. 37. Russell, W. M. S., and R. L. Burch. 1959. The principles of humane experimental technique. London: Methuen & Co. Ltd.
84550.indb 96
11/11/10 6:24:13 PM
Chapter 6
Laboratory Animal Science and Service Organizations Patri Vergara and Gilles Demers Contents Introduction....................................................................................................................................... 98 International Organizations.............................................................................................................. 98 International Council for Laboratory Animal Science..................................................................... 98 Background............................................................................................................................. 98 Aims........................................................................................................................................99 Membership.............................................................................................................................99 Programs.................................................................................................................................99 Meetings..................................................................................................................................99 Laboratory Animal Science Associations...................................................................................... 100 Membership............................................................................................................................... 100 Aims........................................................................................................................................... 100 Laboratory Animal Science Organizations by Regions of the World............................................ 100 Europe........................................................................................................................................ 100 Federation of European Laboratory Animal Science Associations...................................... 100 FELASA Constituent and Affiliate Members....................................................................... 101 Other European Associations................................................................................................ 101 The Americas............................................................................................................................. 101 American Association for Laboratory Animal Science........................................................ 101 Other American Associations............................................................................................... 104 Asia............................................................................................................................................ 106 Asian Federation of Laboratory Animal Science Associations............................................ 106 Oceania...................................................................................................................................... 106 Africa......................................................................................................................................... 106 Professional Organizations............................................................................................................. 108 International Association of Colleges of Laboratory Animal Medicine................................... 108 Background........................................................................................................................... 108 Other Professional Organizations................................................................................................... 108 Animal Care and Welfare Organizations....................................................................................... 108 Miscellaneous Organizations.......................................................................................................... 108 Acknowledgments........................................................................................................................... 113 Reference........................................................................................................................................ 113 97
84550.indb 97
11/11/10 6:24:13 PM
98
Handbook of Laboratory Animal Science
Introduction In parallel with the development of biomedical research, laboratory animal science and service organizations have been in constant evolution and development over the last 50 years. Demands for higher quality animals, together with greater concern for animal welfare, are the driving forces behind the development of organizations that provide support for people working in the field of laboratory animal science. The first laboratory animal organizations were created in the 1950s and 1960s in North America, Japan, and Europe: the American Association for Laboratory Animal Science (AALAS; formerly ACP) in 1950, Japanese Association for Laboratory Animal Science (JALAS) in 1952, Laboratory Animal Science Association (LASA) in 1963, International Council for Laboratory Animal Science (ICLAS) in 1956 (originally the International Council for Laboratory Animals, ICLA), and Canadian Council on Animal Care (CCAC) in 1968. Since then, increasing levels of biomedical research in other countries, mainly in Asia and Central and South America, have created an explosion of new laboratory animal science and service organizations around the world. The role of the International Council for Laboratory Animal Science (ICLAS) as an international umbrella organization is important in this worldwide development. In several parts of the world, regional organizations were created to maintain links between national scientific organizations and to lead the field in providing policies and guidelines related to laboratory animal care and use. The FELASA (Federation of European Laboratory Animal Science Associations) in Europe, ACCMAL (Central American, Caribbean, and Mexican Association of Laboratory Animal Science) in Central America, FESSACAL (Federation of South American Societies and Associations of Laboratory Animal Science Specialists) in South America, and AFLAS (Asian Federation of Laboratory Animal Science Associations) in Asia have played important roles in this respect. Several countries now have more than one laboratory animal science organization serving different goals—namely, continuing education and training, production of guidelines, scientific communication, accreditation, and certification programs. In order to give the reader a useful and practical overview of the principal laboratory animal science organizations around the world, the authors have classified them according to their primary aims and scope—that is, international organizations, laboratory animal science associations, professional organizations, animal care and welfare organizations, and miscellaneous associations. However, this chapter is limited to providing basic information about the main laboratory animal science organizations. It is neither all inclusive nor exhaustive because the continuous growth of laboratory animal science and the ongoing creation of new organizations make this impossible. The reader should visit organizations’ Web sites for further information. When a Web site was not available at the time of publication, a mailing or e-mail address was given. International Organizations International Council for Laboratory Animal Science Background Established in 1956 under the auspices of the Council for International Organizations of Medical Sciences (CIOMS), the International Union of Biological Sciences (IUBS), and the
84550.indb 98
11/11/10 6:24:13 PM
Laboratory Animal Science and Service Organizations
99
United Nations Educational, Scientific, and Cultural Organization (UNESCO), the International Council for Laboratory Animal Science (ICLAS) is a nongovernmental organization for international cooperation in laboratory animal science. Aims As stated in its constitution, the aims of the ICLAS are to • • • • •
Promote and coordinate the development of laboratory animal science throughout the world Promote international collaboration in laboratory animal science Promote quality definition and monitoring of laboratory animals Collect and disseminate information on laboratory animal science Promote the humane use of animals in research through recognition of ethical principles and scientific responsibilities
Membership The ICLAS has national, scientific/union, and associate members, which number about 36, 43, and 41, respectively. National members represent national perspectives, scientific/union members represent national/regional laboratory animal science associations and international nongovernmental unions, and associate members represent commercial, academic, and scientific organizations that support the aims of ICLAS. Programs The ICLAS has always emphasized international harmonization (Demers et al. 2006) in respect of both the quality of laboratory animals and the ethics of animal experimentation in order to ensure that animal experimentation is conducted both scientifically and ethically. With regard to ethics, ICLAS has facilitated international recognition and acceptance of guidelines on several topics, including the harmonization on euthanasia and humane end points contained in a published ICLAS statement paper. ICLAS has also been involved in the development of programs to improve the quality of laboratory animals by providing the scientific community with information and tools in the areas of health and genetic monitoring. These programs have been developed by the ICLAS Laboratory Animal Quality Network comprising members of ICLAS and well-recognized international professionals. Current projects include the Performance Evaluation Program (PEP), which is designed to enable participating laboratories to self-assess their health monitoring and diagnostic techniques. Meetings An international scientific meeting is held in association with the general assembly every 4 years. It is organized by a national or scientific member and is often held in association with regional or local organizations. Other regional meetings and courses are organized by individuals in the various regions of the world under the auspices of five ICLAS regional committees for the following regions: Europe, Asia, Africa, Oceania, and the Americas. This has allowed ICLAS to focus on each region and to apply local intellectual resources to issues within each of the regions. The ICLAS publication is the ICLAS FYI Bulletin and its address is www.iclas.org.
84550.indb 99
11/11/10 6:24:13 PM
100
Handbook of Laboratory Animal Science
Laboratory Animal Science Associations Membership In terms of membership, laboratory animal science associations consist of professionals whose work is related to laboratory animals (e.g., scientists, veterinarians, animal technicians, educators, etc.). Although there are some exceptions, members are normally from the geographical area where the association is based. Some associations, such as the AALAS or LASA, also have an international membership. Aims The aims shared by most laboratory animal science associations are to • Promote a better and more rational use of laboratory animals, following the ethical principles of the three Rs • Provide training and education in laboratory animal science and welfare • Advance the knowledge, skills, and status of those who care for and use laboratory animals • Promote informed public discussion regarding the use of animals in research • Inform scientists about the appropriate use and care of animals in research
Laboratory Animal Science Organizations by Regions of the World Europe Federation of European Laboratory Animal Science Associations Background The Federation of European Laboratory Animal Science Associations (FELASA) is composed of independent European national and regional laboratory animal associations and was established by them in 1978. The FELASA is managed solely by representatives of its constituent associations. Aims • To represent common interests of constituent associations in the furtherance of all aspects of laboratory animal science by coordinating the development of education, animal welfare, health monitoring, and other aspects of laboratory animal science in Europe by such means as meetings, study groups, and publications • To act as a focus for the exchange of information on laboratory animal science among European states • To establish and maintain appropriate links with international and national bodies, as well as with other organizations concerned with laboratory animal science • To promote the recognition and consultation of FELASA as the specialist federation in laboratory animal science and welfare throughout Europe
Programs Within Europe, FELASA sees as its role not only to respond rapidly to European Union and Council of Europe developments, but also to guide the thinking of these bodies by offering them timely and authoritative advice. Within the purpose of establishing harmonization and developing
84550.indb 100
11/11/10 6:24:14 PM
Laboratory Animal Science and Service Organizations
101
high standards of education and training in laboratory animal science in Europe, FELASA has established an accreditation system for teaching programs that follows the four categories for which FELASA has published guidelines. The FELASA has published guidelines on education and training for persons responsible for the welfare of laboratory animals, for persons responsible for the design and conduct of studies involving animals, and for specialists in laboratory animal science. It has published several recommendations for the health monitoring of breeding and experimental units and for the detection, relief, and control of any pain or suffering in them, and a guidance paper for the accreditation of laboratory animal diagnostic laboratories. It has also recently approved guidelines for accreditation of health monitoring in breeding and experimental units. Meetings A FELASA international scientific meeting is organized every 3 years by each of the constituent organizations in turn. Awards In 2007, FELASA established a triennial award to recognize and reward a defined piece of original scientific work on any aspect of laboratory animal science, housing, husbandry, or welfare. This work must have led to or have the potential to lead to changes or improvements in the utilization of animals in research or testing or in laboratory animal science education within Europe. Publications The FELASA’s recommendations and guidelines are published on FELASA’s Web site, www. felasa.eu, and in Laboratory Animals (www.lal.org.uk), the official journal of FELASA, GV-SOLAS, ILAF, LASA, NVP, SECAL, SGV, and SPCAL. Published quarterly, Laboratory Animals is a peerreviewed journal and is a leading publication in the field of laboratory animal science. FELASA Constituent and Affiliate Members The FELASA is currently composed of 17 European associations and one affiliate member, and it represents laboratory animal scientists and technicians in at least 22 European countries. Details of these associations can be found in Table€6.1. Other European Associations For details of other European associations not affiliated with FELASA, see Table€6.2. The Americas American Association for Laboratory Animal Science Background Established in 1950, the American Association for Laboratory Animal Science (AALAS) is the world’s largest laboratory animal association and, in consequence, a premier forum for the exchange of information and expertise in the production, care, and use of laboratory animals.
84550.indb 101
11/11/10 6:24:14 PM
102
Handbook of Laboratory Animal Science
Table€6.1╅FELASA Constituent and Affiliate Members FELASA Constituent Associations Name of Organization
Web Site
Number of Members
Publications
Association Française des Sciences et Techniques de l’Animal de Laboratoire: AFSTAL (French Association of Laboratory Animal Sciences and Techniques)
www.afstal.com
500
Journal Sciences et Techniques de l’Animal de Laboratoire, STAL
Associazione Italiana per Scienze degli Animale da Laboratorio: AISAL (Italian Association for Laboratory Animal Sciences) Baltic Laboratory Animal Science Association: Balt-LASA (serves organizations, societies, and individual specialists of Estonia, Latvia, and Lithuania) Belgian Council for Laboratory Animal Science: BCLAS Croatian Laboratory Animal Science Association: CroLASA (founded 1981) Czech Laboratory Animal Science Association: CLASA Gesellschaft für Versuchstierkunde: GV-SOLAS (German Society for Laboratory Animal Science) Hellenic Society of Biomedical and Laboratory Animal Science: HSBLAS Hungarian Laboratory Animal Science Association: HLASA Laboratory Animal Science Association: LASA (founded 1963) Nederlandse Vereniging voor Proefdierkunde: NVP (Dutch Association for Laboratory Animal Science) Scandinavian Society for Laboratory Animal Science: Scand-LAS
www.aisal.org
162
AISAL News
Schweizerische Gesellschaft für Versuchtierkunde: SGV (Société Suisse pour la Science des Animaux de Laboratoire; Swiss Laboratory Animal Science Association) Serbian Association for Laboratory Animal Science SLASA Sociedad Española para las Ciencias del Animal de Laboratorio: SECAL (Spanish Society for Laboratory Animal Science)
Sociedade Portuguesa de Ciências em Animais de Laboratório (Portuguese Society of Laboratory Animal Sciences): SPCAL
E-mail: osvaldas.
[email protected] [email protected]
15
www.bclas.org
175
E-mail:
[email protected] E-mail:
[email protected]
200
www.gv-solas.de
43 1001
Shares Laboratory Animals
E-mail:
[email protected];
[email protected] E-mail:
[email protected] [email protected] www.lasa.co.uk
30
262
LASA Newsletter
www.proefdierkunde.nl
121
NVP Newsletter
www.scandlas.org
250
www.sgv.uzh.ch
200
Scandinavian Journal of Laboratory Animal Science Shares Laboratory Animals
E-mail:
[email protected] www.secal.es
402
Shares Laboratory Animals; also publishes a newsletter, Animales de Laboratorio
http://www.spcal.pt
FELASA Affiliate Association Israeli Laboratory Animal Forum ILAF
84550.indb 102
www.ilaf.org.il
69
11/11/10 6:24:14 PM
Laboratory Animal Science and Service Organizations
103
Table€6.2╅Other European Laboratory Animal Associations Not Affiliated with FELASA Name of Organization Finland Laboratory Animal Science: FinLAS
Web Site www.uku.fi/FinLAS/
Number of Members
Publications
47
Aims The AALAS advances responsible laboratory animal care and use to benefit people and animals. Membership The AALAS now comprises more than 13,000 clinical veterinarians, technicians, technologists, biomedical scientists, educators, and business people representing all aspects of the laboratory animal research field. The three levels of membership—bronze, silver, and gold—are available to individuals, businesses, and institutions. AALAS Annual Meeting Over 200 educational presentations are made each year at the AALAS National Meeting. These seminars, platform sessions, special lectures, workshops, and roundtable discussions cover a wide range of topics of interest to AALAS members and other laboratory animal professionals. Certification Program Nationally recognized as the authoritative endorsement of technician competence, the AALAS technician certification program certifies three levels of technical knowledge. The technician certification registry is a voluntary program for technicians to demonstrate a current, credible level of knowledge. Laboratory animal facility managers will be interested in the AALAS certified manager of animal resources (CMAR) program, which is designed to raise competency and professionalism in the field. The annual CEU requirement ensures that certified animal resource managers are abreast of the latest developments in management techniques and theory. The Institute for Laboratory Animal Management (ILAM) is an education program for professional laboratory animal managers focusing on management concepts applicable to the animal resource industry. National Awards Eight professional and technical awards for excellence in the field of laboratory animal science are given each year at the AALAS National Meeting. Publications The AALAS offers numerous publications to its members. Comparative Medicine, a bimonthly journal listed in Index Medicus, is a leading publication in the field of comparative and experimental
84550.indb 103
11/11/10 6:24:14 PM
104
Handbook of Laboratory Animal Science
medicine. The bimonthly Journal of the American Association for Laboratory Animal Science features articles on clinical, technical, managerial, and philosophical subjects, as well as association news. Tech Talk is a quarterly newsletter offering current information and technology. The AALAS Reference Directory includes a “Products and Services Guide” section, technical information, and certification and education information and forms. AALAS also publishes training manuals for each certification level and other materials for professional development and education, with Spanish versions and CD-ROMs currently in development. Listservs The AALAS manages three listservs related to laboratory animal science: CompMed, TechLink, and IACUC-FORUM. These enable professionals involved in the care and use of laboratory animals to get quick answers and talk about issues relating to comparative medicine, laboratory animal medicine, or biomedical research. Details of how to enroll on these listservs may be found at www. aalas.org. Public Outreach The AALAS Foundation provides funding for projects to promote awareness of biomedical research and its contributions to society. Through the foundation, AALAS provides a variety of public outreach tools in the forms of flyers, posters, Web sites, videos and DVDs, and classroom materials. The address is www.aalas.org. Related sites maintained by AALAS are www.IACUC. org, www.kids4research.org, and http://foundation.aalas.org. Other American Associations Details of the North American organizations are in Table€6.3. For details of Central and South American laboratory animal science organizations, see Tables€6.4 and 6.5, respectively.
Table€6.3╅North American Laboratory Animal Science Associations Name of Organization
84550.indb 104
Web Site
American Association for Laboratory Animal Science: AALAS (established 1950)
www.aalas.org www.IACUC.org www.kids4research.org www.foundation.aalas. org
Canadian Association for Laboratory Animal Science—Association canadienne pour la science des animaux de laboratoire: CALAS/ ACSAL (founded 1962)
www.calas-acsal.org
Number of Members 13,000
887
Publications Comparative Medicine Journal of the American Association for Laboratory Animal Science Tech Talk AALAS Reference Directory Member’s magazine (quarterly)
11/11/10 6:24:14 PM
Laboratory Animal Science and Service Organizations
105
Table€6.4â•… Central American and Mexican Laboratory Animal Science Associations Name of Organization Asociación Centroamericana del Caribe y Mexicana de la Ciencia de Animales de Laboratorio: ACCMAL (Central American, Caribbean, and Mexican Association of Laboratory Animal Science) with members from Costa Rica, Mexico, Honduras, El Salvador, Nicaragua, Panama, Cuba, and Colombia Asociación Cubana de la Ciencia de los Animales de Laboratorio: SCCAL (Cuban Association for Laboratory Animal Science) Asociación Mexicana para las Ciencias del Animal de Laboratorio: AMCAL (Mexican Association for Laboratory Animal Science)
Web Site E-mail:
[email protected]
Number of Members
Publications
200
www.bioterios.com
http://amcal-ac.blogspot.com
111
Animales de Experimentación Small Species (AMMVEPE) Journal
Table€6.5â•…South American Laboratory Animal Science Associations Name of Organization Federation of South American Societies and Associations of Laboratory Animal Science Specialists: FESSACAL Asociación Argentina de Ciencia y Tecnología de Animales de Laboratorio: AACyTAL (Argentinean Association of Science and Technology of Laboratory Animals) FESSACAL Constituent Association Asociación Chilena de Ciencias del Animal de Laboratorio: ASOCHICAL (Chilenian Association for Laboratory Animal Science) FESSACAL Constituent Association Asociación Uruguaya de Ciencia y Tecnología de Animales de Laboratorio: AUCyTAL (Uruguayan Association of Science and Technology of Laboratory Animals) FESSACAL Constituent Association
Web Site
www.aacytal.com.ar
Number of Members
Publications
100
www.bioterios.com
www.bioterios.com
(continued)
84550.indb 105
11/11/10 6:24:14 PM
106
Handbook of Laboratory Animal Science
Table€6.5â•…South American Laboratory Animal Science Associations (continued) Name of Organization Asociación Venezolana de Ciencias del Animal de Laboratorio: SOVECAL (Venezuelan Association for Laboratory Animal Science) FESSACAL Constituent Association
Web Site
Number of Members
Publications
www.bioterios.com E-mail:
[email protected]
Asia Asian Federation of Laboratory Animal Science Associations The Asian Federation of Laboratory Animal Science Associations (AFLAS) was established in 2003 by six national or regional associations or societies for laboratory animal science in Asia. Aims of AFLAS The AFLAS aims to promote, through the Asian Congresses on Laboratory Animal Science, the development of laboratory animal science in Asia; to review scientific, technical, and educational problems in laboratory animal science; to develop other relevant activities in the interests of laboratory animal science; and to contribute to animal welfare. Membership Membership comprises Asian national and regional associations or societies of laboratory animal science that conform to the aims of AFLAS. For a list of AFLAS members, see Table€6.6. Work of AFLAS The AFLAS organizes the Asian Congress on Laboratory Animal Science, held once every 2 years, and encourages the participation of scientists and technical workers in the field of laboratory animal science in the congress. To this end, AFLAS selects a host member country to host the congress and collaborates with it in the preparation of the congress. AFLAS also advises and nominates membership of working groups, symposia, or other functions when invited by appropriate organizations. Another area of AFLAS’s work is to facilitate the exchange of information among member associations and societies on activities and functions of common interest and to collaborate closely with other organizations with similar fields of activity. The address is www. aflas-office.org/. Oceania Australia and New Zealand have well-established laboratory animal science associations. For details, see Table€6.7. Africa For details of African associations, see Table€6.8.
84550.indb 106
11/11/10 6:24:15 PM
Laboratory Animal Science and Service Organizations
107
Table€6.6╅AFLAS Constituent Associations Name of Organization
Web Site
Chinese Association for Laboratory Animal Science: CALAS Chinese Taipei Society for Laboratory Animal Science: CSLAS Japanese Association for Laboratory Animal Science: JALAS (established 1952) Korean Association for Laboratory Animal Science: KALAS Laboratory Animal Scientist’s Association of India: LASAI Laboratory Animal Science Association of Malaysia: LASAM (founded 1995) Philippine Association for Laboratory Animal Science: PALAS
www.calas.org.cn/
Singapore Association for Laboratory Animal Science: SALAS Thai Association for Laboratory Animal Science: TALAS
www.salas.sg
www.cslas.org/
wwwsoc.nii.ac.jp/jalas/ index.html
www.kalas.or.kr/
Number of Members
Publications
2500
310
1367
Experimental Animals (quarterly)
500
www.lasaindia.org
www.medic.ukm.my/laru/ LASAM.htm www.aflas-office.org/
The PALAS Code of Practice for the Care and Use of Laboratory Animals in the Philippines (1993)
E-mail: vsprt@mahidol. ac.th
Table€6.7╅Oceania Name of Organization Australian and New Zealand Laboratory Animal Association: ANZLAA
Web Site www.anzlaa.org
Number of Members
Publications
400
ANZLAA was originally known as ASLAS (Australian Society for Laboratory Animal Science) and as ANZLAS (Australian and New Zealand Society for Laboratory Animal Science)
84550.indb 107
11/11/10 6:24:15 PM
108
Handbook of Laboratory Animal Science
Table€6.8╅Africa Name of Organization
Web Site
South African Association for Laboratory Animal Science: SAALAS
Number of Members
E-mail: paulineh@ savp.co.za
Publications Quarterly newsletter and annual SAALAS bulletin
Tunisian Association of Laboratory Animal Science: ATSAL
Professional Organizations This section lists the organizations formed by professional groups (i.e., veterinarians, technicians, etc.) who work in the field of laboratory animal science. Some of these organizations have a well-established accreditation system. International Association of Colleges of Laboratory Animal Medicine Background The International Association of Colleges of Laboratory Animal Medicine (IACLAM) brings together national and regional colleges of laboratory animal medicine and provides a common platform at the global level for communication by and representation of these colleges and their diplomates. Current members are the American College of Laboratory Animal Medicine (ACLAM), the European College of Laboratory Animal Medicine (ECLAM), the Japanese College of Laboratory Animal Medicine (JCLAM), and the Korean College of Laboratory Animal Medicine (KCLAM). For details of these colleges, see Table€6.9. IACLAM is an association of associations—specifically the member colleges of laboratory animal medicine. Appointed representatives of these colleges serve on the IACLAM board. IACLAM is an associate member of the World Veterinary Association (WVA). The Web site is www.iaclam.org. Other Professional Organizations In addition to college organizations, several professional associations are all veterinary associations apart from the Institute of Animal Technology. For details of these organizations, see Table€6.10. Animal Care and Welfare Organizations Table€6.11 gives details of the principal organizations around the world that focus on animal care and welfare. Miscellaneous Organizations Table€6.12 provides details of other organizations that also have programs related to laboratory animals.
84550.indb 108
11/11/10 6:24:15 PM
Laboratory Animal Science and Service Organizations
109
Table€6.9╅ IACLAM Members Name and Web Address of€Organization
Aims
Membership
American College of Laboratory Animal Medicine: ACLAM
To encourage education, training, and research in laboratory animal medicine
Web site: www.aclam.org
To establish standards of training and experience for qualification of specialists in this field, and to recognize qualified specialists by certification
Open to all veterinarians who are graduates of a college or school of veterinary medicine accredited or approved by the AVMA, or who possess an educational commission for foreign veterinary graduate (ECFVG) certificate, or who are qualified to practice veterinary medicine in a state, province, or possession of the United States, Canada, or other country; have satisfactory moral character and impeccable professional behavior; and have been certified as diplomates in accordance with ACLAM bylaws
European College of Laboratory Animal Medicine: ECLAM (established 2000)
To promote high standards for laboratory animal medicine by providing a structured framework to achieve certification of professional competence
Achieved full recognition by EBVS (European Board of Veterinary Specialization) in 2008 Web site: www.eclam.org
To promote scientific inquiry and exchange via progressive continuing-education programs
Japanese College of Laboratory Animal Medicine: JCLAM
The college is formed by charter diplomates selected to establish the college and the certification program
Web site: plaza.umin.ac.jp/JALAM/ Korean College of Laboratory Animal Medicine: KCLAM (established 2006)
To enhance humane techniques in animal experiments
Website: www.kclam.org/
To develop fields of disease of experimental animals, operation, anesthesia, pain relief, animal welfare, and animal protection
E-mail:
[email protected]
Only veterinarians may become diplomates of European veterinary specialty colleges. The ECLAM constitution defines a diplomate as a veterinarian who satisfies the ECLAM requirements with regard to training, experience, and competence in laboratory animal medicine, as required by the constitution
Veterinarians who have obtained the KCLAM certification
To provide training for the development of laboratory animal veterinarians To establish technical standards
84550.indb 109
11/11/10 6:24:15 PM
110
Handbook of Laboratory Animal Science
Table€6.10╅Other Professional Organizations Name and Web Address of€Organization American Society of Laboratory Animal Practitioners: ASLAP (founded 1966) Website: www.aslap.org
Aims
Publications
To provide a mechanism for the exchange of scientific and technical information among veterinarians engaged in laboratory animal practice To encourage the development and dissemination of knowledge in areas related to laboratory animal practice To act as a spokesperson for laboratory animal practitioners within the American Veterinary Medical Association (AVMA) To work with other organizations involved in the care and use of laboratory animals in representing common interests and concerns to the scientific community and the public at large To actively encourage its members to provide training for veterinarians in the field
Canadian Association for Laboratory Animal Medicine: CALAM—L’Association canadienne pour la médecine des animaux de laboratoire: ACMAL Web site: www.uwo.ca/animal/ website/CALAM
European Society of Laboratory Animal Veterinarians: ESLAV Website: www.eslav.org
To advise interested parties on all matters pertaining to laboratory animal medicine
Interface (quarterly, within the CALAS/ACSAL newsletter)
To further the education of its members and to promote ethics and professionalism in the field To promote appropriate veterinary care for all animals used in research, teaching, or testing To give veterinarians working in the field of laboratory animal medicine a forum to discuss issues
LAVA Briefing (newsletter)
To support ECLAM and to promote and disseminate expert veterinary knowledge within the field of laboratory animal science Institute of Animal Technology: IAT Web site: www.iat.org.uk
To advance and promote excellence in the technology and practice of laboratory animal care and welfare To enable animal technicians, technologists, and others professionally engaged in the field of animal science to receive appropriate training and qualifications
84550.indb 110
Animal Technology (worldwide publication); a monthly bulletin is also published; IAT has also published educational aids, books, videos, etc.
11/11/10 6:24:15 PM
Laboratory Animal Science and Service Organizations
111
Table€6.10╅Other Professional Organizations (Continued) Name and Web Address of€Organization Japanese Association for Laboratory Animal Medicine: JALAM Web site: plaza.umin.ac.jp/JALAM
Aims
Publications
To promote the advancement of research and education for laboratory animal medicine and the health of laboratory animals
The JALAM Newsletter is published biannually in Japanese, with contents in English available on the Web site
Table€6.11╅Animal Care and Welfare Organizations Name and Web Address of Organization
Aims
Publications
Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC International)
To promote the humane treatment of animals in science through voluntary accreditation and evaluation programs
Connection newsletter and AAALAC Update
Web site: www.aaalac.org
AAALAC’s Council on Accreditation conducts site visits and program evaluations that determine which institutions are awarded AAALAC accreditation
Australian and New Zealand Council for the Care of Animals in Research and Teaching: ANZCCART
To foster and promote best practice in ethical, social, and scientific issues relating to the use and well-being of animals in research and teaching
ANZCCART News (quarterly newsletter)
To publish guidelines on the care and use of animals in science
Guide to the Care and Use of Experimental Animals (two volumes in English and French; volume 1 is also available in Spanish)
AAALAC International i-brief
Web site: www.adelaide.edu.au/ ANZCCART/ Canadian Council on Animal Care: CCAC—Conseil canadien de protection des animaux: CCPA Web site: www.ccac.ca
To assess scientific institutions using animals To provide educational materials and activities to those who use and care for animals, as well as information to the public To provide a focus for the implementation of replacement, reduction and refinement alternatives
The Universities Federation for Animal Welfare: UFAW (founded 1926)
To promote and support developments in the science and technology that underpin advances in animal welfare
Web site: www.ufaw.org.uk
To promote education in animal care and welfare To provide information; organize symposia, conferences, and meetings To provide expert advice to governments and other organizations and help draft and amend laws and guidelines
84550.indb 111
Resource (newsletter available in English and French) Educational documents and guidelines on various aspects of animal care and use (available on the CCAC Web site)
Animal Welfare (quarterly) Feeding Garden Birds Best Practice Guidelines Handbooks on the care and management of animals Video programs useful to those who have responsibilities for animals and for teaching about their needs
11/11/10 6:24:15 PM
112
Handbook of Laboratory Animal Science
Table€6.12â•… Miscellaneous Organizations Name and Web Address of€Organization World Organization for Animal Health: OIE (Organization mondiale pour la santé animale, OIE) Website: www.oie.int
Aims To improve animal health worldwide To provide an international reference and to develop international standards and guidelines for animal health and welfare
Publications The OIE Guiding Principles on Animal Welfare (included in the OIE Terrestrial Animal Health Code, 2004)
To elaborate recommendations and guidelines covering animal health and welfare Institute of Laboratory Animal Research: ILAR (founded 1952) Web site: http://dels.nas.edu/ilar/
To evaluate and disseminate information on issues related to the scientific, technological, and ethical use of animals and related biological resources in research, testing, and education To provide independent, objective advice to the federal government, the international biomedical research community, and the public
Japanese Society for Laboratory Animal Resources: JSLAR (founded 1985) E-mail:
[email protected]
ILAR Journal (quarterly, peer-reviewed) Other publications and ILAR reports are listed on the ILAR Web site LABIO 21 (bulletin)
To supply high-quality laboratory animals
Web site in Japanese
To provide training and certify standards of laboratory animal technicians
Laboratory Animal Management Association: LAMA (founded 1984)
To enhance the quality of management and care of laboratory animals worldwide through education, knowledge exchange, and professional development
LAMA Review (quarterly)
To promote an information exchange among laboratory animal welfare trainers on training programs, systems, materials, and services for the purpose of promoting the highest standards of laboratory animal care and use
Materials, donated by LAWTE members, are available in various formats for download from the LAWTE Web site
Website: www.lama-online.org/ Laboratory Animal Welfare Training Exchange: LAWTE Web site: www.lawte.org
84550.indb 112
To foster the growth of the laboratory animal industry
Guide for the Care and Use of Laboratory Animals (1996) (available in several languages and currently being updated)
LAWTE also maintains a listserv (for members only) to facilitate discussion on the issues and methods of training research personnel
11/11/10 6:24:15 PM
Laboratory Animal Science and Service Organizations
113
Acknowledgments Special thanks to all of our colleagues from the various organizations who provided us with most of the information in this chapter. Reference Demers, G., G. Griffin, G. De Vroey, J. R. Haywood, J. Zurlo, and M. Bédard. 2006. Animal research. Harmonization of animal care and use guidance. Science 312 (5774): 700–701.
84550.indb 113
11/11/10 6:24:15 PM
84550.indb 114
11/11/10 6:24:16 PM
Chapter 7
Laboratory Animal Allergies and Zoonoses
Richard M. Preece and Anne Renström Contents Introduction..................................................................................................................................... 116 Laboratory Animal Allergy............................................................................................................ 116 Epidemiology and Pathogenesis..................................................................................................... 117 Prevalence: How Many People Have LAA?.............................................................................. 117 Incidence: The Risk of Getting LAA......................................................................................... 118 Exposure–Response Relationship.............................................................................................. 118 Animal Allergens....................................................................................................................... 119 Pathogenesis............................................................................................................................... 120 Clinical Manifestations................................................................................................................... 120 Symptoms................................................................................................................................... 120 Diagnosis.................................................................................................................................... 122 Prognosis.................................................................................................................................... 122 Case Management...................................................................................................................... 123 Control Strategy.............................................................................................................................. 124 Engineering Controls...................................................................................................................... 124 Separation................................................................................................................................... 126 General Ventilation.................................................................................................................... 126 Task Ventilation......................................................................................................................... 127 Automation................................................................................................................................. 127 Cage Systems............................................................................................................................. 128 Procedural Controls........................................................................................................................ 128 Reduction of the Number of Exposed Persons.......................................................................... 129 Animals, Stock Density, and Bedding....................................................................................... 129 Housekeeping............................................................................................................................. 130 Movements within the Facility.................................................................................................. 131 Work Permits and Visitors......................................................................................................... 131 Environmental Monitoring........................................................................................................ 131 Personal Controls............................................................................................................................ 132 Personal Hygiene........................................................................................................................ 132 Protective Clothing.................................................................................................................... 133 115
84550.indb 115
11/11/10 6:24:16 PM
116
Handbook of Laboratory Animal Science
Respiratory Protective Equipment............................................................................................. 133 Training and Education.............................................................................................................. 133 Preplacement Assessment: Assessing Risk................................................................................ 134 Health Surveillance.................................................................................................................... 135 Integrated LAA Risk Management................................................................................................ 135 Zoonoses......................................................................................................................................... 136 Prevention................................................................................................................................... 136 Disease-Free Animals................................................................................................................ 137 Awareness.................................................................................................................................. 137 Personal Protective Equipment.................................................................................................. 137 Health Care of Workers.............................................................................................................. 137 Protecting Animals from Human Disease................................................................................. 138 Important Zoonotic Diseases..................................................................................................... 138 Summary......................................................................................................................................... 139 References....................................................................................................................................... 139 Introduction Laboratory animals pose a continual biological hazard to exposed workers. They are a source of infection that can be transmitted to workers (zoonoses) and a source of biological materials that can induce allergy. Understanding of the mechanisms by which allergy develops and might be prevented continues to improve. In the past 5 years, there have been noteworthy advances in the understanding of sensitization and regarding the effectiveness of control technologies by measurement of allergen levels. Exposure to organic materials may also give rise to airway or skin symptoms without evidence of allergic sensitization. Laboratory animal work may furthermore entail contact with airway or skin irritants, such as latex gloves or chemicals—risks that may all be reduced by working conscientiously to diminish these exposures. Good laboratory hygiene means that zoonoses are uncommon. However, the hazard remains constant and, with the potential for life-threatening disease, vigilance is essential.
Laboratory Animal Allergy Laboratory animal allergy (LAA) is a common health problem in biomedical research; the prevalence of allergy has been found to be as high as 56% of animal-exposed workers (Bryant et al. 1995). Usually, the risk of developing allergic symptoms or sensitization to laboratory animals at work is between 20 and 30%. Sensitization to laboratory animal allergens can give rise to severe, acute (anaphylactic) reactions (Teasdale, Davies, and Slovak 1993) and disabling chronic illnesses (dermatitis and asthma). Both acute and chronic reactions to allergen exposure can have a significant adverse impact on affected workers and their employers. Several authors have reviewed this topic (Hunskaar and Fosse 1990; Bush and Stave 2003; Gordon and Preece 2003). Allergy to many different animals has been described, although the most commonly used rodents, mice and rats, are responsible for the majority of LAA cases (NRC [National Research Council] 1997). The allergens are present in urine, hair, skin, feces, and other material from the animals (Hunskaar and Fosse 1990). The most important laboratory animal allergens have been identified and characterized (Wood 2001). These include the major rat and mouse allergens, which are urinary proteins. The immunological mechanisms that give rise to symptoms of allergy have been characterized (Gordon and Preece 2003). Allergen exposure can lead to the production of specific
84550.indb 116
11/11/10 6:24:16 PM
Laboratory Animal Allergies and Zoonoses
117
immunoglobulin E (IgE) against allergens. The interaction between an allergen and specific IgE initiates a cascade of events that leads to the symptoms of allergy. This cascade will be repeated on each subsequent exposure to the allergen. The allergy may present as disorders of the nose and eyes (rhinoconjunctivitis), skin (urticaria and contact dermatitis), and chest (asthma) or, more rarely, as acute anaphylactic reactions (Hunskaar and Fosse 1990; Bush and Stave 2003; Gordon and Preece 2003). There is conflicting evidence for personal risk factors for development of LAA, especially tobacco smoking and a family history of allergic reactions (Seward 2001). The majority of studies have indicated that workers who have a tendency to produce IgE against allergens and to develop allergic symptoms (atopics) (Johansson et al. 2004) are about three- to fivefold more likely to develop LAA (Seward 1999). Also, atopics appear to develop LAA after a shorter time of exposure (Kruize et al. 1997). However, even atopy is not a sufficiently good predictor of LAA to be used in preplacement selection (NRC 1997; Botham et al. 1995; Newill, Evans, and Khoury 1986; Renström et al. 1994). A number of occupational risk factors for exposure to allergens have been identified, including high-risk tasks (such as handling animals or cleaning cages) (Nieuwenhuijsen et al. 1995; Hollander, Van Run, et al. 1997) and some work practices (e.g., choice of bedding) (Gordon et al. 1992). Studies have demonstrated a positive relationship between exposure to animal allergens and the prevalence of sensitization (Hollander, Heederik, and Doekes 1997; Cullinan et al. 1994; Heederik et al. 1999; Schweitzer et al. 2003; Elliott et al. 2005a). The combination of preexisting allergy or atopy and environmental risk factors (e.g., exposure level) may further increase the risk for LAA (Hollander, Heederik, and Doekes 1997; Hollander et al. 1999; Renström et al. 2001). The exposure–response relationship is complex and not yet fully understood (Jones 2008). The prevention of allergy is a significant challenge while uncertainty remains about which are the critical characteristics of exposure to control. Allergy-prevention strategies have generally focused on measures to control total exposure to allergens, through a combination of engineering, procedural, and personal controls. The cost of measures to reduce allergen exposure, such as robots and sophisticated ventilation, may be prohibitive, while their ability to reduce the incidence of allergy is uncertain. No studies of the cost effectiveness of allergen-control programs have been reported. Although a high incidence and prevalence of LAA have been reported, where comprehensive measures have been introduced to reduce personal exposure to allergens, this has led to a decrease in the incidence of allergy to low levels (Botham et al. 1995; Fisher et al. 1998). When this has happened, there is a low incidence and high prevalence, which is typical of a stable workforce (Pacheco 2007). These reports suggest that for the majority of workers exposed to animals, the development of allergy can be prevented by effective control of allergen exposure.
Epidemiology and Pathogenesis Prevalence: How Many People Have LAA? Many cross-sectional studies have investigated the proportion of cases of work-related allergy in laboratory animal-exposed populations. Pooled data from the studies of 4,988 persons at risk have indicated a prevalence of LAA of about 20%, although the spread of reported prevalence was broad (Hunskaar and Fosse 1990). Many factors contribute to the wide variation in prevalence, including differences in investigators’ methods and definitions of allergy, the tendency to retain unaffected workers in the exposed population, employment policies for workers already allergic to laboratory animals, and the natur e and degree of allergen exposure (Seward 1999).
84550.indb 117
11/11/10 6:24:16 PM
118
Handbook of Laboratory Animal Science
Cumulative Proportion of Cases %
100 80 Skin prick test Skin Any Eye/Nose Chest
60 40 20 0
0
6
12 18 24 36 Time since First Employment (months)
84
Figure 7.1â•…Latency of new, work-related symptoms or development of sensitization to rat urinary allergens. (Cullinan, P. et al. 1999. European Respiratory Journal 13:1139–1143. Reproduced with permission.)
Incidence: The Risk of Getting LAA There have been few studies of the number of newly developing cases of LAA in exposed populations, but in those published, incidences of LAA among subjects in the first years of animal work vary from 5 to 40% when no special prevention strategies have been employed (Seward 1999; Renström et al. 1994; Gautrin et al. 2000). When comprehensive prevention programs have been introduced, a reducing incidence of allergy has been observed (Botham et al. 1995; Fisher et al. 1998). Nose, eye, and skin symptoms have a higher incidence than chest symptoms (Cullinan et al. 1999). In a study of 373 students of animal health technology, the incidence of probable occupational asthma was 2.7% after 8–44 months of follow-up (Gautrin et al. 2001). This incidence is similar to that reported elsewhere (Cullinan et al. 1999; Kibby, Powell, and Cromer 1989; Fuortes et al. 1997). Most workers who develop LAA do so within the first 3 years of exposure (see Figure€ 7.1) (Hunskaar and Fosse 1990; Aoyama et al. 1992). The mean exposure before symptoms develop is shortest for nasal symptoms and longest for chest symptoms (Cullinan et al. 1994). However, there is great variation in the length of the period before allergy is clinically present. While some workers develop symptoms almost immediately, others may have contact with the animals for 15–20 years before they react (Hunskaar and Fosse 1990). However, the time to development of symptoms appears to depend upon atopic status, as was demonstrated in a retrospective study in which atopics developed LAA after a median time of 2.2 years and nonatopics after 8.2 years (Kruize et al. 1997). Exposure–Response Relationship Exposure as a risk factor for LAA has been studied; however, only recently have aeroallergen measurements been used in cohorts of exposed subjects. Exposure proxies, such as job title, work years, or hours of animal work per week, have been used in many studies, with partially contradictory results. For instance, some studies have found an inverse relationship between exposure intensity or length of employment and LAA (Kibby et al. 1989; Venables et al. 1988). This has usually been attributed to healthy worker selection: Those in the highest exposure groups who develop
84550.indb 118
11/11/10 6:24:18 PM
Laboratory Animal Allergies and Zoonoses
119
LAA have left the workforce. This may, however, suggest that length of exposure (i.e., cumulative exposure) is less important than intensity. Exposure assessments that are solely based on working day averages underestimate the peaks of exposure that are significant for symptoms and disease (Pacheco 2007). As methods to measure aeroallergen exposure have been developed, studies have begun to explain the relationship of measured allergen exposure with development of allergy, symptoms of allergy, and biological evidence of sensitization (Hollander, Heederik, et al. 1997; Heederik et al. 1999; Cullinan et al. 1999). The prevalence of LAA has been shown to be associated with length and intensity of exposure to aeroallergens (Hollander, Heederik, et al. 1997; Heederik et al. 1999). Similar findings were reported in the one study that has examined the relationship between exposure and incidence of LAA (Cullinan et al. 1999). Although an exposure–response relationship has now been demonstrated for animal allergens, the nature of this relationship is not fully understood. The total dose of allergens accumulated over a period of animal exposure is a function of exposure intensity and exposure duration. In the case of LAA, it has been suggested that intensity and duration are not equivalent. It appears that the clearest exposure–response relationship is with intensity (Nieuwenhuijsen et al. 2003). The risk of developing occupational allergy is not a simple exposure relationship. It is more complex and also depends on genetic predisposition and other exposures in the working environment, such as endotoxin (Jones 2008). At high exposure, there appears to be at least an exposure–response plateau and, possibly, reduction in risk. In part, this is explained by a healthy worker effect, but other biological factors are relevant. Some individuals may develop immunological tolerance during high or constant exposure (Platts-Mills et al. 2001). Recent research has shown that high exposure to rats is associated with lower rates of specific IgE and symptoms but an increased frequency of high specific IgG and IgG4 production. An interaction between IgE and IgG4 may result in a reduction in risk of IgE-associated, work-related respiratory disease (Jeal et al. 2006). However, the absence of identifiable IgE sensitization to animal allergens does not mean that workers are free from airway symptoms. Exposure to endotoxins may cause bronchoconstriction (Rylander 2006). Endotoxins have been measured in animal facilities, but the relative importance for airway symptoms is not clear (Lieutier-Colas et al. 2002); in one study, nonsensitized mouse workers’ airway symptoms were associated with higher endotoxin levels (Pacheco et al. 2003). Apart from inhalation, it is not known how other mechanisms of exposure (such as mucocutaneous contact) are involved in the development of LAA. Skin contact is certainly relevant in the development of skin symptoms, such as contact urticaria (Agrup and Sjöstedt 1985), and may have a role in respiratory tract symptoms. Respiratory sensitization following skin exposure to chemicals has been described (Kimber 1996), and percutaneous exposure to animal allergens has been associated with acute (anaphylactic) respiratory symptoms (Teasdale et al. 1993; Hesford, Platts-Mills, and Edlich 1995; Watt and McSharry 1996). These exposure–response observations have important implications for strategies to control allergen risks. Studies suggest that reducing total exposure to aeroallergens may prevent allergy (Hollander, Heederik, et al. 1997; Heederik et al. 1999; Schweitzer et al. 2003; Cullinan et al. 1999). However, reductions in aeroallergen exposure alone are likely to be inadequate. Measures should also be taken to reduce the number and intensity of peaks of exposure and the possibility of exposure by and through skin contact and by ingestion. Exposure to other potentially hazardous factors in the working environment should also be considered when prevention strategies are developed. Animal Allergens Several sources of allergens have been found in rodents, such as urine, fur, saliva, skin, and serum (Hunskaar and Fosse 1990). However, about 90% of those who are allergic to rats or mice
84550.indb 119
11/11/10 6:24:18 PM
120
Handbook of Laboratory Animal Science
react to the closely related allergens Rat n 1.01 (alpha2u-globulin) and Rat n 1.02 (prealbumin) from rats (Bayard, Holmquist, and Vesterberg 1996) and Mus m 1 (or the MUP complex) from mice (Lorusso, Moffat, and Ohman 1986). Sensitization to one of the species is often associated with sensitization to the other (Renström et al. 2001). Thus, developing an allergy against, for instance, rats increases the risk of soon becoming allergic also to mice. The proteins are mainly excreted in urine, but minor amounts are also excreted in saliva and by perianal and other glands (Mancini et al. 1989). The excretion depends on sex, age, and diet, and postpubertal male urinary levels may be more than 100-fold higher than in mature females or prepubertal animals (Lorusso et al. 1986; Vandoren et al. 1983). These allergens are transport pheromones, which are important for sexual communication and will influence other males, pregnant and nonpregnant females, and prepubertal animals in a range of ways (Keverne 1998). The allergens belong to a rapidly growing group of molecules called lipocalins. Many other major fur-animal allergens, such as Equ c 1 from horses and Can f 1 from dogs, also belong to the lipocalin superfamily (Virtanen, Zeiler, and Mäntyjärvi 1999). The molecules are structurally similar and have the capacity to bind or transport hydrophobic molecules within a barrel-shaped pocket. Another allergen from rodents is, for instance, rat albumin (68 kD), to which about 30% of rat-allergic subjects react (Wahn, Peters, and Siraganian 1980; Gordon, Tee, and Newman 1993). Allergens of 10–40 kD in urine, saliva, fur, and dander from guinea pigs and rabbits also cause LAA symptoms (Price and Longbottom 1988; Walls, Newman, and Longbottom 1985; Swanson et al. 1984). Pathogenesis Subjects who are hypersensitive to animals usually have an immediate type, IgE-mediated allergic reaction. This reaction is preceded by a sensitization process. Upon exposure to allergens, antigen-presenting cells internalize the allergens. The molecules are processed, and pieces of the antigens are presented to T-helper lymphocytes. The production of cytokines, especially interleukin-4 (IL-4), will induce B-lymphocytes to proliferate and produce immunoglobulin E (IgE) antibodies with specificity against these allergen epitopes. IgE antibodies are released and bound to IgE receptors on the surfaces of certain white blood cells, notably mast cells. Next, mast-cell contact with the relevant allergens initiates a series of events. The cross-linking of IgE receptors on the mast-cell surface through binding of the allergen causes degranulation: The numerous granules within the mast cell rapidly eject their contents. The mast-cell granules contain several preformed molecules, such as histamine, tryptase, and cytokines, and new synthesis of, for example, leukotrienes and prostaglandins is stimulated. Histamine and other released factors have immediate effects on local blood vessels, mucous membranes, and airway smooth-muscle tissue, causing leakage and swelling, bronchoconstriction, etc. In highly sensitive subjects, the allergic reaction can cause rapid systemic effects (anaphylactic shock). The released molecules also mobilize other types of white blood cells. A late-phase response may follow, peaking at 4–8 hours after exposure, and may involve inflammatory cells, such as eosinophils, attracted to the target tissues. Immune cells are activated, which potentiates and prolongs the allergic inflammation. Clinical Manifestations Symptoms Symptoms during work with laboratory animals are mainly, but not exclusively, mediated by mechanisms initiated by the interaction of allergens with specific IgE. Symptoms of LAA usually arise within 10 minutes of contact with laboratory animals (Lutsky and Neuman 1975), from the
84550.indb 120
11/11/10 6:24:18 PM
Laboratory Animal Allergies and Zoonoses
121
release of biochemical mediators that lead to an inflammatory response. The nature of the symptoms varies from person to person, but forms three principal patterns that affect the nose and eyes, chest, and skin. The most commonly reported symptoms affect the nose and eyes (rhinitis and conjunctivitis). The symptoms include sneezing, nasal congestion and discharge, redness of the conjunctiva, and itching, watery eyes. If the lower airways are affected, then the presenting symptoms are those of asthma, with cough, wheezing, production of sputum, and shortness of breath. Skin reactions include urticaria, with itching and circumscribed red lesions, and contact dermatitis (Agrup and Sjöstedt 1985). Typically, localized lesions appear quickly on the skin exposed to an allergen, such as the face, neck, and arms, but skin reactions can be more generalized (Rudzki, Rebandel, and Rogozinski 1981). Anaphylaxis was defined by the World Allergy Association in 2004 as a “severe, life-threatening generalized or systemic hypersensitivity reaction” (Johansson et al. 2004). In LAA, anaphylactic reactions are rare, but have been reported in association with both rat and mouse bites (Teasdale et al. 1993; Hesford et al. 1995) and a puncture wound from a needle used on a rabbit (Watt and McSharry 1996). The reactions are caused by the release of mediators from mast cells and can lead to generalized itching and urticaria; swelling of the face, lips, and tongue (angioedema); obstruction of the airways due to contraction of smooth muscle in the bronchi and increased mucus secretion; and shock (low blood pressure). Pooled data from 13 studies revealed a consistent picture of symptom distribution (Hunskaar and Fosse 1990). Of 10 persons with symptoms of LAA, about eight will have rhinoconjunctivitis (range 53–100%), about four will have skin reactions (13–70%), and about three or four will have asthma (13–71%). Subsequent studies of symptom incidence suggest this 2:1:1 ratio of symptoms remains typical (Cullinan et al. 1999; Aoyama et al. 1992). There is inevitably overlap between symptoms; most subjects have more than one affected target organ, and asthma rarely occurs in the absence of the prior development of rhinoconjunctivitis (see Figure€7.2) (Aoyama et al. 1992; Fuortes et al. 1996). A factor to consider in this context is the effect on quality of life of acquiring an occupational laboratory animal allergy or asthma. Having work-related allergic symptoms has a negative impact on several aspects of quality of life as assessed by SF-36, as does having asthma, whereas neither atopy nor sensitization to laboratory animals influenced quality of life (Renström, Blidberg, and Larsson 2007). Furthermore, occupational asthma sufferers may also be financially disadvantaged (Malo et al. 1993). Nasal/Eye
39.7 11.1
19.3 11.6 Skin
13.1
1.6
3.6
Respiratory
Figure 7.2â•…Combinations of symptoms in LAA subjects (%). (Aoyama, K. et al. 1992. British Journal of Industrial Medicine 49:41–47. Reproduced with permission.)
84550.indb 121
11/11/10 6:24:19 PM
122
Handbook of Laboratory Animal Science
Diagnosis Strong evidence for LAA is provided by a clear history of symptoms that are associated with work. When there are immediate symptoms on contact with the relevant animal, the diagnosis is usually self-evident. However, care must be taken to make sure that it is LAA rather than symptoms arising from contact with another allergen found in the animal facility, such as latex from gloves or wood dust from bedding materials. Workers who have only delayed reactions to an allergen (a feature not uncommon in asthmatics) may only experience symptoms several hours after leaving the animal facility. They may not attribute these symptoms to allergen exposure. When symptoms are those of LAA, a careful history will normally reveal a temporal association with workplace exposure. Symptoms are typically worse at the end of the working week and after periods of intense animal handling and usually improve during periods away from work, such as weekends and vacations. It is common for persons who care for animals at work to keep pets at home; this should be taken into account when considering the pattern of symptoms during “exposure-free” periods. Health workers will easily recognize the symptoms of allergy, such as the typical rashes, although there may be few findings of allergy on examination in the absence of obvious symptoms. Immunological tests are valuable because they provide additional confirmatory evidence of the diagnosis. The most widely used tests are skin-prick tests and immunoassay tests for specific IgE in the serum. In skin-prick testing, an extract of allergen is placed on the skin and the underlying skin is then punctured. The skin reaction is then compared with a positive (histamine) and a negative control. About half of workers who experience airway or skin symptoms when working with animals (and most of the asthmatic individuals) will be positive by immunoassay. In cases of asthma with no specific IgE to laboratory animals, the symptoms may be due to reactions to other agents present in the working environment (e.g., dust, ammonia, formaldehyde, or disinfectants) (Das et al. 1992). Suspected occupational asthma is usually confirmed with pulmonary function tests. Animal workers with suspected asthma are invited to record their peak expiratory flow rate every 2 hours throughout the day. They should continue recording this for at least 4 weeks and include a period away from working with animals. A difference between the mean peak flow while at work and away from work of greater than 15% is indicative of occupational asthma. An alternative technique is cross-shift spirometry, in which pulmonary function is measured before and after work. A deterioration during the working day is suggestive of an occupational cause. However, this technique is often not practical and the results may be confounded by the normal diurnal increase in pulmonary function from morning to afternoon. The results may also be confounded in asthmatics with a late-phase bronchoconstriction, which may appear several hours after getting off work. Rarely, pulmonary-challenge testing may be necessary to confirm the diagnosis, but this is a risky technique that is only performed in designated specialist centers. Prognosis The symptoms of rhinoconjunctivitis and urticaria are a nuisance. Unless they are effectively managed, they may make it difficult for the affected persons to continue to work with the animals to which they are allergic. Most workers allergic to laboratory animal seem to wish to continue working with animals (Renström et al. 2001). If the affected persons cease to be significantly exposed to the allergen, then symptoms may be greatly diminished or disappear. However, subjects who have developed sensitivity to one type of animal fur are at increased risk to develop allergies to others even when control measures are good (Perfetti et al. 1998; Goodno and Stave 2002).
84550.indb 122
11/11/10 6:24:19 PM
Laboratory Animal Allergies and Zoonoses
123
An ongoing follow-up study of secondary prevention among workers with animal-workrelated symptoms indicates that it may be possible to decrease symptom prevalence and frequency significantly and reduce airway inflammation through prevention measures (Renström pers. comm.). Continued exposure to allergens may lead to the insidious development of asthma, which is of greater concern. Sensitized workers are much more likely to go on to develop chest symptoms compared to nonsensitized workers (Nieuwenhuijsen et al. 2003; Elliott et al. 2005b). There is emerging evidence that female workers may be at greater risk of developing asthma (Elliott et al. 2005b), although women may be more inclined to contact occupational health services. According to a Swedish questionnaire (n = 671), although workers with symptoms indicative of LAA have a right to a free medical examination, only about 15% of workers who had experienced animal-work-related symptoms had contacted a doctor. There was a significantly higher proportion among women than men, despite a similar prevalence of reported symptoms (Renström pers. comm.). Lung function decline was most pronounced in sensitized subjects who continued to be in contact with the animals to which they were sensitized (Portengen et al. 2003). The reduction in pulmonary function may be persistent, even after exposure ends (Venables 1997), with substantial impact on the LAA sufferer’s quality of life (Malo et al. 1993). Continued exposure may lead to permanent loss of function. Case Management The workplace management of LAA should be focused on the reduction of exposure to a clinically insignificant level. The medical management of LAA is generally focused on the relief of symptoms and is not curative. Successful treatment of symptoms should not be allowed to confound efforts to reduce allergen exposure. The exposure can only be regarded as sufficiently low when the LAA sufferer is free of symptoms in the absence of treatment. The degree of exposure precipitating symptoms in allergic workers varies greatly between individuals (Anon. 1980; Reeb-Whitaker et al. 1999). Employers and employees should be aware that a risk of anaphylaxis exists and that this may be life threatening. Arrangements for the immediate treatment of anaphylactic reactions should be in place in all animal facilities. Although all workers remain at risk of anaphylaxis, the likelihood of a reaction occurring is very low. Any person who develops LAA should be counseled by a knowledgeable physician. The worker’s job should be analyzed; control measures should be reviewed and, if necessary, enhanced so that exposure is reduced (to the benefit of all exposed workers). The capacity of the individual to reduce the risk, through good personal hygiene and careful use of personal protective equipment and by performing tasks in less exposure-generating ways, should be reemphasized. It may be possible to relocate the worker to a lower risk area, such as work with isolators or in an area with lower aeroallergen levels (Thulin et al. 2002), or to transfer the worker to an area where a different species is present. However, the risk for a rodent-sensitive worker to develop allergies against other rodents is considerable (Renström et al. 2001; Goodno and Stave 2002), and if symptoms of LAA persist, the fitness of the worker to continue any animal work should be reviewed carefully. This is especially important when there is any evidence of lung function impairment; in this case, continuing to work with the animals to which workers are sensitized is unadvisable (Portengen et al. 2003). Redeployment into a different role where key skills can be utilized, such as quality assurance or training, may be necessary. The necessary reduction of exposure to a clinically insignificant level can usually be achieved by changes in working practices or redeployment away from the relevant animal. Only in the most extreme cases will the affected persons have to leave their employment.
84550.indb 123
11/11/10 6:24:19 PM
124
Handbook of Laboratory Animal Science
Control Strategy Allergy is costly to employers and employees, and controlling the risks that animal allergens pose is desirable from both a moral and an economic perspective. In most jurisdictions, it is a legal duty. Unfortunately, because the vulnerable people cannot be identified with certainty in advance and the hazard (allergen exposure) cannot be completely eliminated under normal circumstances, there is inevitably residual risk to be managed. Ideally, the risk posed by allergens should be considered in a comprehensive management system that addresses all the safety, health, and environmental risks in the animal facility. The components of this management system should include setting policy, establishing an appropriate organization, making and implementing plans, and measuring and reviewing performance. An effective system will be dependent on the leadership and commitment of senior management and the cooperative involvement of people at all levels of the organization. A control strategy should be introduced prior to the construction of the animal facility so that it can be implemented in the design and influence key purchasing decisions (Thulin et al. 2002). In many cases, this is not possible because the facility is already established; however, the strategy should still influence future refurbishment decisions. The control strategy should be dynamic and responsive to changes in technology and understanding. Even when the initial design has been strongly influenced by the allergen-control strategy, as this strategy changes during the lifetime of the facility, there may be implications for allergen control. The principal mechanism by which an allergen initiates LAA is assumed to be inhalation. Control measures should be mainly, but not exclusively, aimed at the control of aeroallergen in the worker’s breathing zone. Ambient levels of allergens may not be representative of personal exposure. In a microscopy room, for example, ambient levels of allergens could be low, but, because the microscopist works with a source of allergens close to the breathing zone, personal exposure may be significantly greater. Controls should reduce both the intensity and the duration of exposure. Several studies have described the intensity of exposure associated with different tasks (Nieuwenhuijsen et al. 1995; Eggleston et al. 1989; Hollander et al. 1998; Renström and Kallinn 2000). Directly handling animals (especially during close-up, detailed work) and cleaning and changing dirty cages are associated with exposure to high concentrations, whereas work on animal tissues postmortem is associated with lower exposures (see Table€7.1). A simple hierarchy for risk management can be applied. Controls based on engineering solutions are preferable to those based on procedures or people because they are less reliant on human factors. An effective strategy will be based on all three (NRC 1997). It should take account of controls during normal operations, controls for operations when conditions are not normal (e.g., spillage or breakdown), and controls for exposures that may occur beyond the controlled areas (fugitive exposures) (NRC 1997). Engineering Controls Animal facilities should be designed to incorporate engineering controls to the extent feasible. The most likely limitations to the introduction of engineering controls are the constraints imposed by the existing facility and the need for significant capital investment. In existing facilities, the costs of retrofitting may be prohibitive, not least because to do so may mean business operations have to stop temporarily. One of the problems associated with evaluating engineering controls is that there is relatively little evidence that specific building (ventilation or architect design) systems that may contribute to reduced total exposure to allergens actually help prevent allergy. For instance, a general assumption
84550.indb 124
11/11/10 6:24:19 PM
Laboratory Animal Allergies and Zoonoses
125
Table€7.1╅The Likelihood of Exposure in the Absence of Specific Control Measuresa Exposure Low
Medium
High
Task Postmortem and surgery Slide preparation Laboratory work (low number of animals) Automated cage cleaning Cleaning Indirect contact in animal room Feeding Taking specimens Injections Handling animals Changing and cleaning cages Changing filters Washing cages Preventing Asthma in Animal Handlersb
Animal Handlers Should: Perform animal manipulations within ventilated hoods or safety cabinets when possible Avoid wearing street clothes while working with animals Leave work clothes at the workplace to avoid potential exposure problems for family members Keep cages and animal areas clean Reduce skin contact with animal products, such as dander, serum, and urine, by using gloves, lab coats, and approved particulate respirators with face shields Employers of Animal Handlers Should: Modify ventilation and filtration systems: Increase the ventilation rate and humidity in the animal-housing areas. Ventilate animal-housing and -handling areas separately from the rest of the facility. Direct air flow away from workers and toward the backs of the animal cages. Install ventilated animal cage racks or filter-top animal cages. Decrease animal density (number of animals per cubic meter of room volume). Keep cages and animal areas clean. Use absorbent pads for bedding. If these are not available, use corncob bedding instead of sawdust bedding. Use an animal species or sex that is known to be less allergenic than others. Provide protective equipment for animal handlers: gloves, lab coats, and approved particulate respirators with face shields. Provide training to educate workers about animal allergies and steps for risk reduction. Provide health monitoring and appropriate counseling and medical follow-up for workers who have become sensitized or have developed allergy symptoms. Principal Elements of an Occupational Health and Safety Programc Administrative procedures Facility design and operations Exposure/control methods Education and training Occupational health services (continued)
84550.indb 125
11/11/10 6:24:20 PM
126
Handbook of Laboratory Animal Science
Table€7.1╅The Likelihood of Exposure in the Absence of Specific Control Measuresa (continued) Exposure
Task
Principal elements of an occupational health and safety programc Equipment performance testing Information management networks Emergency procedures Program evaluation and audit a
b c
Data from Nieuwenhuijsen, M. J. et al. 1995. Annals of Occupational Hygiene 39:819–825; Hollander, A. et al. 1998. Scandinavian Journal of Work, Environment & Health 24:228–235; Renström, A., and C. Kallinn. 2000. Characterization of the work environment in an animal laboratory facility, Stockholm: NIWL (unpublished report). National Institute for Occupational Safety and Health. 1998. Publication no. 97–116. NRC. 1997. Occupational health and safety in the care and use of research animals. Washington, D.C.: National Academy Press.
is that ventilation design contributes to reductions in particle counts and thereby leads to less allergy. However, there is limited published evidence regarding the biological significance of the various forms of technology. Separation Concerning allergen spread within facilities, the first consideration should be separation of the potential population at risk from the hazard. The hazard is not just the animals, but, more important, the allergens they produce. At the facility level, this can be interpreted as construction of the animal facility away from workers who are not exposed to animals. Within the facility, this can be achieved by clear segregation of work with animals from other work, such as administration. Boundaries can be established and, when necessary, access controls introduced to prevent exposure of people who are not directly involved in animal work or support of the animal areas. Even within the areas in which animal work is directly carried out, it may be possible to widely separate the majority of workers from the areas where the potential exposure is highest (e.g., the animal-holding rooms) or use barriers such as curtains in front of cages (Krohn et al. 2006). Separation can be facilitated by a two-corridor system when this is feasible—one corridor used for “clean” and the other for “dirty” activities. Separation should also be considered in the specific context of the allergens. When allergen exposure is foreseeable despite the absence of animals, this should be controlled. Key areas include places where cage waste is handled, in the laundry, and at the exhaust points from ventilation systems. Ventilation inlets, especially those directed to clean and nonanimal areas, should not be placed in proximity to or downwind from outlets from contaminated areas. General Ventilation Facility ventilation, including the control of temperature and humidity, contributes to the general control of allergens. General ventilation has an important influence on the microenvironment in the animal cage, and it is this factor, rather than allergen control, that has usually been more influential in the development of ventilation systems. Task-specific local exhaust ventilation, rather than general ventilation, is the principal control method because it is more effective, less costly, and probably easier to implement. Although an increase in air-change frequency may reduce allergen levels, this is not a consistent finding (Schweitzer et al. 2003; Malo et al. 1993). Many different approaches to the general ventilation of animal facilities have been shown to be effective, and not all are dependent on expensive high-frequency air changes (Harrison 2001). One-way airflow systems with sliding perforated
84550.indb 126
11/11/10 6:24:20 PM
Laboratory Animal Allergies and Zoonoses
127
screens, with cage racks and exhaust vents behind them, have been shown to draw allergens behind the screens effectively, leaving minimal allergen levels in the room (Lindqvist et al. 1996). Pressure gradients are an important adjunct to the control of allergen spread; these are a common feature of animal facilities. However, there are potential conflicts between the gradients required to protect the animals’ health and those required to protect human health. In general, it is desirable to use a gradient that minimizes spread of allergens (and pathogens) from cage-cleaning areas into “clean areas,” such as offices and restrooms. Negative pressure “sinks” adjacent to animal-holding rooms can also be used. Indoor pressure balance can be maintained by making sure that the supply and exhaust systems are equal in capacity. Increasing the relative humidity has been shown to reduce the levels of airborne allergens (Edwards, Beeson, and Dewdney 1983; Jones et al. 1995). Presumably, in conditions of higher humidity, particles weigh more, are more adhesive, and will settle more readily. High humidity, however, is more uncomfortable for workers and increases growth of molds and mites. Exhaust air from animal areas will be contaminated with allergens (and possibly pathogenic organisms). In some circumstances, it will be necessary or desirable to filter exhaust air. Exhaust air should not be recirculated without filtration. Exhaust air and filters are important sources of fugitive exposure. Controls should be in place to prevent exposure to exhaust air and to minimize the risk to people involved in the maintenance of ventilation systems and changing of filters. Task Ventilation Task ventilation, or local exhaust ventilation, is one of the most important control measures. These systems remove allergens at the source and can be designed (usually at relatively low cost) to accommodate the tasks with potential for the highest exposure. They contribute to reductions in the spread of allergens and other contaminants. Task ventilation includes biosafety cabinets, fume cupboards, and ventilated workstations that use downdraft or back-draft systems (Skoke 1995). Often, these systems have the advantages of being mobile and suitable for installation in established facilities. However, it may be difficult to demonstrate the effectiveness of these systems, especially novel designs, such as downdraft benches that are reliant upon undisturbed laminar flow. It is easier to demonstrate that novel ventilation systems function effectively when not in use than when they are used by operators under normal work conditions. For instance, covering too great a proportion of the ventilated surface of a downdraft table is likely to reduce the effectiveness of the exhaust system. If the effectiveness of the ventilation system cannot be confirmed under operational conditions, then it should not be relied upon as a primary control measure. Class I cabinets may not eliminate exposure during procedures that generate high levels of aeroallergens (Gordon et al. 2001). Automation New technology is enabling the automation of many tasks. This benefit of automation is especially interesting when the tasks are labor intensive and pose significant risk. These risks may be high allergen levels or other factors, such as exposure to potentially harmful pathogens or test substances or ergonomical risks. For instance, when cages and bottles are being cleaned, the risks will be due to both allergens and ergonomics. Automated cage cleaning and waste handling systems have now been introduced in some animal facilities (Harrison 2001). Automated cage cleaning systems have been shown to reduce ambient levels of allergens and personal exposure of operators under normal operating conditions greatly (Renström and Kallinn 2000). Automated cage cleaning systems can make a substantial difference
84550.indb 127
11/11/10 6:24:20 PM
128
Handbook of Laboratory Animal Science
Rat Urinary Aeroallergen (µg/m3)
3
2 1.43 1 0.33 0 Filter top (n = 30)
Open top (n = 12)
Figure 7.3â•…Comparison of cage design. Rat urinary aeroallergen concentrations measured when 30 rats were housed on woodchip bedding in filter-top and open-top cages, excluding those measurements made on cleaning-out days for the open-top cages. The geometric mean (GM) is indicated. (Gordon, S. et al. 1992. British Journal of Industrial Medicine 49:416–422. Reproduced with permission.)
to control strategies; this is otherwise a particularly hazardous task that can generate high levels of airborne allergens (Table€7.1). Cage Systems The introduction of filters to conventional open-top cages is associated with substantial reductions of greater than 75% in allergen concentrations (see Figure€7.3) (Gordon et al. 1992; Reeb-Whitaker et al. 1999; Hollander et al. 1998; Sakaguchi et al. 1990; Platts-Mills et al. 2005). Individually ventilated cage systems are now widely available (Lipman 1999), and these have been shown in a number of studies to reduce background aeroallergen levels effectively (Reeb-Whitaker et al. 1999; Gordon et al. 2001; Platts-Mills et al. 2005; Clough et al. 1995; Gordon et al. 1997). The most impressive reductions in aeroallergen levels in undisturbed animal rooms—almost 100%—arise when the system is operated with the cages under negative pressure (Schweitzer et al. 2003; Reeb-Whitaker et al. 1999; Gordon et al. 1997; Renström, Höglund, and Björing 2001). The effectiveness is further enhanced by filtered exhaust ports (Platts-Mills et al. 2005). Using individually ventilated cages at positive pressure does not confer the same reductions in aeroallergens, unless the system is sealed (Gordon et al. 2001). Caging systems operated at negative pressure may also inhibit the transmission of infectious diseases (Myers et al. 2003). Procedural Controls The objective of engineering controls is to minimize the influence of human factors. Because the risk of allergy cannot be eliminated by completely removing the source of the allergen hazard or by engineering solutions, additional controls are essential. The emphasis of control procedure techniques is to manage work choices so that essential job duties are performed in a way that minimizes the levels and spread of environmental allergens.
84550.indb 128
11/11/10 6:24:20 PM
Laboratory Animal Allergies and Zoonoses
129
Reduction of the Number of Exposed Persons A first strategy of prevention is to minimize the number of animal-exposed personnel. This can be accomplished by reducing the use of animals, by increasing the use of alternative methods (such as using cell lines in the production of antibodies), or by using toxicity tests that do not involve animals. Another option is to concentrate animal work so that fewer individuals are exposed. For example, the installation of a labor-saving device, such as a cage wash robot, simultaneously reduces ergonomics problems and allergen exposure and allows technicians more time to increase their involvement in other work activities. These workers, who have excellent animal-handling competence that minimizes exposure to allergens, can then perform work procedures that would otherwise expose less experienced postgraduate students. Animals, Stock Density, and Bedding Mature male animals have been shown to generate higher concentrations of allergens in urine (Vandoren et al. 1983; Gordon et al. 1993) and in animal rooms (Sakaguchi et al. 1990). A recent study suggests that working with male animals may increase the risk for LAA (Renström et al. 2001). If it is feasible, considering the scientific question at hand, substitution with younger or female animals may reduce aeroallergen levels and, possibly, LAA. Several studies have shown an association between stock density and allergen levels (see Figure€7.4) (Gordon et al. 1992; Eggleston et al. 1989; Swanson et al. 1990; Edwards et al. 1983). The usefulness of this information is slight, as density is likely to be dictated by business factors. 14
A
Rat Urinary Aeroallergen (µg/m3)
12 10 8.09
8 6 4
3.59
2
1.84 0.85
0.12
0 60
45
30 15 Number of Rats
0
Figure 7.4â•…Effect of reducing stock density. Rat urinary aeroallergen concentrations measured when the stock density was reduced. The measurements made on cleaning-out days are shown as open circles. The geometric mean (GM) is indicated. (Gordon, S. et al. 1992. British Journal of Industrial Medicine 49:416–422. Reproduced with permission.)
84550.indb 129
11/11/10 6:24:21 PM
130
Handbook of Laboratory Animal Science
18
Rat Urinary Aeroallergen (µg/m3)
16 14 12 10 8 6
7.79 6.16
4 2.47
2 1
Woodchip (n = 40)
Sawdust (n = 36)
Absorbent pads (n = 28)
Figure 7.5â•…Comparison of litter type. Rat urinary aeroallergen concentrations measured when 60 rats were housed on woodchip and sawdust in contact litter, and absorbent pad, noncontact litter. The measÂ� urements made on cleaning-out days are shown as open circles. The geometric mean (GM) is indicated. (Gordon, S. et al. 1992. British Journal of Industrial Medicine 49:416–422. Reproduced with permission.)
The advent of individually ventilated cage systems now means that it is feasible to increase stock density while maintaining control of ambient levels of aeroallergen (Gordon et al. 1997). However, the potentially allergen- and pathogen-releasing cage changing task still requires practical solutions to minimize contamination. Bedding has an important influence on allergen concentrations (Kaliste et al. 2004). Absorbent pads are associated with lower allergen levels than wood chips or sawdust (see Figure€7.5) (Gordon et al. 1992). Wood chips give lower aeroallergen concentrations than sawdust (Platts-Mills et al. 1986). Crushed corncob was found to give lower levels than wood shavings (Sakaguchi et al. 1990). The impact of animal cage enhancements on allergen exposure has not been reported; however, preliminary findings suggest that enrichment measures may increase allergen levels during cage changing (Renström unpublished data). Housekeeping Animal facilities should be designed so that they can be effectively and safely cleaned. Examples of this are the use of closed vacuum cleaners that deliver dust into a closed-pipe conveyor with deposition into a sealed container, the use of moist mopping, or damp sweeping. Dry-cleaning procedures, such as brushing and the use of portable vacuum cleaners, should be avoided because they can create high amounts of airborne particles contaminated with allergens. The use of power-washing systems, using high-pressure water, can generate contaminated aerosols and should be avoided when possible. When power washing is used, appropriate personal protective equipment should be worn. The contaminated outputs from animal work must be controlled. There should be procedures to control exposure to allergens in the handling of contaminated documents and in the disposal and
84550.indb 130
11/11/10 6:24:22 PM
Laboratory Animal Allergies and Zoonoses
131
handling of animal carcasses and tissues, animal waste and bedding, and contaminated personal protective equipment. Measures should be implemented to minimize the need for documents in animal holding and handling areas, such as using computers with washable plastic-covered keyboards for documentation, connected to printers outside the animal department. Clean documents can be created in a scanning (but not photocopying) process. If documents need to be retained (and archived), then measures should be taken to minimize spread of allergens from the facility. Contaminated records and archives are an important potential source of allergen exposure because they provide a mechanism by which workers with serious LAA can be inadvertently exposed. Procedures should be implemented to reduce exposure to allergens during the laundry of reusable protective clothing. The risk to laundry workers, who may not be aware of animal allergen risks, should be considered. Soluble laundry bags that can be sealed in the animal facility and dissolve during the laundering process are available. Cage cleaning has the potential to expose workers to high aeroallergen levels (Nieuwenhuijsen et al. 1995; Lieutier-Colas et al. 2001). Care must be taken to reduce allergen spread from this source at all points of handling. Dedicated equipment that minimizes contact with soiled bedding should be provided in the cage-washing area. Several commercially available systems have been described. Common to all of these is that cages are emptied into closed transport systems that deliver bedding to different types of sealed containers. Vacuum cleaning systems that are designed not to generate local contaminated exhaust have been used and result in low levels of allergens during floor cleaning (Thulin et al. 2002). However, these measures will not necessarily contribute to reductions in personal exposure if they do not enable efficient completion of the task (Gordon et al. 1997). Movements within the Facility Procedures should be designed to prevent the spread of allergens into the environment and to adjacent areas, such as corridors, offices, and rest areas. Transport of soiled cages from the animal room to the cleaning area should be done using a closed or covered system, preferably only in designated “dirty corridors.” Animals should be transported in suitable transport cabinets equipped with filters that prevent the spread of allergens along corridors and in elevators. Single cages with animals should be transported using a filter top. Single cages should likewise be covered with a filter top when standing freely in laboratory or procedure rooms. Work Permits and Visitors A mechanism should exist whereby unusual exposures are adequately controlled. These can involve regular animal workers carrying out irregular tasks, such as during equipment breakdown, or they can involve workers not normally exposed to animals (e.g., ventilation maintenance technicians). These tasks should be individually assessed and may need to be controlled under a workpermit procedure. Environmental Monitoring Hygiene studies have improved understanding of the exposure characteristics of different workplaces and activities. Hygiene monitoring can give confidence that controls are reducing allergen concentrations and that changes in controls do not have a negative effect. Environmental monitoring can guide animal room management routines and choice of respiratory protection (Ooms et al. 2008; Renström, Karlsson, and Tovey 2002). In some circumstances, it may be necessary to demonstrate that a change in working practice for a reason other than improved allergen control does not have a significantly adverse effect on
84550.indb 131
11/11/10 6:24:22 PM
132
Handbook of Laboratory Animal Science
A. Monday Number of Particulates × 100
100 50 Lunch
Break
B. Sunday
100 50
0900
1200
1500
1800
2100
Figure 7.6â•…Variations in particles during the day in a typical mouse room. (Kacergis, J. B. et al. 1996. American Industrial Hygiene Association Journal 57:634–640. Reproduced with permission.)
allergen levels. Unfortunately, the analysis of allergen samples is expensive and time-consuming. Analytical services are still not widely available for many animal allergens. It is relatively simple to measure concentrations of particles, and this has the advantage that it can be done in real time, giving a clearer indication of the fluctuating personal exposure (to particles) during the working day (see Figure€7.6) (Kacergis et al. 1996). However, there is no correlation between the concentrations of particles and allergens. For example, handling clean or dirty bedding may generate high concentrations of particles, but in the first case, no animal allergen is present at all. If the level of airborne particles is used to inform risk assessment and controls, then it is important to consider the extent to which the particles may be contaminated with allergens. At present, there are no internationally recognized standards for analytical methods or acceptable aeroallergen exposure levels. While a standard may seem desirable, the challenge of establishing a robust and meaningful exposure limit is great. Air sampling is variable and the levels relevant to sensitization are so low that a standard may be impractical (Platts-Mills et al. 2005). A single threshold would not be likely to protect every worker because of the importance of other factors, such as genetic predisposition, and environmental influences, such as endotoxins (Schweitzer et al. 2003).
Personal Controls In addition to engineering and procedural controls, for some tasks, it may be necessary to implement controls that are targeted at the workers themselves. Although some of these, such as choice of personal protective equipment, may be incorporated into procedures, their effectiveness is wholly dependent upon the behavior of the individual. The focus of personal control measures is on influencing behavior, and it includes the correct use of personal protective equipment and training. Personal Hygiene Eating, drinking, smoking, and the application of cosmetics should not be permitted in animal facilities. Dressing routines should be established that minimize the spread of allergens around and
84550.indb 132
11/11/10 6:24:23 PM
Laboratory Animal Allergies and Zoonoses
133
from the animal areas and the introduction of pathogens into the facility. Workers heavily exposed to allergens should shower before leaving the facility to remove allergens absorbed by exposed skin and hair. If gloves are used, care must be taken to ensure that allergens do not contaminate the skin by entry via the open-arm end of the glove or through puncture holes. Workers should wash their hands regularly and always after handling animals without gloves. However, care must be taken to prevent small skin fissures from developing through frequent washing, since these will increase the risk of systemic allergen exposure. Barrier creams and hand-friendly washing substances are widely available. Protective Clothing Most modern laboratory animal facilities require personnel to use special clothing while working in the facility. The main objective of this is to prevent microbiological contamination of the animals within the facility. Failure to use protective clothing effectively can lead to distant allergen spread (Krop et al. 2007). The type of clothing will vary according to the degree of protection required and may range from full surgical clothing with gown, cap, mask, and shoe covers to simply changing a conventional laboratory coat. The design of protective clothing is important. Coat or coverall arms should be designed so that allergens are not trapped inside the arm surface. The sleeves should have an elastic hem or other suitable mechanism. Alternatively, coverall arms should be rolled back so as to allow personnel to wash both hands and forearms. Several types of gloves are available; nonlatex gloves are preferable to latex gloves, due to the risk of developing a latex allergy (Brehler and Kutting 2001). Respiratory Protective Equipment Half-face particle filter respirators are a significant control measure used in successful programs (Botham et al. 1995; Fisher et al. 1998). Disposable respirators of type P2 have been shown to reduce inhaled allergens by 90% (Renström and Kallinn 2000). Several types of ventilated masks and helmets have been developed in which filtered air is delivered to the operator. These can be highly effective in reducing exposure to airborne allergens. Air-stream respirator helmets have been shown to relieve symptoms of LAA in sensitized workers (Slovak, Orr, and Teasdale 1985). Supplied air systems have an important place in both the prevention of allergy and the control of symptoms. They should be available for workers at high risk and for workers carrying out high-risk tasks, such as cleaning up spills of heavily contaminated bedding or maintaining cage emptying systems. Workers with LAA can continue to work in animal areas as long as they remain symptom free and if they use these devices to minimize their exposure to allergens to the point that it is negligible. Any person using a ventilated device should be trained in its use. Equipment should be cleaned and stored and filters changed without spread of allergens or cross-contamination. Although these air-supplied devices can reduce personal exposure to aeroallergens to negligible levels, they do have major drawbacks: They do not easily facilitate detailed work in close proximity to the animal when the risk of exposure is high; they may be uncomfortable if used for long periods, leading to eye irritation, headaches, and neck discomfort; and they may not be suitable when needing to communicate with colleagues or to perform tasks for socialization purposes with animals (e.g., primates). Training and Education The success of any risk-control program is dependent on the support of those at risk. Effective training and education, at the start of working with animals and regularly repeated, is
84550.indb 133
11/11/10 6:24:23 PM
134
Handbook of Laboratory Animal Science
an important feature of successful allergy prevention programs (Botham et al. 1995; Fisher et al. 1998). Workers at risk of LAA must understand the nature of the risk if they are to be able to appreciate the actions they can take to protect themselves. These actions include wearing protective equipment when handling the animals, reducing allergen spread, taking care with personal hygiene, and promptly reporting symptoms of allergy. Improved use of protective equipment has been reported following the implementation of education programs (Fisher et al. 1998). Failure to take care can lead to distant spread of allergens that put other workers and even family members at risk (Krop et al. 2007). Preplacement Assessment: Assessing Risk The prevention of allergy should begin before exposure to an allergen occurs. Assessment of potential animal workers prior to commencing work with animals is good practice, as well as a legal duty in some jurisdictions. Partly, the purpose of this assessment is to consider individual vulnerability to allergy. It is also to consider the worker’s capability to do the proposed work and determine the need for interventions and adjustments (such as immunization against infectious diseases and provision of lifting aids). If individuals who will definitely develop allergy could be identified reliably prior to exposure, then they could be excluded from the workplace. The identification of these vulnerable people is highly desirable because they could then be given advice on additional precautions. But is it feasible? Atopy is a genetic predisposition to develop specific IgE and allergic reactions (e.g., skin rashes, rhinoconjunctivitis, asthma) (Johansson et al. 2004). The majority of studies have indicated that workers who have a personal history of atopy are more likely to develop LAA, although this has not been a universal finding (Seward 1999). Some investigators have identified an association between family history of atopy and the development of LAA, but others have found no association (Seward 1999). It is likely that any association between family history and the development of LAA is weak. Some studies have examined the association between biological indicators of atopy (skin testing and immunoassay) and allergy. Most of these studies have been a cross-sectional design and have examined the association of atopy in established cases of LAA. This limits their usefulness in establishing the predictive value of these indicators in workers without LAA. There is a clear association between skin-test reactivity to animal allergens and allergic symptoms. Pooled data from seven studies have shown that 51–69% of people allergic to rats or mice have positive skin tests to allergens from these animals (Hunskaar and Fosse 1993). An association between total IgE and the later development of allergy has been reported (Renström et al. 1994). Only one study has examined the predictive value of radioallergosorbent tests (RAST) for IgE specific to the animal allergen. A combination of a positive RAST and positive skin test was 87.4% predictive of the development of LAA (Botham et al. 1995). Even using the best predictive tests (personal history and biological indicators of atopy) to exclude vulnerable workers, more people who would never develop symptoms would be excluded than people who would become allergic (Botham et al. 1995; Newill et al. 1986). While it is possible to identify asymptomatic workers at the start of employment who are at increased risk of developing LAA, it is neither practical nor ethical to implement effective screening criteria and exclude them from work. If the implementation of a comprehensive allergen-control program has reduced the incidence of allergy to low levels, the value of these tests as predictors of LAA is similarly reduced. Preplacement assessment is still worthwhile. It is the first opportunity to assess the vulnerability of the candidate and counsel him or her on the measures he or she should take to minimize the risk of developing allergy. It is an opportunity to establish baseline data and carry out baseline investigations against which future assessments can be compared. Serum banking is not recommended,
84550.indb 134
11/11/10 6:24:24 PM
Laboratory Animal Allergies and Zoonoses
135
however (NRC 1997). Some candidates will have a history of exposure to laboratory animals (either from their studies, work, or pets) and may report preexisting allergies. Many of these will be able to begin their intended occupation (with appropriate adjustments and restrictions), but some will not. If a candidate reports anaphylactic reactions or occupational asthma, then the risk of continued exposure to the relevant allergen is likely to be unacceptable. Counseling at health assessment should be an integral part of any allergy-prevention program. In addition to helping people exposed to allergens understand the potential health effects and the need for early reporting, it is an opportunity to explain the steps they can take to protect their health and to reinforce the importance of engineering, procedural, and behavioral controls. It provides an opportunity to explain individual risk in the context of the proposed work (and exposure pattern), the local experience of allergy incidence, and the individual’s tendency to develop allergy. This information allows the candidates to make an informed decision about the risks of the proposed employment. Health Surveillance Regular health surveillance of significantly exposed workers is worthwhile. It provides an opportunity to raise awareness of the potential effects of allergens on health, investigates symptoms of allergy that the worker reports, and reinforces the need to report relevant symptoms if they do develop. Annual surveillance of exposed workers is typical. The majority of workers who develop allergic symptoms do so within 2 years of first exposure (Hunskaar and Fosse 1990). During this period, more frequent surveillance may be warranted. The basis of surveillance should be a questionnaire (Bush, Wood, and Eggleston 1998). This can help identify workers at higher risk (Meijer, Grobbee, and Heederik 2004). Some centers may perform lung function tests, skin testing, and immunoassay, but there is no evidence that these have any predictive value as routine screening tests. Individuals who have developed signs of LAA or asthma who do not cease working with animals should be carefully monitored at regular intervals while major efforts are made to reduce exposure and to equip them with effective personal protection. Additional tests may have value in helping to reinforce the educational messages. Integrated LAA Risk Management It is imperative that efforts to reduce staff exposure to animal allergens, with the goal of reducing incidence of LAA, do not give rise to new health problems, either among staff or animals. If allergen levels are reduced—but at the cost of increased ergonomic strain and compromised animal welfare (potentially affecting research)—nothing has been gained. Few studies have as yet focused on these balances. Recently, cage systems for housing mice were compared; allergen levels, ergonomics features, biting frequency, weight gain, and cage climate were evaluated. It was shown that although allergen levels in undisturbed mouse rooms were low when individually ventilated cages (IVC) were used, they were less ergonomically suitable, compared to regular open-shelving or ventilated cabinets (Renström, Höglund, and Björing 2001). Also, some evidence was found that cage climate and animal welfare may vary between IVC systems (Höglund and Renström 2001). We conclude that all planned installations should be evaluated beforehand regarding both animal and worker health and the relative advantages and disadvantages of alternatives thoroughly weighed. Investments are usually costly, and the potential costs of worker relocation and compensation, as well as the loss of research animals through contamination or stress, should be taken into account.
84550.indb 135
11/11/10 6:24:24 PM
136
Handbook of Laboratory Animal Science
Zoonoses Zoonoses are diseases that can be transmitted from animals to humans. Most organisms carried by laboratory animals are harmless to both the animal and handler. Some have the potential to cause infection and these are usually relatively specific to a species of animal. Good hygiene means that reports of laboratory animal workers becoming infected with a zoonosis are rare; however, vigilance is essential. Zoonoses can be transmitted directly by contact with an animal (e.g., rabies), via a contaminated environment (e.g., anthrax), or by way of ingestion (e.g., campylobacteriosis) or indirectly via vectors, such as ticks. The organisms causing zoonoses include viruses, bacteria, fungi, and parasites including protozoa. Zoonoses in the general population are surprisingly common. Of the agents known to infect humans and of those considered to be emerging infections, 61 and 75%, respectively, are zoonotic (Taylor, Latham, and Woolhouse 2001). Some important human infections are believed to have originated in animals, but have now become so commonplace that they are no longer considered zoonotic. Notable examples of human diseases of this type include HIV from primates (Sharp, Shaw, and Hahn 2005) and the 1918 influenza pandemic from birds (Taubenberger et al. 2005). In general terms, the likelihood that an animal pathogen can also result in human disease is greater the closer the phylogenetic distance between the animal and humans is (Wolfe, Panosian Dunavon, and Diamond 2007). While the likelihood of transmission may be considered highest in primates, zoonotic infections can be occasionally transmitted from nonmammals. For example, reptiles are an important carrier of salmonella (Bertrand et al. 2008). The mere transmission of an infectious agent does not cause illness. The pathogenesis is dependent on factors such as the means of entry (skin, oral, or respiratory), the availability of suitable receptors, the integrity and effectiveness of host defenses, the access to target organs, and the ability of the agent to replicate. When the infectious agent is poorly adapted to a human host, it is normally dealt with efficiently by the immune system. If the immune system’s defenses are overcome, then even poorly adapted zoonoses can cause severe disease. Zoonoses acquired from laboratory animals are uncommon (Sewell 1995; Weigler, Di Giacomo, and Alexander 2005). There is a very low incidence of zoonoses transmission in modern laboratories (Walker and Campbell 1999). However, severe illness can arise. It has been suggested that herpes B infection has caused more deaths among laboratory workers than any other virus (Pike 1976, 1979). Prevention Close collaboration is needed between veterinarians, physicians, and animal workers to identify and manage the risk of zoonotic diseases. This includes identifying the infection hazards associated with a species, the means to prevent infection, and the identification of early symptoms (Fox and Lipman 1991). Standard biosafety practices reduce the likelihood of infections of any type in laboratory science (NRC 1997). These should include: • • • • •
84550.indb 136
Establishing the biosafety level Using aseptic techniques and procedures Practicing good personal hygiene and using PPE Using biosafety cabinets Decontaminating
11/11/10 6:24:24 PM
Laboratory Animal Allergies and Zoonoses
• • • •
137
Appropriately managing waste Learning from incidents Immunizing personnel when suitable vaccines exist Providing postexposure treatment after incidents
Disease-Free Animals Whenever practical, laboratory animals should be selected that are at low risk of disease. Special measures are needed when animal–human infections are being studied (i.e., when animals are known to be infected). Measures to prevent the transmission of infection between animals will normally also prevent transmission from animals to humans. For example, an important benefit of procuring animals that are regularly screened and are at a low risk of important animal diseases is that the animals then also pose a very low risk of transmitting disease to humans. Additional measures to prevent zoonoses should include the quarantine of incoming animals, containment facilities, and promptly dealing with infected animals. The use of live vaccines to protect animal health can pose a risk of infection to handlers (e.g., Newcastle disease) (Swayne and King 2003; Nelson et al. 1952). Awareness It is important that animal workers are aware that zoonoses may be transmitted from carrier animals that appear to be in good health (e.g., Cercopithecine herpesvirus 1). They should also be aware that several human diseases may infect laboratory animals (e.g., measles and hepatitis A). Handlers complying with hygiene requirements reduce the risk to themselves and the animals. Effective hygiene should eliminate the risk of transmission of agents normally passed by the oralfecal route (e.g., Salmonella and Shigella). Care is needed when handling both animals and potentially contaminated materials. It is important that incidents are reported, not just to enable learning and improvement, but also to make sure there is no delay in the treatment of exposed workers. It is not unusual for incidents and infections to go unreported (Weigler et al. 2005). Personal Protective Equipment Suitable clothing is essential when working with animals. The main purposes of personal protective equipment are to reduce the commoner risks of microbiological contamination of the animals and allergen exposure. However, this same clothing, especially the use of gloves and masks, will also reduce the likelihood of zoonotic pathogens being transferred from the animal to the worker. Infection has arisen from handling animals’ broken skin without using gloves (Khabbaz et al. 1994), from injury to the skin from contaminated sharp items, or from scratches and bites (Miller, Songer, and Sullivan 1987; Khabbaz et al. 1992). Health Care of Workers The vulnerability of workers to infection should be assessed at the start of employment. This should be reconsidered whenever the worker has a significant change in health status that might increase the likelihood of infection (e.g., reduced immune response) or increase the consequences of infection (e.g., pregnancy) (Grant and Olsen 1999). When health status changes, there may be a need to consider additional precautions, as well as temporary or permanent redeployment.
84550.indb 137
11/11/10 6:24:24 PM
138
Handbook of Laboratory Animal Science
Table€7.2╅Some Important Zoonotic Agents Viruses Herpes B Rabies Hantavirus Lymphocytic choriomeningitis Hepatitis Simian immunodeficiency
Bacteria
Protozoa
Fungi
Tuberculosis Streptobacillus (rat bite fever) Brucellosis Leptospirosisosis Campylobacteri Salmonellosis Shigellosis
Amebiasis Toxoplasmosis
Trichophyton
When suitable vaccines exist, it is possible to protect workers from some zoonoses by immunization (e.g., rabies). Animal workers should seek medical advice promptly following exposure to an infected animal or the development of unexpected symptoms. The consequences of delayed treatment can be catastrophic (Weigler et al. 2005; Cohen et al. 2002). Protecting Animals from Human Disease It is important to recognize that some laboratory animals are also susceptible to human disease. While the vulnerability of primates to human diseases, such as measles, might be predicted from the close phylogenetic relationship, the vulnerability of other species is not always self-evident (e.g., influenza in ferrets) (Marini, Adkins, and Fox 1989). Standard biosafety precautions also reduce the risk to the animals from humans. This sensitivity of some species to human disease offers an opportunity in relevant circumstances to study human diseases and treatments in animal models. Important Zoonotic Diseases A full and detailed description of the medical management of all relevant zoonoses is beyond the scope of this review. Zoonoses important in laboratory animal science have been reviewed in detail elsewhere (Hankenson et al. 2003). All zoonoses from laboratory animals are very uncommon. Some of the most important ones are shown in Table€7.2. Although rare, there have a number of case reports of zoonoses associated with laboratory animal work. In some cases, specific precautions are available: Herpes B. There have been more than 20 deaths from Cercopithecine herpesvirus 1 (herpes B virus) infection associated with the care of primates (Cohen et al. 2002). Early flu-like symptoms progress rapidly in untreated patients to life-threatening encephalomyelitis. It is essential that exposed workers be treated as soon as possible with antiviral medications (Cohen et al. 2002). Hantavirus. Hantaviruses can be acquired from rodents (Umenai et al. 1979; Lloyd et al. 1984; Centers for Disease Control and Prevention 1994). They can cause a severe lung disease. There is no specific treatment for hantavirus infection. Rabies. No cases have been reported in laboratory animal workers. The disease is noteworthy because a vaccine is available. Early treatment of workers exposed to the virus (usually from the bite of an infected animal) is essential. LCMV. Lymphocytic choriomeningitis is viral infectious disease borne by rodents (usually mice) (Jahrling and Peters 1992). It is rare for people to become ill from LCMV. The disease poses a more serious risk to pregnant workers. SIV. Cases of exposure to simian immunodeficiency virus with development of antibodies have been reported in laboratory workers (Khabbaz et al. 1992, 1994; Centers for Disease Control and Prevention 1992; Sotir et al. 1997). Tuberculosis. Tuberculosis has been reported in relation to work with laboratory animals. A vaccine is available for the immunization of workers at risk.
84550.indb 138
11/11/10 6:24:24 PM
Laboratory Animal Allergies and Zoonoses
139
Brucellosis. This was reported relatively frequently in the past, especially among veterinarians (Pike 1976). Vaccines are not available for humans. Dermatophytoses. Animal workers are at risk of fungal infections, especially of the skin (Hironaga, Fujigaki, and Watanabe 1981). Skin should be intact and, when practicable, covered.
Summary The biological risks to laboratory animal workers are significant. Allergy is common and workers are at high risk of developing disease if thorough precautions are not in place. Zoonoses are rarely reported but, in some cases, pose a severe risk of very serious and life-threatening illness. Adequate assessment and control measures will not eliminate the risk completely but will substantially reduce the likelihood of serious morbidity. References Agrup, G., and L. Sjöstedt. 1985. Contact urticaria in laboratory technicians working with animals. Acta Dermato Venereologica 65:111–115. Anon. 1980. Allergy to laboratory animals. Veterinary Record 107:122–123. Aoyama, K., A. Ueda, F. Manda, T. Matsushita, T. Ueda, and C. Yamauchi. 1992. Allergy to laboratory animals: An epidemiological study. British Journal of Industrial Medicine 49:41–47. Bayard, C., L. Holmquist, and O. Vesterberg. 1996. Purification and identification of allergenic alpha2u-globulin species of rat urine. Biochimica et Biophysica Acta 1290:129–134. Bertrand, S., R. Rimhanen-Finne, F. X. Weill, W. Rabsch, L. Thornton, J. Perevoscikovs, W. van Pelt, and M. Heck. 2008. Salmonella infections associated with reptiles: The current situation in Europe. Eurosurveillance 13(24). Botham, P. A., C. T. Lamb, E. L. Teasdale, S. M. Bonner, and J. A. Tomenson. 1995. Allergy to laboratory animals: A follow-up study of its incidence and of the influence of atopy and preexisting sensitization on its development. Occupational and Environmental Medicine 52:129–133. Brehler, R., and B. Kutting. 2001. Natural rubber latex allergy: A problem of interdisciplinary concern in medicine. Archives of Internal Medicine 161:1057–1064. Bryant, D., L. M. Boscato, P. N. Mboloi, and M. C. Stuart. 1995. Allergy to laboratory animals among animal handlers. Medical Journal of Australia 163:415–418. Bush, R. K., and G. M. Stave. 2003. Laboratory animal allergy: An update. ILAR Journal 44 (1): 28–51. Bush, R. K., R. A. Wood, and P. A. Eggleston. 1998. Laboratory animal allergy. Journal of Allergy and Clinical Immunology 102:99–112. Centers for Disease Control and Prevention. 1992. Seroconversion to simian immunodeficiency virus in two laboratory workers. Morbidity and Mortality Weekly Report 41:678–681. ———. 1994. Laboratory management of agents associated with hantavirus pulmonary syndrome: interim biosafety guidelines. Morbidity and Mortality Weekly Report 43 (RR-7): 1–7. Clough, G., J. Wallace, M. R. Gamble, E. R. Merryweather, and E. Bailey. 1995. A positive, individually ventilated caging system: A local barrier system to protect both animals and personnel. Laboratory Animals 29:139–151. Cohen, J. I., D. S. Davenport, J. A. Stewart, S. Deitchman, J. K. Hilliard, and L. E. Chapman. 2002. Recommendations for prevention of and therapy for exposure to B virus (Cercopithecine herpesvirus 1). Clinical Infectious Disease 35:1191–1203. Cullinan, P., A. Cook, S. Gordon, M. J. Nieuwenhuijsen, R. D. Tee, K. M. Venables, J. C. McDonald, and A. J. Newman Taylor. 1999. Allergen exposure, atopy, and smoking as determinants of allergy to rats in a cohort of laboratory employees. European Respiratory Journal 13:1139–1143. Cullinan, P., D. Lowson, M. J. Nieuwenhuijsen, S. Gordon, R. D. Tee, K. M. Venables, J. C. McDonald, and A. J. Newman Taylor. 1994. Work-related symptoms, sensitization, and estimated exposure in workers not previously exposed to laboratory rats. Occupational and Environmental Medicine 51:589–592.
84550.indb 139
11/11/10 6:24:24 PM
140
Handbook of Laboratory Animal Science
Das, R., I. B. Tager, T. Gamsky, M. B. Schenker, S. Roycce, and J. R. Balmes. 1992. Atopy and airways reactivity in animal-health technicians. A pilot study. Journal of Occupational Medicine 34:53–60. Edwards, R. G., M. F. Beeson, and J. M. Dewdney. 1983. Laboratory animal allergy: The measurement of airborne urinary allergens and effects of different environmental conditions. Lab Anim 17:235–239. Eggleston, P. A., C. A. Newill, A. A. Ansari, A. Pustelnik, S.-R. Lou, R. Evans III, D. G. Marsh, J. L. Longbottom, and M. Corn. 1989. Task-related variation in airborne concentrations of laboratory animal allergens: Studies with Rat n I. Journal of Allergy and Clinical Immunology 84:347–352. Elliott, L., D. Heederik, S. Marshall, D. Peden, and D. Loomis. 2005a. Incidence of allergy and allergy symptoms among workers exposed to laboratory animals. Occupational and Environmental Medicine 62 (11): 766–671. ———. 2005b. Progression of self-reported symptoms in laboratory animal allergy. Journal of Allergy and Clinical Immunology 116 (1): 127–132. Fisher, R., W. B. Saunders, S. J. Murray, and G. M. Stave. 1998. Prevention of laboratory animal allergy. Journal of Occupational and Environmental Medicine 40:609–613. Fox, J. G., and N. S. Lipman. 1991. Infections transmitted by large and small laboratory animals. Infectious Disease Clinics of North America 5 (1): 131–163. Fuortes, L. J., L. Weih, M. L. Jones, L. F. Burmeister, P. S. Thorne, S. Pollen, and J. A. Merchant. 1996. Epidemiologic assessment of laboratory animal allergy among university employees. American Journal of Industrial Medicine 29:67–74. Fuortes, L. J., L. Weih, P. Pomrehn, P. S. Thorne, M. Jones, L. Burmeister, and J. A. Merchant. 1997. Prospective epidemiologic evaluation of laboratory animal allergy among university employees. American Journal of Industrial Medicine 32:665–669. Gautrin, D., H. Ghezzo, C. Infante-Rivard, and J.-L. Malo. 2000. Incidence and determinants of IgE-mediated sensitization in apprentices. A prospective study. American Journal of Respiratory and Critical Care Medicine 163:1222–1228. Gautrin, D., C. Infante-Rivard, H. Ghezzo, and J.-L. Malo. 2001. Incidence and host determinants of probable occupational asthma in apprentices exposed to laboratory animals. American Journal of Respiratory and Critical Care Medicine 163:899–904. Goodno, L. E., and G. M. Stave. 2002. Primary and secondary allergies to laboratory animals. Journal of Occupational and Environmental Medicine 44 (12): 1143–1152. Gordon, S., and R. Preece. 2003. Prevention of laboratory animal allergy. Occupational Medicine (Lond) 53 (6): 371–377. Gordon, S., W. Stanley, S. W. Fisher, and R. H. Raymond. 2001. Elimination of mouse allergens in the working environment: Assessment of individually ventilated cage systems and ventilated cabinets in the containment of mouse allergens. Journal of Allergy and Clinical Immunology 108:288–294. Gordon, S., R. D. Tee, D. Lowson, J. Wallace, and A. J. Newman Taylor. 1992. Reduction of airborne allergenic urinary proteins from laboratory rats. British Journal of Industrial Medicine 49:416–422. Gordon, S., R. D. Tee, and A. J. Newman Taylor. 1993. Analysis of rat urine proteins and allergens by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting. Journal of Allergy and Clinical Immunology 92:298–305. Gordon, S., J. Wallace, A. Cook, R. D. Tee, and A. J. Newman Taylor. 1997. Reduction of exposure to laboratory animal allergens in the workplace. Clinical and Experimental Allergy 27:744–751. Grant, S., and C. W. Olsen. 1999. Preventing zoonotic diseases in immunocompromised persons: The role of physicians and veterinarians. Emerging Infectious Diseases 5:159–163. Hankenson, F. C., N. A. Johnston, B. J. Weigler, and R. F. Di Giacomo. 2003. Zoonoses of occupational health importance in contemporary laboratory animal research. Comparative Medicine 53 (6): 579–601. Harrison, D. J. 2001. Controlling exposure to laboratory animal allergens. ILAR Journal 42:17–36. Heederik, D., K. M. Venables, P. Malmberg, A. Hollander, A.-S. Karlsson, A. Renström, G. Doekes, M. Nieuwenhuijsen, and S. Gordon. 1999. Exposure–response relationship for work-related sensitization in workers exposed to rat urinary allergens: Results from a pooled study. Journal of Allergy and Clinical Immunology 103:678–684. Hesford, J. D., T. A. E. Platts-Mills, and R. F. Edlich. 1995. Anaphylaxis after laboratory rat bite: An occupational hazard. Journal of Emergency Medicine 13:765–768. Hironaga, M., T. Fujigaki, and S. Watanabe. 1981. Trichophyton mentagrophytes skin infections in laboratory animals as a cause of zoonosis. Mycopathologia 73 (2): 101–104.
84550.indb 140
11/11/10 6:24:25 PM
Laboratory Animal Allergies and Zoonoses
141
Höglund, U., and A. Renström. 2001. Evaluation of individually ventilated cage systems for laboratory rodents. Cage environment and animal health aspects. Lab Anim 35:51–57. Hollander, A., S. Gordon, A. Renstrom, J. Thissen, G. Doekes, P. H. Larsson, P. Malmberg, K. M. Venables, and D. Heederik. 1999. Comparison of methods to assess airborne rat and mouse allergen levels. I. Analysis of air samples. Allergy 54:142–149. Hollander, A., D. Heederik, and G. Doekes. 1997. Respiratory allergy to rats: Exposure-response relationships in laboratory animal workers. American Journal of Respiratory and Critical Care Medicine 155:562–567. Hollander, A., D. Heederik, G. Doekes, and H. Kromhout. 1998. Determinants of airborne rat and mouse urinary allergen exposure. Scandinavian Journal of Work, Environment & Health 24:228–235. Hollander, A., P. Van Run, J. Spithoven, D. Heederik, and G. Doekes. 1997. Exposure of laboratory animal workers to airborne rat and mouse urinary allergens. Clinical and Experimental Allergy 27:617–626. Hunskaar, S., and R. T. Fosse. 1990. Allergy to laboratory mice and rats: A review of the pathophysiology, epidemiology, and clinical aspects. Lab Anim 34:358–379. ———. 1993. Allergy to laboratory mice and rats: A review of its prevention, management, and treatment. Lab Anim 27:206–221. Jahrling, P. B., and C. J. Peters. 1992. Lymphocytic choriomeningitis virus. Laboratory Medicine 116:486–488. Jeal, H., A. Draper, J. Harris, A. Newman Taylor, P. Cullinan, and M. Jones. 2006. Modified Th2 Responses at high-dose exposures to allergen. American Journal of Respiratory and Critical Care Medicine 174:21–25. Johansson, S. G., T. Bieber, R. Dahl, P. S. Friedmann, B. Q. Lanier, R. F. Lockey, C. Motala, J. A. Ortega Martell, T. A. Platts-Mills, J. Ring, F. Thien, P. Van Cauwenberge, and H. C. Williams. 2004. Revised nomenclature for allergy for global use: Report of the Nomenclature Review Committee of the World Allergy Organization, October 2003. Journal of Allergy and Clinical Immunology 113:832–836. Jones, M. G. 2008. Exposure–response in occupational allergy. Current Opinion in Allergy and Clinical Immunology 8 (2): 110–114. Jones, R. B., J. B. Kacergis, M. R. MacDonald, F. T. McKnight, W. A. Turner, J. L. Ohman, and B. Paigen. 1995. The effect of relative humidity on mouse allergen levels in an environmentally controlled mouse room. American Industrial Hygiene Association Journal 56:398–401. Kacergis, J. B., R. B. Jones, C. K. Reeb, W. A. Turner, J. L. Ohman, M. R. Ardman, and B. Paigen. 1996. Air quality in an animal facility: Particulates, ammonia, and volatile organic compounds. American Industrial Hygiene Association Journal 57:634–640. Kaliste, E., M. Linnainmaa, T. Meklin, E. Torvinen, and A. Nevalainen. 2004. The bedding of laboratory animals as a source of airborne contaminants. Laboratory Animals 38:25–37. Keverne, E. B. 1998. Vomeronasal/accessory olfactory system and pheromonal recognition. Chemical Senses 23:491–494. Khabbaz, R. F., W. Heneine, J. R. George, B. Parekh, T. Rowe, T. Woods, W. M. Switzer, H. M. McClure, M. Murphey-Corb, and T. M. Folks. 1994. Infection of a laboratory worker with simian immunodeficiency virus. New England Journal of Medicine 330 (3): 172–177. Khabbaz R. F., T. Rowe, M. Murphey-Corb, W. Heneine, C. A. Schalbe, J. R. George, P. Chou-Pong, B. Parekh, M. D. Lairmore, J. W. Curran, J. E. Kaplan, G. Schochetman, and T. M. Folks. 1992. Simian immunodeficiency virus needlestick accident in a laboratory worker. Lancet 340:271–273. Kibby, T., G. Powell, and J. Cromer. 1989. Allergy to laboratory animals. Journal of Occupational Medicine 31:842–846. Kimber, I. 1996. The role of the skin in the development of chemical respiratory sensitivity. Toxicology Letters 86:89–92. Krohn, T. C., G. Gabi Itter, R. Fosse, and A. K. Hansen. 2006. Controlling allergens in animal rooms by using curtains. Journal of the American Association for Laboratory Animal Science 45 (3): 51–53. Krop, E. J., G. Doekes, M. J. Stone, R. C. Aalberse, and J. S. van der Zee. 2007. Spreading of occupational allergens: Laboratory animal allergens on hair-covering caps and in mattress dust of laboratory animal workers. Occupational Environmental Medicine 64 (4): 267–272. Kruize, H., W. Post, D. Heederik, B. Martens, A. Hollander, and E. van der Beek. 1997. Respiratory allergy in laboratory animal workers: A retrospective cohort study using preemployment screening data. Occupational and Environmental Medicine 54:830–835.
84550.indb 141
11/11/10 6:24:25 PM
142
Handbook of Laboratory Animal Science
Lieutier-Colas, F., P. Meyer, P. Larsson, P. Malmberg, N. Frossard, G. Pauli, and F. De Blay. 2001. Difference in exposure to airborne major rat allergen (Rat n 1) and to endotoxin in rat quarters according to tasks. Clinical & Experimental Allergy 31 (9): 1449–1456. Lieutier-Colas, F., P. Meyer, F. Pons, G. Hedelin, P. Larsson, P. Malmberg, G. Pauli, and F. De Blay. 2002. Prevalence of symptoms, sensitization to rats, and airborne exposure to major rat allergen (Rat n 1) and to endotoxin in rat-exposed workers: A cross-sectional study. Clinical & Experimental Allergy 32 (10): 1424–1429. Lindqvist, C., L. Persson, K. Iwarsson, G. Lustig, A. Renström, and P. H. Larsson. 1996. A sliding curtain system for improving air distribution of animal rooms in relation to working environment and cage climate. Scandinavian Journal of Laboratory Animal Science 23:135–143. Lipman, N. S. 1999. Isolator rodent caging system (state of the art): A critical view. Contemporary Topics in Laboratory Animal Science 38:9–17. Lloyd, G., E. T. Bowen, N. Jones, and A. Pendry. 1984. HFRS outbreak associated with laboratory rats in UK. Lancet 1:1175–1176. Lorusso, J. R., S. Moffat, and J. L. Ohman, Jr. 1986. Immunologic and biochemical properties of the major mouse urinary allergen (Mus m 1). Journal of Allergy and Clinical Immunology 78:928–937. Lutsky, I. I., and I. Neuman. 1975. Laboratory animal dander allergy: I. An occupational disease. Annals of Allergy 35 (4): 201–205. Malo, J.-L., L.-P. Boulet, J.-D. Dewitte, A. Cartier, J. L’Archevêque, J. Côté, G. Bédard, S. Boucher, F. Champagne, G. Tessier, A-P. Constanopoulos, E. F. Juniper, and G. H. Guyalt. 1993. Quality of life of subjects with occupational asthma. Journal of Allergy and Clinical Immunology 91:1121–1127. Mancini, M. A., D. Majumda, B. Chatterjee, and A. K. Roy. 1989. Alpha2u-globulin in modified sebaceous glands with pheromonal functions: Localization of the protein and its mRNA in preputial, meibomian, and perianal glands. Journal of Histochemistry and Cytochemistry 37:149–157. Marini, R. P., J. A. Adkins, and J. G. Fox. 1989. Proven or potential zoonotic diseases of ferrets. Journal of the American Veterinary Medical Association 195 (7): 990–994. Meijer, E., D. E. Grobbee, and D. Heederik. 2004. A strategy for health surveillance in laboratory animal workers exposed to high molecular weight allergens. Occupational Environmental Medicine 61 (10): 831–837. Miller, C. D., J. R. Songer, and J. F. Sullivan. 1987. A twenty-five year review of laboratory acquired human infections at the National Animal Disease Center. American Industrial Hygiene Association Journal 48:271–275. Myers, D. D., E. Smith, I. Schweitzer, J. D. Stockwell, B. J. Paigen, R. Bates, J. Palmer, and A. L. Smith. 2003. Assessing the risk of transmission of three infectious agents among mice housed in a negatively pressurized caging system. Contemporary Topics in Laboratory Animal Science 42 (6): 16–21. National Institute for Occupational Safety and Health. 1998. NIOSH alert: Preventing asthma in animal handlers. Publication no. 97–116. NRC (National Research Council). 1997. Occupational health and safety in the care and use of research animals. Washington, D.C.: National Academy Press. Nelson, C. B., B. S. Pomeroy, K. Schrall, W. E. Park, and R. J. Lindeman. 1952. An outbreak of conjunctivitis due to Newcastle disease virus (NDV) occurring in poultry workers. American Journal of Public Health and the Nation’s Health 42:672–678. Newill, C. A., R. Evans III, and M. Khoury. 1986. Preemployment screening for allergy to laboratory animals: Epidemiologic evaluation of its potential usefulness. Journal of Occupational Medicine 28:1158–1164. Nieuwenhuijsen, M. J., S. Gordon, J. M. Harris, R. D. Tee, K. M. Venables, and A. J. Newman Taylor. 1995. Variation in rat urinary aeroallergen levels explained by differences in site, task, and exposure group. Annals of Occupational Hygiene 39:819–825. Nieuwenhuijsen, M. J., V. Putcha, S. Gordon, D. Heederik, K. M. Venables, P. Cullinan, and A. J. NewmanTaylor. 2003. Exposure-response relations among laboratory animal workers exposed to rats. Occupational and Environmental Medicine 60:104–108. Ooms, T. G., J. E. Artwohl, L. M. Conroy, T. M. Schoonover, and J. D. Fortman. 2008. Concentration and emission of airborne contaminants in a laboratory animal facility housing rabbits. Journal of the American Association for Laboratory Animal Science 47 (2): 39–48. Pacheco, K. A. 2007. New insights into laboratory animal exposures and allergic responses. Current Opinion in Allergy and Clinical Immunology 7 (2): 156–161.
84550.indb 142
11/11/10 6:24:25 PM
Laboratory Animal Allergies and Zoonoses
143
Pacheco, K. A., C. McCammon, A. H. Liu, P. S. Thorne, M. E. O’Neill, J. Martyny, L. S. Newman, R. F. Hamman, and C. S. Rose. 2003. Airborne endotoxin predicts symptoms in non-mouse-sensitized technicians and research scientists exposed to laboratory mice. American Journal of Respiratory and Critical Care Medicine 167 (7): 983–990. Perfetti, L., J. Hébert, Y. Lapalme, H. Ghezzo, D. Gautrin, and J. L. Malo. 1998. Changes in IgE-mediated allergy to ubiquitous inhalants after removal from or diminution of exposure to the agent causing occupational asthma. Clinical & Experimental Allergy 28:66–73. Pike, R. M. 1976. Laboratory-associated infections: Summary and analysis of 3,921 cases. Health Laboratory Science 13:105–114. ———. 1979. Laboratory associated infections: Incidence, fatalities, causes, and prevention. Annual Review of Microbiology 33:41–46. Platts-Mills J., N. Custis, A. Kenney, A. Tsay, M. Chapman, S. Feldman, and T. Platts-Mills. 2005. The effects of cage design on airborne allergens and endotoxin in animal rooms: High-volume measurements with an ion-charging device. Contemporary Topics in Laboratory Animal Science 44 (2): 12–16. Platts-Mills, T. A. E., P. Heyman, J. L. Longbottom, and S. R. Wilkins. 1986. Airborne allergens associated with asthma: Particle size carrying dust mite and rat allergens measured with a cascade impactor. Journal of Allergy and Clinical Immunology 77:850–857. Platts-Mills, T., J. Vaughan, S. Squillace, J. Woodfolk, and R. Sporik. 2001. Sensitization, asthma, and a modified Th2 response in children exposed to cat allergen: A population-based cross-sectional study. Lancet 357:752–756. Portengen, L., A. Hollander, G. Doekes, G. de Meer, and D. Heederik. 2003. Lung function decline in laboratory animal workers: The role of sensitization and exposure. Occupational Environmental Medicine 60 (11): 870–875. Price, J. A., and J. L. Longbottom. 1988. ELISA method for measurement of airborne levels of major laboratory animal allergens. Clinical Allergy 18:95–107. Reeb-Whitaker, C. K., D. J. Harrison, R. B. Jones, J. B. Kacergis, D. D. Myers, and B. Paigen. 1999. Control strategies for aeroallergens in an animal facility. Journal of Allergy and Clinical Immunology 103:139–146. Renström A., K. Blidberg, and K. Larsson. 2007. Quality of life in laboratory animal workers with and without allergic symptoms. Abstract 3731, ERS Stockholm. Renström, A., U. Höglund, and G. Björing. 2001. Evaluation of individually ventilated cage systems for laboratory rodents. Occupational health aspects, Lab Anim 35: 42–50. Renström, A., and C. Kallinn. 2000. Characterization of the work environment in an animal laboratory facility, Stockholm: NIWL (unpublished report). Renström, A., A.-S. Karlsson, P. Malmberg, P.H. Larsson, and M. van Hage-Hamsten. 2001. Working with male rodents may increase risk for laboratory animal allergy. Allergy 56:964–970. Renström A., A.-S. Karlsson, and E. Tovey. 2002. Nasal air sampling used for the assessment of occupational allergen exposure and the efficacy of respiratory protection. Clinical and Experimental Allergy 32 (12): 1769–1775. Renström, A., P. Malmberg, K. Larsson, B.-M. Sunblad, and P. H. Larsson. 1994. Prospective study of laboratoryanimal allergy: Factors predisposing to sensitization and development of allergic symptoms. Allergy 49:548–552. Rudzki, E., P. Rebandel, and T. Rogozinski. 1981. Contact urticaria from rat tail, guinea pig, streptomycin, and vinyl pyridine. Contact Dermatitis 7:186–188. Rylander, R. 2006. Endotoxin and occupational airway disease. Current Opinion in Allergy and Clinical Immunology 6 (1): 62–66. Sakaguchi, M., S. Inouye, H. Miyazawa, M. Kimura, and S. Yamazaki. 1990. Evaluation of countermeasures for reduction of mouse airborne allergens. Laboratory Animal Science 40:613–615. Schweitzer, I. B., E. Smith, D. J. Harrison, D. D. Myers, P. A. Eggleston, J. D. Stockwell, B. Paigen, and A. L. Smith. 2003. Reducing exposure to laboratory animal allergens. Comparative Medicine 53 (5): 487–92. Seward, J. P. 1999. Occupational allergy to animals. Occupational Medicine 14:285–302. ———. 2001. Medical surveillance of allergy in laboratory animal handlers. ILAR Journal 42:47–54. Sewell, D. L. 1995. Laboratory-associated infections and biosafety. Clinical Microbiology Reviews 8 (3): 389–405.
84550.indb 143
11/11/10 6:24:25 PM
144
Handbook of Laboratory Animal Science
Sharp, P. M., G. M. Shaw, and B. H. Hahn. 2005. Simian immunodeficiency virus infection of chimpanzees. Journal of Virology 79 (7): 3891–3902. Skoke, H. H. 1995. Ventilated tables’ impact on workstation and building design. Lab Anim 24:22–25. Slovak, A. J. M., R. G. Orr, and E. L. Teasdale. 1985. Efficacy of the helmet respiratory in occupational asthma due to laboratory animal allergy (LAA). American Industrial Hygiene Association Journal 46:411–415. Sotir, M., W. Switzer, C. Schable, J. Schmitt, C. Vitek, and R. F. Khabbaz. 1997. Risk of occupational exposure to potentially infectious nonhuman primate materials and to simian immunodeficiency virus. Journal of Medical Primatology 26:233–240. Swanson, M. C., M. K. Agarwal, J. W. Yunginger, and C. E. Reed. 1984. Guinea-pig-derived allergens. Clinicoimmunologic studies, characterization, airborne quantitation, and size distribution. American Review of Respiratory Disease 129:844–849. Swanson, M. C., A. R. Campbell, M. T. O’Hollaren, and C. E. Reed. 1990. Role of ventilation, air filtration, and allergen production rate in determining concentrations of rat allergens in the air of animal quarters. American Review of Respiratory Disease 141:1578–1581. Swayne, D. E., and D. J. King. 2003. Avian influenza and Newcastle disease. Journal of the American Veterinary Medical Association 222:1534–1540. Taubenberger, J. K., A. H. Reid, R. M. Lourens, R. Wang, G. Jin, and T. G. Fanning. 2005. Characterization of the 1918 influenza virus polymerase genes. Nature 437:889–893. Taylor, L. H., S. M. Latham, and M. E. J. Woolhouse. 2001. Risk factors for human disease emergence. Philosophical Transactions of the Royal Society of London B 356:983–989. Teasdale, E. L., E. G. Davies, and R. Slovak. 1993. Anaphylaxis after bites by rodents. British Medical Journal 286:1480. Thulin, H., M. Björkdahl, A.-S. Karlsson, and A. Renström. 2002. Reduction of exposure to laboratory animal allergens in a research laboratory. Annals of Occupational Hygiene 46 (1): 61–68. Umenai, T., H. W. Lee, P. W. Lee, T. Saito, T. Toyoda, M. Hongo, K. Yoshinaga, T. Nobunaga, T. Horiuchi, and N. Ishida. 1979. Korean hemorrhagic fever in staff in an animal laboratory. Lancet 1:1314–1316. Vandoren, G., B. Mertens, W. Heyns, H. van Baelen, W. Rombauts, and G. Verhoven. 1983. Different forms of α2u-globulin in male and female rat urine. European Journal of Biochemistry 134:175–181. Venables, K. M. 1997. Occupational asthma. Lancet 349:1465–1469. Venables, K. M., J. L. Upton, E. R. Hawkins, R. D. Tee, J. L. Longbottom, and A. J. Newman-Taylor. 1988. Smoking, atopy, and laboratory animal allergy. British Journal of Industrial Medicine 45:667–671. Virtanen, T., T. Zeiler, and R. Mäntyjärvi. 1999. Important animal allergens are lipocalin proteins: Why are they allergenic? International Archives of Allergy and Immunology 120:247–258. Wahn, U., T. Peters, Jr., and R. P. Siraganian. 1980. Studies of the allergenic significance and structure of rat serum albumin. Journal of Immunology 125:2544–2549. Walker, D., and D. Campbell. 1999. A survey of infections in United Kingdom laboratories, 1994–1995. Journal of Clinical Pathology 52:415–418. Walls, A., A. Newman-Taylor, and J. Longbottom. 1985. Allergy to guinea pigs. II. Identification of specific allergens in guinea pig dust by crossed radio-immunoelectrophoresis and investigation of possible origin. Clinical Allergy 15:535–546. Watt, A. D., and P. McSharry. 1996. Laboratory animal allergy: Anaphylaxis from a needle injury. Occupational Environmental Medicine 53:573–574. Weigler, B. J., R. F. Di Giacomo, and S. Alexander. 2005. A national survey of laboratory animal workers concerning occupational risks for zoonotic diseases. Comparative Medicine 55 (2): 183–191. Wolfe, N. D., C. Panosian Dunavon, and J. Diamond. 2007. Origins of human infectious diseases. Nature 447:279–283. Wood, R. A. 2001. Laboratory animal allergens. ILAR Journal 42:12–15.
84550.indb 144
11/11/10 6:24:25 PM
Chapter 8
Laboratory Animal Facilities and Equipment for Conventional, Barrier, and Containment Housing Systems
Jack R. Hessler Contents Introduction..................................................................................................................................... 146 Facilities.......................................................................................................................................... 148 General Considerations.............................................................................................................. 148 Location................................................................................................................................. 148 Arrangement.......................................................................................................................... 148 Circulation............................................................................................................................. 148 Function Areas........................................................................................................................... 150 Support Areas........................................................................................................................ 150 Animal Housing Areas.......................................................................................................... 156 Architectural Features............................................................................................................... 165 Interior Surfaces.................................................................................................................... 165 Vermin Control..................................................................................................................... 168 Noise Control........................................................................................................................ 168 Engineering Features................................................................................................................. 169 Heating, Ventilation, and Air Conditioning (HVAC)........................................................... 169 Power and Lighting............................................................................................................... 177 Plumbing............................................................................................................................... 178 Equipment....................................................................................................................................... 182 Cage Sanitation and Sterilization............................................................................................... 182 Batch Washers....................................................................................................................... 183 Continuous Belt Washers...................................................................................................... 185 Robotic Cage Washing.......................................................................................................... 187 Sterilization Equipment......................................................................................................... 189 Animal Watering........................................................................................................................ 190 Caging........................................................................................................................................ 192 Rodents.................................................................................................................................. 192 Rabbits................................................................................................................................... 197 Canines, Swine, and Small Ungulates.................................................................................. 198 145
84550.indb 145
11/11/10 6:24:25 PM
146
Handbook of Laboratory Animal Science
Nonhuman Primates.............................................................................................................. 199 Cats........................................................................................................................................ 199 Equipment and Caging for Biological Control (Barrier and Containment)...............................200 Biological Safety Cabinets....................................................................................................200 Microisolation Caging System.............................................................................................. 201 Isolators................................................................................................................................. 201 Autoclaves.............................................................................................................................203 Conclusion.......................................................................................................................................203 References.......................................................................................................................................203 Introduction The objective of this chapter is to provide an overview of facilities and equipment required for housing laboratory animals and to be a resource for planning and designing research animal facilities. An additional objective is to provide insight into facility operations for planners and designers with minimal experience working in research animal facilities. Many other publications provide additional information on the subject, including: • Planning and Designing Research Animal Facilities, edited by Hessler and Lehner (2009) • A chapter in The Mouse in Biomedical Research, 2nd edition (2007) by Lipman • The DHHS National Institutes of Health Biomedical and Animal Research Facilities Design Policies and Guidelines (2009) • The Canadian Council on Animal Care Guidelines on: Laboratory Animal Facilities— Characteristics, Design, and Development (2003) • A chapter in Laboratory Animal Medicine, 2nd edition (2002) by Hessler and Leary • A chapter in the Handbook of Laboratory Animal Science, 2nd edition (2002) by Hessler and Höglund • An entire issue of Laboratory Animals (2002) dedicated to animal facility design and planning • Handbook of Facilities Planning: Laboratory Animal Facilities (1991) edited by Ruys
A review of the progress made in research animal facilities and equipment during the latter half of the twentieth century is summarized by Hessler (1999) in a chapter of a book published to celebrate the 50th anniversary of the American Association of Laboratory Animal Science (AALAS). A major objective of laboratory animal science is to control the laboratory animal’s environment (Faith and Huerkamp 2009; Faith and Hessler 2006; Lipman and Perkins 2002; Baker, Lindsey, and Weisbroth 1979). Environmental variables can alter the animal’s biology, resulting in background “noise” that can mask the biological response to experimental variables, thus confounding the interpretation of the experimental data. Of course, animal comfort and well-being are paramount, not only for moral reasons, but also for scientific reasons. Distressed animals make poor research subjects, but many biological responses to environmental factors are not manifested in terms of stress, distress, or any overt pathologies. For this reason, the degree of control required for the research animal’s environment is primarily science driven, going well beyond that required to assure the animal’s well-being. It is the responsibility of laboratory animal specialists not only to provide for the comfort and well-being of the animals, but also to assist the scientist with controlling animal-related variables that may confound the science. Figure€8.1 illustrates the many environmental factors that must be considered, including genetic, microbial, chemical, and physical. Control of genetic variables is primarily a matter of biology, but control of other variables is dependent to a significant degree on the design and management of the research animal facility and available equipment. Environmental standards and design concepts for animal facilities are constantly evolving toward higher levels of performance with regard to controlling the animal’s environment and operational efficiency (Hessler 1999). Properly designed and equipped facilities greatly facilitate effective management and the consistent day-to-day animal care
84550.indb 146
11/11/10 6:24:26 PM
Laboratory Animal Facilities and Equipment
147
Environmental Factors
Cage
Food
Humidity
H2O
l
a ic em Ch
Ph ysi cal
Temperature
Genetic Factors
Lighting
Air
Insecticides
Noise
Pheromones Population per Cage
Qu
alit
y
Misc. Chemical
l tro
n
Co
Bedding Pathogens
NH3
Microbial
Figure 8.1â•…Conceptual depiction of the laboratory animal and how its biology, while based on genetics, is influenced by environmental factors. Controlling these environmental factors, any of which may confound animal research data, is a major component of laboratory animal science. (From Baker, H. J., J. R. Lindsey, and S. H. Weisbroth. 1979. In The Laboratory Rat: Vol. I: Biology and Diseases, ed. H. J. Baker, J. R. Lindsey, and S. H. Weisbroth, 169–192. New York: Academic Press. With permission.)
required for optimal support of animal research and testing. In spite of the many choices and multiple opportunities for creativity in designing research animal facilities, the general research animal facility and caging standards have evolved to become well defined (ILAR 1996; Council of Europe 1985; Canadian Council on Animal Care 1993; Federation of Animal Science Societies 1999). There are a variety of ways to categorize facilities for housing laboratory animals. The three most common are addressed in this chapter—conventional, barrier, and containment facilities. For the sake of clarity, because these are not necessarily universally defined terms, they are defined here as follows: • Barrier (keep out) —animal housing systems designed and managed to protect the animals from undesirable microbes • Containment (keep in) —animal housing systems designed and managed to contain experimental or naturally occurring hazards—for example, biological, chemical, and physical (radiation) in order to protect workers, other animals, and the general environment • Conventional—standard housing systems for laboratory animals that do not offer the added level of control provided by barrier and containment systems
84550.indb 147
11/11/10 6:24:28 PM
148
Handbook of Laboratory Animal Science
These categories do not pertain to any particular species. However, the descriptions in this chapter pertain primarily to standard laboratory animals, and the term “barrier” almost always pertains to rodents and is often referred to as a “rodent barrier.”
Facilities General Considerations Location Animal facilities should be located in a secure area isolated from the rest of the research facility for many reasons, including public health, public relations, security, human comfort, animal health, and animal husbandry considerations. Because animal allergens pose a potentially serious health risk (see Chapter 7, this volume), exposure to animals, animal dander, equipment soiled by animals, and animal waste products must be limited to personnel whose jobs require exposure. For these individuals, steps should be taken to minimize such exposure (Huerkamp et al. 2009; National Research Council 1997). Careful planning must reconcile the necessity for isolating the animal facility with the desirability of locating animal facilities as close as possible to the research laboratories. In addition, access and egress patterns for research staff, supplies, animals, and trash need to be planned carefully to facilitate efficiency, reduce contamination between animal rooms, and prevent unnecessary exposure of personnel to animals and animal waste products. Arrangement A single-story, centralized facility with direct access to ground-level transportation is the most efficient facility to operate. Alternative arrangements include a centralized facility located on an upper floor with dedicated elevators to gain access to ground-level transportation, a centralized facility on multiple floors arranged around dedicated elevators, multiple autonomous units that contain all the necessary animal care and use support services, and satellite facilities with convenient access to ground-level transportation that rely to varying degrees on a primary facility for some support services. If properly planned and managed, almost any arrangement can be made to work, but the more an arrangement varies from the single-floor facility with direct access to ground-level transportation, the less the operational efficiency and the greater the operational cost for the overall animal care and use program will be. See Hessler (2009a) for a comprehensive description of functional adjacencies of various spaces within the animal facility. Circulation Circulation within the animal facility must take many factors into consideration, including people, animals, general supplies, chemicals, equipment, and trash (Conti and Hessler 2009). Vertical circulation in facilities without direct access to ground-level transportation and multilevel animal facilities should include a minimum of two dedicated freight elevators: one for transporting “clean” items and one for “soiled” items; more important, one to serve as backup when the other is being serviced. It is important that each elevator have independent mechanical and control systems to reduce chances for both being out at the same time. The focus of traffic flow in an animal facility revolves around the cage sanitation area and the movement of cages between it and the animal rooms. The horizontal circulation pattern to be used is one of the early decisions to be made in the facility planning process.
84550.indb 148
11/11/10 6:24:28 PM
Laboratory Animal Facilities and Equipment
Unidirectional Wash
Cage Wash Clean Side
Wash
Corridor
41%
Cage Wash Clean Side
Dirty Corridor
Wash
Animal Rooms
Dual Corridors D Cage Cage Wash Wash Clean Dirty Side Side
Animal Rooms
Wash
Animal Rooms
Corridor
Cage Wash Dirty Side
Clean Corridor
Wash
B
Animal Rooms
Wash
Animal Rooms
Wash
Cage Wash Clean Side
Dirty Corridor
33%
Cage Wash Dirty Side
Dirty Corridor
C
Animal Rooms
17%
Wash
Corridor
Cage Wash Dirty Side
Single Corridors
Clean Corridor
Bidirectional
Animal Rooms
A
149
56%
% = Corridor Space/Corridor + Animal Room Space × 100
Figure 8.2â•…Schematics of four basic types of circulation patterns. The arrows indicate the direction of cage traffic between the animal rooms and the cage sanitation area. “A” illustrates a single-corridor bidirectional pattern. “B” illustrates a single-corridor unidirectional pattern. “C” illustrates a dualcorridor pattern with relatively large animal rooms. “D” illustrates a dual-corridor pattern with relatively small animal rooms. All four are drawn within the same footprint to illustrate the relative “cost” of the different circulation patterns and small versus large animal room sizes in terms of the ratio of corridor space to corridor plus animal room space. The percentages only serve to illustrate the significance of choosing a combination of circulation patterns and animal room sizes and do not necessarily apply to a particular plan.
There are two basic horizontal circulation patterns: single corridor and dual corridor. Dual corridors are also known as “clean–dirty” corridors. The objective of the dual corridor circulation pattern is to decrease the potential for cross-contamination between animal rooms. Theoretically, dual corridors are superior to single corridors in terms of reducing cross-contamination; however, as compared to single corridors, they come at a high cost due to the reduced ratio of animal housing space to circulation space. Figure€8.2 illustrates this point. Whether or not dual corridors are cost effective is a complex issue, and the answer will vary according to the relative weight assigned to each of the many pros and cons (Hessler 1991a). Clearly, each potential arrangement has advantages, disadvantages, and limitations (Table€8.1). Few would disagree that a dual-corridor plan is the best choice if cost and space are not issues. However, many single-corridor barrier and containment facilities function effectively in terms of providing adequate control of contamination.
84550.indb 149
11/11/10 6:24:31 PM
150
Handbook of Laboratory Animal Science
Table€8.1╅Advantages and Disadvantages of Dual Corridors Compared with Single Corridors Advantages Separation of clean cages and supplies from soiled cages and trash eliminates the potential for cross-contamination in the corridors Facilitates the flow of supplies and cages through the facility
Reduced congestion in the corridors
Disadvantages
Comments
The higher ratio of circulation space to animal room and support space is costly
The smaller the animal rooms, the higher the space cost of dual corridors and vice versa
Labor costs are higher when managed to maintain a strict unidirectional flow of personnel between clean and soiled sides The additional door required in the animal room limits room layout options and decreases space utilization efficiency
Potential for airborne crosscontamination from the corridor to the room is similar for both dual- and single-corridor configurations Contamination control is the primary issue. Most would agree that a dual-corridor system is the best model for contamination control; however, effective contamination control can be provided in a single-corridor system by using an appropriate combination of management procedures and equipment options Which is more cost effective? The answer to this question depends on the individual situation and how much weight is put on the various advantages and disadvantages
Function Areas Animal facility space may be divided into two major functional types: animal housing space and support and use space. The ratio of animal housing to support and use space varies considerably from facility to facility depending on programmatic requirements. In biomedical research facilities, the trend has been toward increasing the amount of animal support and use space within the animal facilities. This is typically driven by policies that do not permit animals to be taken out of the animal facility or returned once removed, as well as by increasing requirements for space for animal imaging, animal behavioral testing, and other types of core research resources involving procedures conducted with animals. Typically, the ratio of housing to support and use space ranges between 30:70 and 70:30. In general, the smaller the facility, the higher the proportion of space devoted to support and use. Support Areas Administrative, Training, and Personnel Health and Hygiene Managing an animal facility is a complex business that requires the coordinated effort of a variety of staff, many of whom require office space, including professional, management, supervisory, training, and clerical staff (Faith, Corey, and Nelan 2009). It is highly desirable to provide the administrative and training space in a consolidated suite adjacent to the animal facility, but outside the animal biosecurity perimeter. The suite should include space for office equipment and storage of office supplies and files, and amenities such as an office kitchen unit. This is also a highly desirable location to place the training space. Given the importance of training for animal care and animal use staff, such space should have high priority. It should include space for conferencing and training, training equipment, Internet access, and storage of library and training materials. Space for animal procedure training may best be located in the animal facility.
84550.indb 150
11/11/10 6:24:31 PM
Laboratory Animal Facilities and Equipment
151
Animal technician supervisor offices may be in the administrative suite or scattered throughout animal housing areas, depending on the size of the facility. Office space for veterinary technicians may be located inside the surgery suite and/or in the diagnostic laboratory space, which may also be included inside the administrative suite. A safe, efficient, and healthy working environment must be provided for personnel working in the facility (National Research Council 1997). The primary safety issues are animal allergens, infectious agents, chemical hazards, and physical hazards (often related to ergonomic issues). To protect personnel and animals and to reduce the potential for transporting hazardous agents between home and the animal facility, animal care technicians are required to wear work uniforms; other personnel working intermittently in the facility are typically required to wear protective outer garments over their street clothes prior to entering animal rooms and often prior to entering the animal facility. All uniforms and protective outerwear are provided by the facility; thus, storage space is required. In addition, eating and drinking are not permitted in animal housing areas or in most support areas. To accommodate these requirements, support facilities should include lavatory, shower, locker rooms, and a break area. A laundry room for laundering uniforms and surgical linens is useful, even if a commercial laundry service is to be used. In addition, amenities and aesthetic considerations throughout the facility, especially in the break area, should be provided to make for a high-quality work environment and to enhance the recruitment and retention of staff (Meyer and Mollo-Christensen 2009). Animal Care Cage Servicing and Sanitation. This is one of the most important spaces in the animal facility. Mobile animal cages are typically transported between animal rooms and the cage sanitation area at frequencies varying from three times a week to once every 2 weeks. This makes the cage sanitation area the busiest and one of the most critical areas in the facility with regard to location, design, and equipment. The location of this area relative to the animal rooms and the way in which it is designed and equipped have a major impact on how effectively and efficiently adequate animal care can be provided. Typically, the main portion of the cage sanitation area is divided into two sides: a soiled side and a clean side, which are separated by pass-through cage sanitation equipment and a wall (Figure€8.3). Single-room cage sanitation areas are not recommended, but if there is no other option, two separate rooms, adjacent to, but outside the cage wash room, should be provided for (1) dumping soiled bedding and (2) clean cage storage. Because of the high moisture and heat level, as well as the potential for bedding dust accumulation on the clean side of cage wash, if bedding is to be dispensed into cages manually or with an automatic bedding dispenser on the clean side, it is recommended to divide the clean side into two areas, separating the area where cages are discharged from the washers and bedding is dispensed into the cages from the clean cage storage area. In some rodent barrier facilities, there may be a bulk autoclave between the clean side of cage wash and the clean/sterilized cage storage area. The type of cage sanitation equipment (described later) and the amount of space required in the cage sanitation area depend on the species housed, cage types, cage rack capacity of the facility, cage sanitation program and type of equipment (Leary, Majoros, and Tomson 1998). Space for bulk storage of cage sanitation chemicals is also required. Often, this space is provided adjacent to the cage sanitation area, but an even more desirable location is close to the facility loading dock, where it can be delivered in bulk and from where it can readily be piped to the cage sanitation equipment. Feed and Bedding Storage. Feed and bedding are typically delivered to the facility on pallets and then taken out of the storage space one bag at a time. Therefore, the ideal storage location is not at the loading dock, but rather in an area that is as close as possible to the actual point of use, which, in the case of bedding, is the clean side of the cage sanitation area (Figure€8.3). This usually proves to be the best location for feed storage as well. Of course, all circulation space and doors in the path between the dock and the storage areas must be wide enough to accommodate pallets of feed and bedding.
84550.indb 151
11/11/10 6:24:31 PM
84550.indb 152
BO20
Soiled Elevator Lobby
Toilet ??
Soiled BO19
Stor. BO21
Rack Washer
Rack Washer
Stair = 1 BO25
BO22
Elevator Equipment Room
Tunnel Washer
Corr BO75
BO24
Cage San. Storage
Clean BO23
BO26
Clean Elevator LOBBY
HSKP. Stor. BO27
Cage Storage BO28
Cage Preparation
BO32
Animal Receiving
Super. Office BO33
CORR BO34
BO35
Cold Box
Vestibule BO43
Bedding Storage BO31
Necropsy BO30
Freezer BO29
Corr. BO96
Storage BO37
A
Figure 8.3â•…Schematic drawing of a cage sanitation area with two cage and rack washers and one tunnel washer. The area consists of the soiled side (B019), sanitation chemical storage (B021), clean side (B023), clean cage storage and cage preparation (B028), bedding storage (B031), and a small cage parts storage room (B024). It is located adjacent to a pair of elevators that connect the cage sanitation area to a corridor just outside a rodent barrier facility on the floor above. One elevator is used for transporting soiled cages (elevator lobby B020) and one for transporting clean cages (elevator lobby B026). The bedding storage has two doors; one connects to the cage sanitation area for convenient access to the automatic bedding dispenser in the clean side at the end of the tunnel washer (Figure 8.17b,c), and the other connects to a corridor that leads to the nearby receiving dock. The floors on the soiled and clean sides slope toward grate-covered drain troughs the width of the rooms. The grate-covered drain pit on the discharge end of the rack washers spans the width of the two washers and extends out 2.4 m (8 ft) from the washers. The floor in the cage storage area gently slopes to a grate-covered drain trough in the center of the room. Immediately to the left is a containment facility (Figure 8.7).
Animal Room BO17
Autoclave
Bding Disposal BO18
152 Handbook of Laboratory Animal Science
11/11/10 6:24:35 PM
Laboratory Animal Facilities and Equipment
153
The maximum recommended storage temperature for natural ingredient feed is 21°C (70°F) (ILAR 1996). Purified and chemically defined diets, though dry, are often less stable, and thus their shelf life may be significantly shorter than that of natural ingredient diets, unless they are stored at 4°C (39°F) (Fullerton, Greenman, and Kendall 1982). Refrigerated storage space is also required for fresh food items (i.e., meats, fruits, and vegetables). The need for food preparation capabilities ranges from none to complex, depending on the animals and research being supported. Safety testing laboratories that administer test compounds in feed require highly specialized preparation areas that allow for safe mixing of potentially hazardous compounds with animal feed. Housekeeping and Supply Storage. This space is required to support sanitation of animal rooms, corridors, and other support areas. These include storage rooms for sanitation supplies and equipment, including floor scrubbers, and janitorial and mop closets strategically located in corridors and self-contained areas, such as the surgery suite, biocontainment, and rodent barriers. Receiving and Shipping. For most facilities, a dedicated, strategically located, and well-designed receiving and shipping area is essential for handling the large volume of supplies (e.g., bedding, feed, sanitation chemicals, disposables) and animals routinely received into an animal facility. A nearly equal volume of materials, mostly in the form of trash, exits the facility. Ideally, a separate dock or, at least, an isolated portion of the receiving and shipping area should be provided for trash disposal. Animal shipments out of the facility are a given for animal production facilities and are becoming increasingly common in research facilities due to the sharing of genetically modified animals between research institutions. At a minimum, the receiving and shipping area should include a dock; a climatecontrolled, enclosed receiving room immediately adjacent to the dock; and a climate-controlled room for short-term housing of animals in transit (either into or out of the facility). In order to accommodate a wide variety of delivery vehicle sizes, the dock should be equipped with a scissor lift that ranges from ground level to the height of large trucks. An overhang extending at least 2 m out from the dock bumper is required to protect animals and supplies from the elements. In very cold climates or in situations where the dock is exposed to a high volume of public traffic, consideration should be given to providing a fully enclosed area in front of the dock that is large enough to hold the delivery vehicles. In addition to automatic rollup doors equipped with flying insect air shields, a standard hinged door for personnel access should be provided. Waste Storage, Removal, and Disposal. A large amount of waste material, including soiled bedding, general trash, and animal carcasses, is generated in animal facilities and needs to be removed from the facility without being transported through common corridors or elevators outside the facility. The preferred point of exit is a dock inside the animal facility. Soiled bedding typically makes up the bulk of the waste. A common disposal method is to dump it into a trash container, preferably inside a high-efficiency particulate air (HEPA) filtered bedding disposal cabinet (Figure€8.4), and then to transport it manually to a trash container outside the facility. Other more efficient methods used for disposing of soiled bedding include dumping it inside the soiled side of cage sanitation, either directly into the sanitary sewage system (Figure€8.5, local codes permitting) or into a vacuum system that transports it directly to a disposal container outside the facility. Similar vacuum systems may be used for transporting clean bedding into the facility. Such vacuum systems require dedicated space for the vacuum equipment that effectively contains the noise generated by the equipment. At one time, incinerators inside the animal facility or on site were commonly used to dispose of soiled bedding, animal carcasses, and certain hazardous wastes, but current environmental protection codes often preclude the use of incinerators. Today, most hazardous waste and animal carcasses are packed into special containers while still inside the animal facility and are then incinerated off site, typically by commercial disposal companies. Space needs to be provided to pack the containers safely and, as was noted previously, refrigerated and/or freezer space preferably near the dock is required to store the containers until they can be picked up for final disposal elsewhere.
84550.indb 153
11/11/10 6:24:35 PM
154
Handbook of Laboratory Animal Science
Figure 8.4â•…A color version of this figure follows page 336. A soiled bedding dump station on the soiled side of cage sanitation at the entrance to the tunnel washer. The bedding dump station protects the operator by using mass air displacement to draw soiled bedding aerosols away from the operator standing in front of the cabinet while dumping soiled bedding from a cage inside the cabinet. The air draws the dust into the back of the cabinet, where it is filtered out from the air as it is first passed through a coarse filter and then a HEPA filter before being returned to the room.
Figure 8.5â•…Photo taken on the soiled side of a cage sanitation facility illustrating a bedding dump/disposal station that disposes of bedding directly to the sanitary sewer system. To the left of the picture is the load end of a tunnel washer. To the near side of the disposal unit is a wall-mounted stainless steel sink typical of a type often recommended for use in animal rooms. The wall and floor finish is ceramic tile with epoxy grout.
84550.indb 154
11/11/10 6:24:36 PM
Laboratory Animal Facilities and Equipment
155
An alternative means for disposing of animal carcasses includes chemical digestion in specially designed equipment that prepares the carcasses for disposal through the sanitation sewage system. Animal Use Surgery. In most localities, dedicated space is required for conducting major survival surgical procedures on nonrodent mammalian species (ILAR 1996; AVMA 1993). Typically, this should be a surgical suite. The design of the surgery suite will depend on the species and number and complexity of procedures likely to be performed. In addition to operating rooms, the surgery suite should include space for preparation and storage of sterile supplies, surgeon preparation, animal preparation, immediate postsurgical recovery, and storage for equipment and supplies. Ideally, separate rooms would be used; at a minimum, it is essential to limit activities in the surgery room to those required to conduct the surgical procedure and to separate “clean” and “dirty” activities. Depending on the size of the surgery suite, office space for veterinary technicians and veterinarians may be required. Ideally, the surgery suite should be located near where the nonrodent mammalian species are likely to be housed and arranged to preclude unnecessary traffic through it. See Hessler (1991b) and Howard and Foucher (2009) for detailed samples of program descriptions for a surgery facility. Standards for conducting survival surgical procedures on rodents may be less stringent in some localities and a dedicated space may not be required. Even then, aseptic procedures are required. If it is anticipated that a large number of rodent surgical procedures will be conducted, a surgical room located near the rodent housing areas should be provided. It can be, but need not necessarily be, part of the surgery suite or even be dedicated to this purpose, but it should be designed so that it can be readily sanitized prior to use as a surgery room. Back-draft work stations make for good rodent surgery tables when volatile anesthetics are used. See Cunliff-Beamer (1993) and Brown (1994) for a description of surgical facilities and management procedures for rodents. Diagnostic Laboratories and Necropsy. Diagnostic laboratory facilities are an essential component of an adequate veterinary care program. The size and complexity of the laboratory space may vary from a simple wet laboratory used to process samples for delivery to an external comprehensive diagnostic laboratory, to an internal comprehensive diagnostic laboratory, or anything in between. It is efficient and convenient for diagnostic laboratory space to be immediately adjacent to or a part of the administrative and training suite. A necropsy laboratory is required in most facilities to support both veterinary and investigative personnel. Ideally, the necropsy area should be located in a relatively isolated area that is adjacent to refrigerated space that can be used to store animal carcasses. Imaging and Special Research Support Facilities. An adequate veterinary care program may require imaging equipment, including x-ray and ultrasound machines. Diagnostic imaging for larger animals is typically located inside the surgery suite. Examples of imaging techniques and equipment that are rapidly becoming essential research tools for supporting contemporary research, especially for protocols that involve genetically modified rodents, include MRI, CT scanners, PET and micro-PET scanners, ultrasound, fluorescence and bioluminescence in vivo imagers, and confocal microscopes. If such equipment is to be used with animals housed in a barrier facility, consideration should be given to including space for the equipment inside the barrier because many rodent barriers are managed so that animals are not returned once they have been removed. An even more useful arrangement would be to locate the imaging suite so that it can be directly accessible from both inside and outside the barrier. If properly managed, this arrangement would increase access to the equipment without compromising the barrier. In addition to imaging equipment, a whole-body irradiator for small rodents is a frequently required research tool that should be located either inside the rodent barrier or, even better, arranged so it can be accessed from both inside and outside the barrier. Animal Procedure Laboratories. Research animal facilities have become increasingly more active extensions of the research laboratory, with most, if not all, survival animal procedures being conducted inside the animal facility. The primary drivers for this change are human health issues raised
84550.indb 155
11/11/10 6:24:36 PM
156
Handbook of Laboratory Animal Science
by taking the animals out of the facility and animal health issues raised by returning animals to the animal facility. Other concerns relate to security and public relations issues and to the biological effects that the transport and change of environment have on the animal, potentially confounding the research data derived from the animals. For these reasons and others, an increasing percentage of animal facility space is being devoted to animal procedure space. For most nonrodent species, shared animal procedure laboratories are useful. A procedure room for every four to eight animal rooms works well. These rooms are equipped with procedure lights and examination tables and other amenities that facilitate performing animal procedures. Shared procedure rooms for use with “clean” mice and rats are not advised because of the increased potential for spreading infectious agents from room to room. For these animals, most procedures are performed in the animal room or in dedicated animal procedure space adjacent to each room housing rodents. In rooms housing rodents in microisolation cages, animal procedures are performed on mobile clean benches or in biosafety cabinets, referred to here as “animal transfer cabinets” that are primarily used for changing cages. Even though most routine procedures on rodents can be performed in the animal transfer cabinets in the animal room, rodent facilities require a considerable amount of procedure space for performing more complex procedures than can practically be performed in the animal room, including ones that involve extensive equipment. Examples of animal procedure space that may be required in a rodent barrier facility include laboratories for surgery, diagnostic and experimental imaging, whole body irradiation, and transgenic and knockout (TG/KO) animal procedures. See Hessler (1991b) and Howard and Foucher (2009) for detailed sample program descriptions and layouts of various types of animal procedure space. Animal Housing Areas General Animal Housing Concepts Types and Sizes of Animal Rooms. Basically, animal rooms can be divided into two types: (1) rooms for housing animals using dry bedding cage systems, generally for housing small animals from rodents to rabbits; and (2) rooms for housing animals using hose-down caging systems, generally for housing nonhuman primates, canines, and small agriculture animals in cages or floor pens. One design approach is to design all animal rooms to accommodate either type of housing system; another is to design rooms for either one system or the other. The hose-down caging system requires floors sloping to floor drains, preferably in troughs, and the presence of a hose, preferably on a hose reel. The dry bedding housing system does not necessarily require floor drains or sloped floors. The obvious advantage of designing for both types of housing systems is maximum flexibility. The disadvantage is that a room designed for both types of housing systems is not optimal for either. If it is known with a reasonable degree of certainty that only dry bedding housing systems will be used in the facility (e.g., a rodent barrier facility), then a reasonable choice is to design it with no floor drains, although some who prefer to routinely hose down the floor even in rodent rooms would disagree with this. The size and shape of the animal room can vary depending on many factors, including the species to be housed, the type of housing system to be used, and the arrangement of the cages and racks in the room. There is no one best or ideal size, but it is important to decide on the cage type to be used as well as the layout of cages and/ or cage racks in the room prior to deciding on sizes and shapes of the animal rooms. For example, double-sided rodent racks are typically arranged library style with multiple racks parked parallel to one another, with the end of each parked rack against a common wall, or two rows on opposite walls with an aisle between them. Single-sided cage racks are typically parked with the back of the rack against a wall. A combination of both types combines the advantages of both. For
84550.indb 156
11/11/10 6:24:36 PM
Laboratory Animal Facilities and Equipment
157
(a)
(b)
Figure 8.6â•…(a) Two animal cubicles side by side. Immediately across the 1.52 m (5 ft) aisle are two identical cubicles (not shown). Note the lights in each back corner of the cubicles, the low air returns, and the pair of hinged doors with bumper guards. (b) Two facing animal cubicles with vertical stacking, three-panel sash doors.
example, a room with both types may have single-sided racks placed against both side walls with double racks placed end to end in the center of the room, forming two aisles between facing cage racks. Animal Cubicles. This is an animal room concept that provides maximum flexibility for animal isolation within minimal space by dividing animal rooms into multiple small spaces, with each typically large enough to hold one cage rack or, occasionally, two cage racks (Figures€8.6a and 8.6b). Cubicles help solve the problem of what to do when a facility has plenty of animal housing space, but too few spaces to provide the necessary separation of animals based on species, source, microbiological status, project, and/or experimental hazards. First described by Dolowy in 1961, animal cubicles have been used extensively since then, especially for specialized housing areas where
84550.indb 157
11/11/10 6:24:40 PM
158
Handbook of Laboratory Animal Science
isolation of small groups of animals and containment of hazardous or potentially hazardous agents are priorities (e.g., quarantine, biocontainment, and chemical and radioisotope containment areas). Animal cubicles typically have three solid sides, with the fourth side composed of full panel glass doors—either vertical telescoping doors or a pair of conventional hinged doors. The most common cubicle size is approximately 1.2 m deep by 1.8 m wide, although larger cubicles (e.g., 2.1 m × 2.1 m) that can hold two racks and in which a person could perform simple tasks with the doors closed are useful. The size of the room depends on the size and number of cubicles. It is recommended that the aisle between facing cubicles be maintained at a minimal width of 1.5 m. Typically, animal cubicles are used to house smaller animals in cages on mobile cage racks using dry bedding caging systems, but the concept can be applied to housing large animals requiring hose-down caging systems as well (Hessler and Britz 2009; Hessler 1991c, 1993). Extensive experience over many years suggests that cubicles effectively prevent airborne infectious agents from spreading between cubicles in the same room. This containment effectiveness is probably related to the brief window of opportunity for cross-contamination when a cubicle door is open and substantial dilution of the contaminant with large volumes of air ventilating the aisle and cubicles. The usefulness of animal cubicles has decreased with the advent of microisolation cages for rodents; however, cubicles continue to be useful for conventional housing of rodents and other species. Animal cubicles can be built in place or commercially prefabricated. Prefabricated cubicles typically come complete with lighting and internal ventilation, with and without HEPA filtration, and the ability to switch between positive and negative relative air pressures. The many options regarding architectural and engineering features for animal cubicles and animal cubicle rooms, along with some of their pros and cons, are described in detail in Hessler and Britz 2009; Hessler 1991c, 1993; Ruys 1988; and Curry et al. 1998. Conventional Animal Housing In this context, “conventional” is a generic term with no specific definition. It refers to almost any type of laboratory animal housing facility, area within a facility, or animal room that is not specially designated otherwise (e.g., as a barrier or containment). All animal rooms, whether conventional, barrier, or containment, should be designed for ease of cleaning and have minimal built-ins. Often, a sink is all that is required. As noted previously, floor slopes and plumbing features such as floor drains and hose reels may or may not be required, depending on the species to be housed and the housing and sanitation system to be used. Barrier Animal Housing In the jargon of laboratory animal science, a “barrier facility” has come to be known as an animal housing system designed and managed to protect animals from undesirable microbes. In other words, “barrier” equates to “keep out.” Until recently, the primary use of barriers was for the production of laboratory rodents; however, the need to maintain a similar level of barrier housing in the research environment has extended the need for barrier housing to the research facility. This need has expanded further with the extensive use of immune-compromised animals and genetically modified rodents. The “barrier” may be at the cage level, the room level, the level of an area within a facility, or encompass the entire facility. For example, it is common to create a barrier in a conventional animal room using various types of cages and equipment, including static microisolation caging systems, ventilated microisolation caging systems (in which air filtered through HEPA filters is delivered to each cage on the rack), and flexible film isolators of the type used for maintaining germ-free animals. All of these approaches work reasonably well, but are much more labor intensive to manage than a barrier designed as an area within an animal facility or as an entire facility. The primary difference is that, with the room level barrier system, the cages and supplies are wrapped and autoclaved
84550.indb 158
11/11/10 6:24:40 PM
Laboratory Animal Facilities and Equipment
159
elsewhere in the facility before being transported to the animal rooms. In a “barrier” facility or “barrier” area within an animal facility, the cages and supplies, after being cleaned and set up outside the barrier, are autoclaved into the barrier area; once inside the barrier, they are handled in a conventional manner, thus eliminating the need for wrapping and unwrapping. Large rodent barriers may include a cage sanitation facility inside the barrier. This offers the option of relying on the level of sanitation provided by the cage washing equipment, operating at a minimum temperature of 82.2°C, and not routinely autoclaving cages unless there is a disease outbreak. In the case of an outbreak, having an autoclave with the capacity to autoclave all the cages used in the barrier is best. A good case can be made for autoclaving all the cages inside the barrier, even when the cage sanitation facility is inside the barrier. The use of irradiated feed and bedding also eliminates the need for autoclaving them into the barrier. Irradiated feed and bedding are often double bagged; the outer bag is removed and the outside of the inner bag is chemically sanitized prior to introducing it into the barrier. Barrier facilities are designed and managed at various levels of microbiological control, affecting the degree of control over the ways in which supplies and personnel enter the facility. The highest level barrier facilities may have one or more double-door, pass-through autoclaves (preferably pit-mounted, floor-loading bulk autoclaves) and one or more ventilated entry and exit vestibules with interlocking doors. Packaged sterile supplies and animals in filtered containers can be passed into the barrier through these vestibules, after having the exterior surface of the package chemically sanitized. Soiled equipment and trash can be passed out of the barrier through the vestibules. In certain circumstances, a pass-through dip tank, filled with high-level disinfectants, may be used to pass sterile items packaged in watertight containers into the barrier. Personnel may be required to shower and change clothing prior to entering the barrier, but more typically, at least in research barrier facilities, personnel enter through a vestibule with interlocking doors where they put on sterile outer garments over street clothes or uniforms, along with head and shoe covers, a face mask, and gloves. Air showers using mass quantities of HEPA-filtered air may be added to a personnel entry vestibule. See Hessler (2009b) for a detailed description of barrier housing facilities for maintaining rodents. Depending on the intended use of the barrier, space may be required inside the barrier for wet laboratories, animal procedure laboratories, TG/KO laboratories, specialized imaging equipment, irradiation equipment, etc. A research rodent barrier may require a quarantine area inside the barrier. This is especially important for a TG/KO facility because quarantine is recommended for all foster mothers coming out of the TG/KO laboratory until the young are weaned and the mother’s health status is determined. Animal cubicles are useful for this purpose, even if the animals are housed in microisolation cages. Containment Animal Housing “Containment” refers to animal housing systems designed and managed to prevent the escape of experimental hazardous agents to which the animals have been exposed. The objective is to protect workers, other animals, and the general environment. In other words, “containment” equates to “keep in.” The hazardous agents may be biological, chemical, or radiological. Like a “barrier,” “containment” can be achieved at the cage level, the room level, for an area within an animal facility, or at the level of the entire facility. When the various levels are used together, increasing levels of containment can be provided. Figure€8.7 is a schematic of a flexible facility designed for the containment of all three classes of hazardous agents. The design features of a containment facility are similar to those of a barrier facility. The most basic design and engineering features of a containment facility focus on providing effective isolation of the animals exposed to a hazardous agent (to facilitate containing the agent) and controlled or limited access to the containment area. At all levels of containment, the primary
84550.indb 159
11/11/10 6:24:40 PM
84550.indb 160
??
Inter Lock BO10
Corr. BOO6
Inter Lock BO12
Procedure Room BO13
Corr. BO14
BOO3
Animal Room
Shower BO15
HSKP BOO5
BOO4
Animal Room
Shower BO16
Animal Room BO17
Autoclave
Bedding Disposal BO18
Soiled Elevator Lobby BO20
Soiled BO19
??
Figure 8.7â•…Schematic drawing of a containment area. This containment area was designed to provide maximum flexibility. It consists of six animal rooms. One is a standard animal room (B004) and five (B001, B003, B007, B008, and B017) are animal cubicles rooms, each of which is divided into four animal cubicles, an area for changing cages and conducting animal procedures in a biosafety cabinet, and an area for a sink and storage of feed containers and sanitation equipment and supplies. There are three entry/exit vestibules with interlocking doors. One vestibule (B010) enters into corridor B006; along with animal rooms B001, B007, and B008, this can be isolated from the rest of the containment area to serve as an animal biosafety level 2 (ABSL-2) facility in which research staff enter and exit through vestibule B010. The second vestibule (B012) enters into corridor B014, which also may be entered from two private shower/locker rooms (B015 and B016). This is an ABSL-3 area. It includes three animal rooms (B003, B004, and B017), a laboratory (B013), a housekeeping closet (B005), and a bedding disposal room (B018). The bedding disposal room is for disposing of bedding soiled with hazardous chemicals or radioisotopes. The third vestibule is between the autoclave and the bedding disposal room. It enters the soiled side of the cage sanitation area so that cages contaminated with hazardous chemicals or radioisotopes can be taken directly to the cage and rack washer to be decontaminated without having to be transported through corridors. The bulk autoclave is large enough to hold two racks. Cages contaminated with biohazards are autoclaved out of the containment area. The door between the ABSL-2 and ABSL-3 area allows for the entire area to be operated as an ABSL-3 facility.
Start
Cage Maintenance BOO9
Animal Room BOO8
Animal Room BOO7
Animal Room BOOI
HSKP BOO2
160 Handbook of Laboratory Animal Science
11/11/10 6:24:44 PM
Laboratory Animal Facilities and Equipment
161
objective is to contain the hazardous agent as close to the source as possible, ideally, at the cage level (e.g., a microisolation cage). Animal cubicles are particularly well suited for use in containment facilities. Of course, the more levels of containment there are, the higher the safety level is. For example, when experimentally infected mice are housed in a microisolation cage in a conventional room, the cage provides the first level of containment and the room door the second level. If they were housed in a microisolation cage inside an animal cubicle in an animal cubicle room in a barrier area located inside a larger animal facility, there would be at least five levels of containment: the microisolation cage, the cubicle doors, the cubicle room door, and the two doors of the entry vestibule (Figure€8.7). It is important to provide appropriate laboratory and animal procedure space inside containment facilities to avoid having to remove live animals from the facility. See Lehner and colleagues (2009) for a detailed description of containment facilities. Biohazard Containment. Microbiological agents are classified into four biosafety levels (BSL-1 to BSL-4) according to the degree of risk to humans (classified by the CDC-NIH in the publication “Biosafety in Microbiological and Biomedical Laboratories,” USDHHS 2007). BSL-1 agents are considered to have very low or no pathogenicity for humans while BSL-4 agents pose the highest risk level; BSL-3 agents have greater potential for aerosol transmission than BSL-2. The CDCNIH publication describes combinations of laboratory practices and techniques, safety equipment, and facilities required for working with agents and animals in each classification level. When animals are infected with microbial agents, the corresponding facilities and management practices are referred to as animal biosafety levels (ABSL) one to four. Animal studies with BSL-2 agents are relatively common and recently have become more so with the use of viral vectors for gene therapy studies, even if they are referred to as being “replication deficient” (Evans and Lesnaw 1999; Webber and William 1999). Animal studies with BSL-3 agents are less common than BSL-2 agents. Studies with BSL-4 agents are very special and rare. Studies with BSL-2 agents can be conducted in conventional animal rooms using appropriate equipment and ABSL-2 practices; however, they are more efficiently and consistently conducted at a higher level of safety in an ABSL-3 facility. This is primarily due to the fact that contaminated cages, supplies, and wastes are autoclaved directly out of the ABSL-3 facility, eliminating the time-consuming and potentially hazardous practice of bagging them before transporting them out of the facility to a remote autoclave. In addition, an ABSL-3 facility is highly desirable for quarantine of rodents infected with adventitious agents or that are of unknown health status. These agents are not hazardous to humans but have the potential to be devastating for many if not most of the rodent studies in the facility. ABSL-2 is the highest level of biocontainment that can practically be achieved in a conventional animal room with appropriate equipment and management practices. An ABSL-3 facility has all the design features of a high-level barrier facility as described previously. In fact, infectious containment facilities are often managed as both a barrier and containment facility, in that cages and supplies are autoclaved in and soiled cages and wastes are autoclaved out. ABSL-3 facilities should have ventilated entry and exit vestibules with interlocking doors, a pass-through autoclave as an integral part of the containment perimeter, and a hand-washing sink in each animal room. In addition, a number of design features are required to facilitate keeping agents in, such as an effective sealed envelope around each room and around the entire facility except for the doors (gasketed doors are not generally required) and air balancing that directs the movement of air from the least contaminated areas to the most contaminated areas. HEPA filtering of exhaust air is not required for ABSL-3 facilities, but is highly recommended, not only because it increases the degree of safety, but also because it helps allay public concerns about the existence of the facility in their neighborhood. In addition, standards may change to require HEPA filtration of exhaust air for some BSL-3 agents. Exhaust air systems should provide 100% redundancy and be on emergency power to assure uninterrupted exhaust and filters should be the bag-in, bag-out type to facilitate safe replacement of contaminated filters. More details regarding animal biosafety facilities and practices can be found in
84550.indb 161
11/11/10 6:24:44 PM
162
Handbook of Laboratory Animal Science
Lehner et al. (2009); USDHHS (2007); Barkley (1997); Barkley and Richardson (1984); Richmond (1991, 1996); Hessler (1995); White (1996); and Hessler et al. (1999a, 1999b). Chemical and Radioisotope Containment As with biohazard containment, appropriate equipment and management practices are critical, but the physical characteristics of the facility can influence the level of safety that can be attained and the consistency at which it can be maintained. Work with chemicals and radioisotopes in animals may often be carried out safely in conventional animal rooms; however, there are some exceptions. For example, HEPA filtering of exhaust air may be required for working with concentrated levels of especially potent carcinogens, or special shielding may be required for working with certain radioisotopes. It is essential to isolate such studies to help prevent cross-contamination. When small numbers of animals are required, it is inefficient to use an entire animal room for a single study. Animal cubicles, semiridged isolators, microisolation cages, etc. can provide the isolation necessary to prevent cross-contamination while housing multiple studies involving small numbers of animals within a relatively small area, as compared with using conventional animal rooms for each study. An area planned for supporting chemical and radioisotope studies may utilize one or more rooms with cubicles or be equipped with other containment equipment, along with one or more procedure rooms equipped with radioisotope and chemical fume hoods. Decontamination of cages can usually be accomplished safely with the use of conventional mechanical cage washers, especially cage and rack washers, taking advantage of the dilution factor that occurs due to the large volume of water used by the washers. Ideally, the chemical and radioisotope containment area should be near the dirty side of the cage sanitation area to minimize the need to transport contaminated cages through corridors (Figure€8.7). Also recommended is a separate room, where contaminated bedding can be removed from cages or pans inside a laminar air flow cabinet in which the aerosolized contaminant is drawn away from the operator into a HEPA filter (Figure€8.4). Quarantine Most laboratory animals used today are purposely bred using disease control measures equal to or superior to those in the research facility; therefore, most research facilities do not require special quarantine for the vast majority of the animals received into the facility. However, there typically are exceptions. One common exception is the result of the increased use of genetically modified rodents and the sharing of these unique animals (usually mice) between research institutions. Most facilities today require a high-level rodent quarantine facility for holding animals until they can be documented to be “clean” or the genetic line can be rederived by C-section and fostering or, preferably, by embryo transfer. Ideally, rodents of unknown health status or animals known to be infected with agents hazardous to other rodents in the facility are best maintained in an ABSL-3 facility. At a minimum, the rodent quarantine facility should be well isolated from other rodent housing areas. Animal cubicles are a good option for quarantine areas, even when used in conjunction with microisolation cages. Housing for Nonhuman Primates Housing for nonhuman primates can be considered somewhere between conventional and biohazard containment because of their potential for carrying zoonotic diseases. For this reason, the ideal arrangement is to house them in an isolated area under ABSL-2 standards. At a minimum, rooms housing nonhuman primates should be arranged and located to avoid the necessity of transporting animals or cages and equipment soiled by the animals through corridors or on elevators outside the animal facility. The objective is to avoid exposing individuals who do not have an occupational requirement to be exposed to nonhuman primates.
84550.indb 162
11/11/10 6:24:44 PM
Laboratory Animal Facilities and Equipment
163
Some species can be very noisy, so it is recommended to take this into consideration when designing the location of nonhuman primate rooms. Rooms housing nonhuman primates require plumbing to facilitate hose-down room and cage-sanitation, even if litter pans with dry absorbent materials are to be used. Other special considerations for a nonhuman primate housing area or room may include additional entry security, and an entry vestibule for each animal room, typically made of chain-link fencing inside the room, that prevents animals that get out of their primary enclosure from escaping when the room door is opened. Lights and any other fixtures in the animal room must be mounted so that animals free in the room cannot damage them and so they do not impede capturing the animals. Housing for Canines, Swine, and Small Ungulates In laboratory animal science parlance, canines, swine, and small ungulates (sheep and goats) are often referred to as “large animals.” Like housing for nonhuman primates, rooms for housing these large animal species require plumbing to facilitate hose-down sanitation. Preferably, housing for these species is isolated from other animal housing and human occupancy areas because of their relatively “dirty” microbial status as compared with rodents and the fact that some species (e.g., dogs and swine) generate high noise levels. This area may be considered a candidate for being an isolated area of the facility with controlled access. Animal procedure space should be provided in this area. Because these animal species are commonly used as surgical research models, it is desirable to house them near the surgical suite. Rooms may be provided for postoperative recovery and intensive care of surgical patients. Generally, these species are housed in mobile double-tiered cages, mobile single-tiered pens, or fixed-floor pens (Figures€8.8–8.10). Even when dry bedding systems are used to house these species, routine cage and room sanitation still requires floors sloped to floor drain troughs. With these species, the zoonotic disease concern, while present, is not as great as with nonhuman primates; however, sheep and goats are an exception because of concerns for Q fever, especially in connection with pregnant sheep. For this reason, pregnant sheep and goats are often maintained under ABSL-2 standards.
Figure 8.8â•…Dogs that are being held individually when fed. This strategy ensures that dogs housed in groups for most of the day may be fed without disturbances from fellow dogs. It also allows caretakers to identify if some dogs have a decreased appetite.
84550.indb 163
11/11/10 6:24:45 PM
164
Handbook of Laboratory Animal Science
Figure 8.9â•…Indoor pen for dogs. Wood shavings are spread on the floor to facilitate cleaning. Note the door on the right side of the picture connecting the indoor with the outdoor pen.
Figure 8.10â•…Given a large area, dogs and pigs may be held together. Outdoor pens such as the one shown here stimulate the animals to exercise and to engage in social interactions.
84550.indb 164
11/11/10 6:24:46 PM
Laboratory Animal Facilities and Equipment
165
Architectural Features The primary focus for the following architectural features, especially the interior surface features, is to create a durable surface that is easy to maintain and sanitize capable of withstanding scrubbing, chemical cleaning and disinfecting agents, and impact from high-pressure water. All surface junctions, penetrations, and wall-mounted apparatus should be caulked and sealed to facilitate air balancing and vermin control. Animal facility interior surfaces are exposed to much abuse in the normal conduct of animal care and use. Selections made with an eye toward “saving” money on architectural features rarely prove to be wise and could easily cost many times more in long-term maintenance costs than the initial cost “savings.” It is also true that “expensive” does not necessarily guarantee a satisfactory performance. Interior Surfaces Interior surfaces are definitely not a place to cut costs. All interior surfaces should be durable; impact resistant; moisture proof; highly resistant to cleaning agents, scrubbing, hot water, and high-pressure sprays; and relatively smooth, free of pits, and pinholes to facilitate efficient and hygienic operations. Seamless applications are most desirable. Surface materials should repair easily and safely under ongoing facility operations. Materials that do not give off gas volatile organic compounds or other hazardous air pollutants should be used whenever possible. See Leverage and Roberts (2009) for detailed information regarding interior finishes. All interior surface penetrations as well as floor-to-wall, wall-to-wall, and wall-to-ceiling junctions in animal room surfaces should be sealed to eliminate access for vermin and to facilitate balancing relative air pressures. Floors Floors should be monolithic and slip resistant even when wet, yet relatively smooth and easy to sanitize. Commonly used flooring materials include troweled on or broadcast composite resinous floors (typically, resins include epoxy, urethane, methyl-methacrylates, and, more recently, ultraviolet light cured materials) ranging in thickness from 3 to 6 mm ( 81 to 14 in). Many floor coverings work well in a rodent room, including resinous coverings and sheet vinyl with sealed seams; however, caution is required since double-sided, high-density ventilated racks may be heavy enough to dent soft floors. Few flooring materials consistently hold up when challenged with the high traffic, heavy loads, physical abuse, and high moisture that floors are subject to in a cage sanitation area. Ceramic tile installed with a nonporous, chemical- and moisture-resistant grout (Figure€8.5) has proven to be a relatively maintenance-free floor for cage sanitation areas where seamless composite resinous floors have a long history of too often failing, especially around drains and other resin-to-metal junctions. Grouted tile floors are not suitable for corridors or animal rooms because the joints cause excessive noise when cage racks are rolled across it. There should be a minimum 10 cm (4 in.) high, 1.3 cm ( 12 in.) radially coved base to form a watertight seal at the floor-to-wall junction and facilitate sanitation. The most important part of any flooring installation using material that is intended to bond to the concrete base is the quality of installation and the attention paid to proper preparation of the base and studious attention to the amount of vapor transmissions out of the concrete. Too high a level of vapor transmission at the time of installation may cause the flooring material to delaminate. See Leverage and Roberts (2009) for detailed information regarding flooring materials and technology, and an explanation of the vapor transmission issue.
84550.indb 165
11/11/10 6:24:47 PM
166
Handbook of Laboratory Animal Science
Walls Historically, the most commonly used wall material in the United States has been concrete masonry units (CMU) coated with block filler and then sealed with two epoxy or urethane coatings with a goal to create a relatively smooth surface without pits and pin holes, a goal too often not met. This wall performs well in most areas of the facility, but is more likely to fail in high-moisture areas, such as animal rooms (in which hose-down caging systems are used) and cage sanitation areas (where, over time, the coating too often peels from the block). Structural glazed facing blocks with nonporous, chemical- and moisture-resistant grout provide a maintenance-free wall that performs exceptionally well in these high-moisture areas, as do fiberglass-reinforced panels (FRP) laminated onto masonry. Studded walls with common gypsum wallboard, water resistant or not, rarely prove suitable for animal care or use areas of an animal facility; however, the availability of more durable and waterresistant types of wallboard panels has made metal studded hollow walls a viable option. These walls are often finished with wet lay up, resin-filled fiberglass mat or laminated with fiberglass- reinforced panels that are approximately 24 mm ( 323 in.) thick. More recently, fiberglass panels fabricated offsite and then applied to metal stud walls in lieu of gypsum or other hard boards have proven to be very durable. These walls are approximately 13 to 16 mm ( 12 to 85 in.) thick and offer the advantage of being significantly smoother than masonry block walls. They also do not require painting, either at the time of construction or during downstream maintenance, and can reduce construction time and coordination efforts. In earthquake-prone locations, CMU walls may not meet code, in which case metal studded walls with appropriate wall panels are used. Protective guardrails or wall curbs are required in corridors and guardrails may also be cost effective in the cage sanitation area, animal rooms, and other areas where wall damage from caging and other mobile equipment is likely. Guardrails should withstand sanitizing, be sturdy, and be constructed to avoid providing harborage for cockroaches and other vermin. Extruded solid aluminum rails fastened to the wall with I-beam standoffs have proven very useful in animal facilities (Figure€8.11), but aluminum can oxidize when exposed to certain chemicals. Stainless steel rails are more expensive, but they do withstand more aggressive chemical exposure. Guardrail height should be carefully matched to the equipment used in the facility. A double row of guardrails is sometimes provided in corridors; however, one is generally effective if its height from the floor is carefully considered to protect the wall from all rolling equipment to be routinely used in the facility. Generally, this turns out to be a 7.6 to 10 cm (3 to 4 in.) wide rail centered 25 to 30 cm (10 to 12 in.) from the floor. Ceilings Gypsum board ceilings sealed with a resinous coating are adequate for relatively dry areas of the facility, including rodent rooms, but are generally not suitable for high-moisture areas like cage sanitation or animal rooms where hose-down sanitation programs are routinely used. A drop ceiling with lay-in panels is generally not recommended for animal housing rooms because they impede sanitation and vermin control. However, in recent years, composite panels made of lightweight, water-impervious materials sealed to fiberglass “T” bars with gaskets and clamps have proved to be a satisfactory, virtually maintenance-free choice for ceilings. These ceilings are particularly cost effective for use in highmoisture areas such as cage sanitation and animal rooms with hose-down type animal housing systems. In all cases, the ceiling-to-wall junction should be sealed. The minimal recommended ceiling height is 2.7 m (9 ft) and may need to be higher in rodent and nonhuman primate rooms, depending on the height of rodent racks or nonhuman primate cages to be used.
84550.indb 166
11/11/10 6:24:47 PM
Laboratory Animal Facilities and Equipment
167
Figure 8.11â•…A corridor in an animal facility showing extruded solid aluminum bumper guardrails at two levels to protect the walls and the door. If only one guardrail is to be used, the lower one may provide the most protection from the widest variety of equipment.
Doors and Hardware Doors and hardware are a critical component of the planning and design process (MolloChristensen 2009). The minimum animal room door size should be 107 cm (42 in.) wide by 2.1 m (7 ft) high; however, nonhuman primate cages and ventilated high-density rodent cage racks may require wider and higher openings. Doors measuring 122 cm (48 in.) wide by 2.4 m (8 ft) high frequently prove useful for animal rooms. If 2.4 m high doors are provided for animal rooms, it is important to make certain that all doors in the facility through which the higher cage racks will be transported are also at least 2.4 m high. This includes all corridor doors, doors in and out of the cage sanitation area, the cage rack washer doors, dock doors, and elevator doors, if applicable. Stainless steel or fiberglass-reinforced polyester door frames in the long run are a more cost-effective choice than less durable high-maintenance painted steel frames. Door frames should have hospital stops to facilitate floor cleaning. Jamb guards may be mounted on the corridor side. There must be no doorsill at any interior doors because this seriously impedes the movement of cage racks through the door. Like the frames, stainless steel or fiberglass-reinforced polyester doors prove more cost effective than less durable and higher maintenance materials, including painted hollow metal doors. The doors should be sealed and have flush finished tops and bottoms. If the doors are not SS or fiberglass, they should be outfitted with stainless steel kick plates on both sides and edge guards on the strike side. Automatic drop bottoms should be surface mounted on the animal room side of the door, leaving no gaps larger than ¼ in. A view panel is highly desirable, if not essential, for security and personnel safety. Size and shape of the view panel are a matter of choice, but it should provide a clear view of the room from the corridor.
84550.indb 167
11/11/10 6:24:48 PM
168
Handbook of Laboratory Animal Science
To maintain light and dark cycles, the view panel should be glazed with red tinted glass carefully specked to block out light in the visible range for rodents while allowing enough light transmission in the red range to provide visibility for humans (e.g., “rose-chocolate 3” from Solar Graphics or “vivarium red” from Aegis Applied Films). If red-tinted glass is not used, the view panels must be covered. Cover options include a variety of solid blackout view panel coverings attached with magnets or hinged and latched window covers, most of which are inconvenient and high maintenance. Hinges should be stainless steel, heavy duty, standard, or continuous. Swing-clear hinges can be used to optimize door width. Hospital, lever-type door openers are a good choice. Other door hardware options include heavy-duty automatic door closers with variable delays and hold opens, push and pull plates mounted on both sides of the door, and guard rails the width of the door, similar to the ones on the wall, designed and placed to protect the door and the door handle. If fire codes permit, it may be preferable to eliminate the latch. If access to the animal room is controlled via a security system, magnetic locks are cleaner and generally found to require less maintenance than electric strikes. Door seals of various types may be required to control air movement around the door to facilitate balancing the ventilation system. Automatic sliding or hinged doors with full panel glazing should be provided in doorways with a high traffic of rolling stock, such as cage sanitation, the loading dock receiving and shipping area, and selected corridor doors. Depending on the situation, they may be opened with sensors that detect movement or with wall-mounted push plates or ceiling-mounted pull chords. Vermin Control Careful planning and construction will go a long way toward facilitating the control of vermin and insects without the use of organic insecticides and baits, especially wild or escaped rodents and cockroaches. Organic insecticides and baits should not be used in research animal facilities because they have the potential to change biological baselines and alter the animal’s response to experimental variables. The basic control approach is to seal vermin and insects out of the facility and eliminate hiding and nesting places within the facility. All cracks, joints, utility penetrations, lights, wall switches, communication, and power outlets must be sealed. Animal rooms should have a minimal amount of “built-ins” consisting of little more than a paper towel dispenser, utility hangers, and possibly a sink. These should be sealed to the wall or mounted away from the wall to eliminate hiding places and allow cleaning between the wall and the mounted item. Animal rooms should not have casework. Casework for animal procedure rooms and other laboratory spaces in the animal facility should be wall mounted and of an open design to reduce hiding places under and in back of them and to facilitate cleaning. Boxed-in casework should be avoided. The control of cockroaches and vermin starts during construction by keeping the construction site free of the garbage on which they feed. This requires having a zero tolerance for eating or drinking in the facility during construction. In addition, all hollow dead spaces in the facility, including inside concrete blocks and studded walls, should be treated with amorphous silica dust to preclude the harborage of cockroaches. There should be high-pressure sodium (not mercury vapor) lamps or dichrome yellow (not incandescent flood) lamps located at exterior doors or vents to reduce the influx of vermin and insects into the facility. Air curtains with a velocity of 490 m (1600 ft) per minute can help reduce the influx of flying insects at frequently used exterior entrances that may be open for extended periods of time, such as loading dock doors. Noise Control Noise is another potential variable in the animal’s environment that can also be stressful for the staff. The primary noise producers are the cage sanitation area, and canine, porcine, and potentially, depending on the species, nonhuman primate rooms. Design features such as strategically locating
84550.indb 168
11/11/10 6:24:48 PM
Laboratory Animal Facilities and Equipment
169
these areas to buffer them from other animal rooms and human occupancy areas, along with architectural measures that reduce sound transmission should be carefully considered, including doubleentry doors, soundproof walls, locating corridors and support areas around the noise-generating areas, and locating the noise-generating areas next to outside walls or mechanical spaces. Conventional acoustical materials on animal room surfaces impede sanitation and vermin control and should be avoided; however, sound-attenuating panels that can easily be removed, washed, and sanitized in mechanical cage washers are commercially available and should be considered for use in especially noisy areas of the facility (Carlton 2002). All in-room activities, including cage changing, must be conducted in a manner that generates as little noise as possible. Background noise (e.g., soft music) can help buffer unavoidable noise inherent in routine care and use procedures. Other common sources of avoidable excessive noise include improperly sized ventilation ducts and outlets, improper air balancing that produces whistling around the room door, and improperly sealed room penetrations that also result in whistling. Vacuum equipment and the conduit used to transport bedding generate a large amount of noise and should be isolated or insulated or both to assure adequate sound attenuation. Fire alarms selected for animal housing areas should disturb the animals as little as possible. Most rodent species cannot hear frequencies below 1000 kHz, although guinea pigs are capable of hearing down to 200 kHz. Fire alarms that operate between 400 and 500 kHz should be used in facilities that house rodents. Engineering Features Heating, Ventilation, and Air Conditioning (HVAC) The function of the HVAC system is to control the laboratory animal’s macroenvironment (the room) and microenvironment (the cage) and to maintain a healthy work environment for personnel. The HVAC system must supply clean air to the animal rooms and maintain a consistent temperature and humidity throughout the year, while effectively removing excessive heat, particulate, and gaseous contaminants generated in animal rooms. Many of the animal facility planning- and design-related references cited earlier in this chapter cover HVAC systems to some degree (Lipman 2007; USPH 2009; Canadian Council on Animal Care 2003; Hessler and Leary 2002; Hessler and Höglund 2002; Shalev 2001; Ruys 1991), and the book Planning and Designing Research Animal Facilities (Hessler and Lehner 2009) includes a complete chapter (Hessler and Frasier 2009) describing special considerations for animal facility HVAC systems. The American Society of Heating, Refrigerating, and Air Conditioning Engineers, Inc. recognized the unique design requirements of HVAC systems for research animal facilities and included a separate section, “Laboratory Animal Rooms,” in the ASHRAE handbook of HVAC applications (ASHRAE 2007). Air Quality The quality of air delivered to the facility is determined to a large extent by the source of the air and the degree of filtration. The source of the supply air must be selected to avoid contamination with exhaust air from other buildings or the same building, especially the animal faculty, incinerator smokestacks, vehicle exhaust fumes, etc. The quality of filters used for filtering incoming air varies from 85 to 99.97% efficiency HEPA (high-efficiency particulate air) filters, depending on the type of facility or area of the facility. For example, the air delivered to rodent barrier facilities and surgery rooms may be HEPA filtered, while the air to other areas of the facility may be filtered with 95% efficient filters. The need for HEPA-filtered air, even in rodent barrier facilities, is not well documented, and its cost effectiveness is questionable. Task-directed HEPA filtering (e.g., using HEPA filters on ventilated racks and in cage change cabinets) is probably not only more cost effective than HEPA-filtering all the air coming into the facility but also more effective.
84550.indb 169
11/11/10 6:24:48 PM
170
Handbook of Laboratory Animal Science
Another use of task-directed HEPA-filter air is mass air displacement (MAD) clean rooms, which are similar to but typically at a lower quality than that used in electronic fabrication plants. In MAD rooms, air is recirculated within the room through HEPA filters at volumes sufficient to change the air 150 (most common) to 600 times per hour, depending on the type of system and clean room class desired (Hessler and Moreland 1984). Fresh air exchanges are superimposed over the recirculated air at a rate similar to that in a conventional room. MAD rooms effectively control the animal’s airborne microbial environment, thereby reducing cross-contamination; they may be “hard wall” or “soft wall” units, with the size of rooms large enough to house multiple cage racks or soft wall units just large enough to house a single cage rack. MAD rooms for animal housing were considered state of the art during the 1970s and 1980s, but declined in popularity with the advent of individually ventilated rodent cages. More recently, there has been an increased interest in using multiple soft wall units with HEPA-filtered air in large, open-warehouse type spaces to gain high-quality animal housing space with maximum flexibility at minimal cost. Ventilation Animal facilities must have dedicated supply and exhaust air handling units. The supply air must be 100% outside makeup air. The ventilation rate that has proven effective for most animal rooms, expressed in terms of fresh air changes per hour (cph), is around 15 cph. However, this varies between 10 and 20 cph, depending on the heat load as well as microbial, particulate, and gaseous contaminants generated in the room, which is dependent on the species and density of animals to be housed in the room. Control of the heat load in the room is the most critical concern because high temperatures are stressful for all animals and may be lethal for laboratory species (especially rodents), even at temperatures not normally dangerous for most species. The minimal lethal temperature for laboratory rodents depends on time and relative humidity, but may start at temperatures as low as 29.4°C (85°F). It is important to note that the temperature in the animal’s microenvironment inside microisolation caging can be several degrees higher than in the macroenvironment. The prominent gaseous contaminant is ammonia, which is generated by urease-positive bacteria from the feces splitting each urea molecule from urine to form two ammonia molecules. Ammonia production depends on many factors, including the species, intestinal microflora, density of animals, the sanitation level, and the relative humidity in the room and cage. As a general rule applicable primarily to rodents, a ventilation rate that adequately controls the heat load when air is delivered to the room at 12.8°C (55°F) is adequate to control the gaseous and particulate contaminants. Heat loads for various species of animals are listed in the ASHRAE handbook (ASHRAE 2007). The ventilation rates noted before are not necessarily to be considered absolutes. Variable air volume (VAV) ventilation systems in which ventilation volume is based on actual heat load may achieve the objective while conserving energy. This would be consistent with the idea of using performance standards as noted in the Guide (ILAR 1996), as opposed to inflexible engineering standards. The same applies to other rooms in the facility (e.g., the cage sanitation area, where loads range from very high when the sanitation equipment is being used to very low when it is not). If VAV is to be used, consideration must be given to how varying the volume may alter the ventilation efficiency in terms of distribution of fresh air in the room and removing particulates, including allergens and infectious agents. Room ventilation patterns with regard to the location and type of supply diffusers and location of return and exhaust grills significantly affect the room ventilation efficiency; however, the most efficient pattern has yet to be defined definitively. The dogma for many years has been to supply high (typically, from the ceiling down the center of the room) and exhaust low near the floor, preferably in all four corners. This dogma has been called into question by the result of some computational fluid dynamic (CFD) studies but is supported by other studies.
84550.indb 170
11/11/10 6:24:48 PM
Laboratory Animal Facilities and Equipment
171
CFD is the use of highly complex mathematical models to predict air circulation patterns in a space (Reynolds 2009; Hughes, Reynolds, and Rodrigues 1996; Memarzadeh 1998; Jackson et al. 2002). CFD appears to be a powerful design tool for determining the optimal animal room ventilation pattern, given the room configuration, the species and number of animals to be housed, and the type of caging. One published study suggests that high returns, preferably in each corner or above each cage rack, are the most effective (Hughes et al. 1996); another suggests that low returns, one in each corner, are the most effective (Memarzadeh 1998). The problem is complex, and these CFD studies used different assumptions for key features; thus, additional study will be required to clarify this important issue. A study by Hughes and colleagues (1996) suggests that an even more efficient configuration is to supply and exhaust room air from a soffit mounted in the center of the ceiling extending the full length of the long axis of the room. In this CFD model, supply air is directed from radial diffusers in the bottom of the soffit toward the floor. Exhaust inlets located along both sides of the soffit capture the air as it curls from the floor, up the wall parallel with the soffit, across the ceiling, and into the soffit, where it is removed from the room. A full-scale test model of an animal room fitted with this type of soffit is reported to have performed even better than predicted by the CFD model (Hughes et al. 1996). Given the uncertainty, the high returns in each corner or the soffit configurations are tempting options in that they are less costly to construct than low returns and do not take up floor space; however, the best current answer is to do CFD studies specifically for the animal rooms in the facility being planned. Animal facilities typically have fixed equipment with special ventilation requirements. Fume hoods and certain types of biosafety cabinets require independent direct exhaust systems. Autoclaves require canopy exhaust hoods immediately above the autoclave doors with sufficient airflows to capture the heat, moisture, and odors that emanate from the autoclave when opened. This is especially important if the organic materials are to be autoclaved because they generate high odor levels when autoclaved. The cage sanitation area has unique ventilation requirements because high heat and moisture loads are generated in the room by cage wash-water temperatures that are 82°C (180°F) or higher. Tunnel washers and often the cage and rack washers are connected directly to the exhaust system; in addition, cage and rack washers must have exhaust canopies above the doors to capture the heat and moisture that emanate from the machines when the doors are opened (Figures€8.17b, 8.19a, and 8.19b). Because of the exceptionally high moisture levels in cage sanitation areas, these areas must be provided with a dedicated independent exhaust system, including exhaust fans and ducts. The canopies and all ducts must be nonferrous and acid resistant, and the ducts must be watertight and slopped and fitted to drain of the large amount of condensate released from the water-saturated hot air coming from the washers. The overall ventilation requirements for the cage sanitation area must take into consideration the enormous heat load in the room, which may include a significant mass of stainless steel coming out of the washers at temperatures of 82°C (180°F) or higher. Ventilated rodent cage racks are an example of mobile equipment that may be connected directly to the ventilation system. Ventilated racks may be used as freestanding equipment with blower and filter units that supply HEPA-filtered room air to the cages. They may also be equipped with blower and filter units that capture air coming from the cages and HEPA filter before blowing it back into the room (Figures€8.13 and 8.14). The blower and filter units can be mounted on top of the cage rack (Figure 8.13), but, ideally, are mounted on wall shelves and connected to the racks with flexible ducting (Figure€8.14). HEPA-filtering the exhaust air from the cages removes particulate contaminants but does not remove gaseous contaminants and heat. This can be accomplished by coupling the rack exhaust directly to the room exhaust (Figures€8.12 and 8.15). There are many strategies for integrating supply and exhaust air of ventilated racks with the ventilation system (Hessler and Frasier 2009; Hessler and Leary 2002; Reynolds 2009; Lipman 1993, 2009; Bilecke 2001). Table€8.2 includes a brief description of five options for integrating ventilated racks with the facility HVAC system.
84550.indb 171
11/11/10 6:24:49 PM
172
Handbook of Laboratory Animal Science
Figure 8.12â•…A color version of this figure follows page 336. Animal room with single-sided ventilated cage racks along both side walls with an animal transfer station (ATS) against the back wall. The height of the ATF work surface can be changed to suit the operator and it can be rolled up and down the aisle to where cages are being changed. The fixed cage racks are 12 rows high, requiring a step ladder to work with the cages in the top rows. Note: the supply air to the rack comes from a fan/ HEPA filter module above the ceiling that draws air from the room, and the air from the racks is exhausted directly into the facility exhaust system.
Figure 8.13â•…An individually ventilated microisolation cage rack with two filter/blower units on top of the rack. One supplies HEPA-filtered air to the cages; the other captures the air coming from the cages and HEPA filters it before blowing it into the room.
84550.indb 172
11/11/10 6:24:50 PM
Laboratory Animal Facilities and Equipment
173
Figure 8.14â•…Upper portion of an individually ventilated microisolation cage rack showing the cage supply and exhaust filter/blower units sitting on a wall-mounted shelf. This arrangement facilitates rack changes by eliminating the need to transfer the filter/blower units from rack to rack.
Figure 8.15â•…An individually ventilated microisolation cage rack with two filter/blower units mounted on a mobile rack alongside the cage rack. The cage exhaust filter/fan unit is connected directly to the room exhaust at the ceiling, similar to that shown in Figure 8.12.
84550.indb 173
11/11/10 6:24:51 PM
174
Handbook of Laboratory Animal Science
Table€8.2╅ Integrating Rodent Ventilated Cage Racks with the Facility HVAC System Option 1
From room air → supply FFM (with HEPA) → FD → rack/cage → FD → exhaust FFM with HEPA (with HEPA) → back into the rooma
Option 2
From room air → supply FFM (with HEPA) → FD → rack/cage → FD → exhaust FFM (without HEPA but with dust filter) → FES via a thimble connectionb From room air → supply FFM (with HEPA) → FD → rack/cage → FD → hard ducted to FES with a PICV/ID controlling airflow from the rackc From room air → a larger FFM supplying room air to multiple (typically all) racks in the room → PICV/ID controlling flow to each rack → FD → rack/cage → FD → hard ducted to FES with a PICV/ID controlling airflow from each rack From room air → through filter on cage → cage → through filter on cage → rack plenum → FD → iris damper → hard ducted to FES
Option 3 Option 4
Option 5
Notes: All these options start with room air because the room serves as a mixing box into which warm or cool air is supplied as required to maintain a constant temperature. Descriptions of other more complicated options that do not start with room air, along with advantages and disadvantages of those described here, can be found in Hessler and Lehner (2009) Chapters 20 and 34. FSS = facility air supply system; FES = facility air exhaust system; FD = flexible duct; FFM = fan/filter module— supply and exhaust (with or without HEPA filter); Rack/cage = rack supply plenum—cage—rack exhaust plenum; PICV/ID = pressure independent air flow control valve or iris damper; → = direction of air flow. a See Figures€8.13 and 8.14. b This is the same as option 1 except that exhaust air from the rack is directed into the building exhaust system. c See Figure€8.12.
Regardless of which strategy is selected, it is important to decide early in the planning process because the design of the room ventilation system must be matched with the equipment to gain maximum benefit. The decision not only affects the physical couplings but it also can have an impact on the cubic feet of air per minute (cfm) of supply air that will be required in the room. In determining cfm for a room with ventilated cage racks, it is important to recognize that exhausting air from the cages and cage racks directly into the facility exhaust system efficiently exhausts all odors but only approximately 20% (based on physical measurement (Hessler personal communication) to 35% (based on CFD; Reynolds 2009) of the animal heat load; the remainder is dissipated into the room air by convection and radiation. Each ventilated rack manufacturer recommends a specific cfm to be exhausted from its racks. The following combines this fact with an energy-saving strategy for ventilating animal rooms with ventilated cage racks. It involves providing controls to modulate airflow to the room depending on room exhaust air temperature. The minimum supply ventilation rate for the room is that required to support the ventilated racks in the room as specified by the rack manufacturer plus 10%. For example, a 250 sq. ft room with a 9 ft ceiling and three racks, each requiring an exhaust rate of 80 cfm, will require a minimum of 240 cfm for the racks plus 24 cfm (10%)—totaling a minimum air supply rate of 264 cfm. This calculates out to a room air exchange rate of seven changes per hour (cph). An additional amount to the minimal rate is required if the room is balanced positively to the corridor. A variable amount of ventilation up to a maximum rate of 15 or 20 cph is then automatically added as required by the heat load in the room. This will automatically adjust for differences in the heat load (animal census) in each room while recognizing that, with ventilated cage racks from which exhaust air is directed into the building exhaust, 65–80% of the animal heat load radiates into the room from the cages and rack exhaust plena. The room exhaust controls track the supply to maintain a set relative air pressure between the room and the corridor. As compared with conventional fixed volume control, which provides the same amount of ventilation regardless of the heat load, this variable volume ventilation design will allow for significant energy savings in animal rooms with a low animal census while providing only the minimal amount of additional ventilation required at any given time.
84550.indb 174
11/11/10 6:24:51 PM
Laboratory Animal Facilities and Equipment
175
Table€8.3╅Relative Air Pressure between the Corridor and the Animal Rooms Dual Corridor Managed as Conventional facility Barrier facility Containment facility
Single Corridor + or – + or – −
Clean
Soiled
+ + +
– – + or –
Notes: + = corridor positive to animal room. – = corridor negative to animal room. + or – single-corridor conventional: in a conventional facility, the air pressure in the corridor is generally maintained positive to the animal rooms. The exceptions are facilities with mixed “conventional” and “barrier” rooms, where the air pressure in the barrier rooms is maintained positive to the corridor and in the conventional rooms is maintained negative to the corridor. + or – single-corridor barrier: both options are used. Following is a rational for each: (1) corridor negative to animal rooms—to keep airborne contaminants out of the animal room; (2) corridor positive to animal rooms—to contain odors and inadvertent contaminants. Following is the rationale for this option: Infectious agents of concern are not ordinarily present in a barrier facility, so the rationale “to keep airborne contaminants out of the animal room, does not ordinarily apply as it does in a mixed facility. However, it must be assumed that a “break” will occur in a barrier room at some time. When this happens, the management objective is to contain the infectious agent, like in a biocontainment facility, until it can be detected and eliminated from the room and the facility. Keeping air pressure in the corridor positive to the animal room has the added benefit of reducing animal allergens and odors in the corridors and throughout the facility. + or – double-corridor containment: both options are used; negative is more common but positive may be preferred in some situations.
Air Balancing. Appropriate relative air pressures throughout the facility must be maintained to control airborne contaminants (Hessler and Frasier 2009; Hessler 1991a; Hessler, Broderson, and King 1999a, 1999b). This involves balancing supply and exhaust to maintain predetermined relative air pressures between adjoining spaces—typically, between the room and corridor. Table€8.3 summarizes various balancing options, depending on facility type and corridor plan. Maintaining proper balance requires proper sealing of the room envelope and maintenance of the appropriate volumetric offset between supply and exhaust air to achieve adequate differential pressures— typically, between 0.08 and 0.2 cm (0.03 and 0.075 in.) of water. Proper air balance is important in controlling contaminants, but it has limitations (Hessler 1991a). Most significant is to realize is that the relative air pressure in the spaces on either side of an opened door is essentially zero, allowing airborne contaminants to move freely between the spaces. The relative air pressures selected for animal rooms of a single-corridor facility will depend on how the facility is to be managed: conventional, containment, or barrier. In a single-corridor conventional facility, animal rooms are typically balanced negatively to the corridor, except for rooms that are designated as “barrier” or “clean” rooms, which are then balanced positively to the corridor. For this reason, the ability to reverse room air pressure automatically relative to the corridor without having to rebalance the entire system is a highly desirable feature in a single-corridor conventional animal facility. In a single-corridor containment facility, where the objective is to contain airborne contaminants, the relative air pressure in the animal rooms will be balanced negatively to the corridor. The opposite does not necessarily hold for a single-corridor barrier facility, where the choice depends more on management philosophy. One philosophy calls for balancing animal rooms positively to the corridor in an effort to keep airborne contaminants out; the other calls for balancing animal rooms negatively to the corridor with the objective of containing a disease break until it can be detected and eliminated. Both
84550.indb 175
11/11/10 6:24:51 PM
176
Handbook of Laboratory Animal Science
management philosophies have merit, and neither is clearly right or wrong. However, one advantage to the latter is that it maintains corridors relatively free of animal allergens, which are well documented as a serious and common occupational hazard (Huerkamp et al. 2009; National Research Council 1997; Reeb-Whitaker and Harrison 1999). In dual-corridor facilities, regardless of facility type, relative air pressures are typically balanced with the clean corridor positively to animal rooms and animal rooms positively to the soiled corridor; however, in some instances, both corridors may be balanced positively to the animal rooms. Temperature and Relative Humidity (RH) Control Each animal room should have individual temperature control using terminal hot water reheats (steam reheats are to be avoided) to allow for environmental temperature requirements for different species and differences in heat loads between rooms because of species differences and animal density. The standard design temperature range for animal rooms is between 18 and 29°C (65 and 85°F). This is not to be confused with temperature variations around a set point. The temperature control system should be capable of maintaining temperature ±1°C (±2°F) around any set point selected from the designed temperature range (Canadian Council on Animal Care 2003). Designing for a narrower temperature range may be acceptable for facilities intended for a single purpose (e.g., rodent production). Room temperatures as low as 18°C (65°F) are desirable for some commonly used species (e.g., rabbits); occasions for room temperatures over 26.6°C (80°F) are rare and usually involve the maintenance of relatively exotic species. Relative humidity (RH) in animal rooms should be maintained between 30 and 70% (Canadian Council on Animal Care 2003) with no specific set point generally required within this range. A well-designed HVAC system that supplies 12.8°C (55°F) air nearly saturated with water vapor can maintain this range of RH in multiple rooms without HR control in each animal room. Zonal control may be desirable in some situations (e.g., separating rooms with dry bedding systems from those where hose-down housing systems will be used). Clean steam free of boiler chemicals should be used for humidification in order to avoid the potentially confounding effects of chemical additives often used in boilers. An additional concern is the seasonal variation in the animal’s environment produced by using treated boiler steam for humidification. When outside air temperature is cold, humidification will almost always be required, thus introducing the boiler chemicals. However, when outside temperatures are warm, additional humidification is not required and thus no boiler chemicals are necessary. Exceptions may be very dry climates where humidification is required throughout the year, but even in this case, chemical additives in boiler steam are undesirable. Redundancy Uninterrupted control of the research animal’s environment is critical for both scientific and animal welfare reasons. One of the primary welfare reasons is the intolerance to high temperatures of rodents, in particular, and, to a lesser extent rabbits. Temperatures as low as in the range of 30–32°C (85–90°F) are potentially lethal for rodents that are not adapted to heat (Gordon 1993). Mechanical systems require routine preventive maintenance, often requiring shutdown of the system. In addition, mechanical systems are prone to fail. To account for such downtime and still assure consistent control of the research animal’s environment, the HVAC system should be designed with redundant critical components such as air handlers, exhaust fans, pumps, chillers, boilers, clean steam generators, and hot water heaters. There are many strategies for supplying redundancy, including parallel totally redundant systems or N + 1 systems (e.g., the number of air handlers required to meet 100% of the requirement plus one additional air handler). Other options for chillers and boilers may include cross-connecting with other
84550.indb 176
11/11/10 6:24:51 PM
Laboratory Animal Facilities and Equipment
177
lower priority sources to access available chilled water or steam and having spare parts available for quick replacement. Many campus-wide central chilling and boiler plants have multiple units that may be sufficient to satisfy the redundancy requirement as long as they can supply what is needed when one of the units is down. If not, a separate boiler or chiller for the animal facility could be provided as backup for chilled water or steam coming from the central plant. HVAC Control and Environmental Monitoring Contemporary animal facilities use direct digital controls (DDCs) to control and monitor the HVAC system. Parameters electronically monitored in each animal room and animal procedure room include temperature and humidity with sensors located in a common exhaust duct collecting air from all exhaust ducts coming from the room, differential air pressure, and supply and exhaust air flows. The ability to monitor and control air flows digitally not only assures maintenance of the proper parameters, but also makes possible variable volume control based on temperature—the ability to change room air pressure relative to the corridor via a keyboard. Environmental monitoring should include providing alarms to responsible offices and individuals when parameters go out of preset ranges. In addition, the monitoring system should be able to provide reports documenting the average, high, and low temperature and humidity; average, high, and low relative air pressure; and room air exchange rates (changes per hour) in every animal room and procedure room for each day, week, or month over any selected period of time. Many advocate having a redundant monitoring system independent of the control system. Energy Conservation Because of the high ventilation rates of 100% fresh makeup air required by animal facilities, an energy recovery system that preconditions incoming air with discharge air often proves to be cost effective, depending on local climatic conditions. Recover systems should be limited to types that preclude contaminating incoming air with outgoing air. Given the sophistication and reliability of modern control systems, variable volume control of ventilation in areas with variable heat loads offers great potential for energy conservation. A prime example of such a space is the cage sanitation area, which has a very high heat load and high ventilation requirement while operational, but a very low heat load and a very low requirement for ventilation when it is not being used. Variable volume control of ventilation in animal rooms driven by the heat load but with a set minimal ventilation rate is also a viable option for energy conservation. An example of this was described in the “Ventilation” section in connection with ventilated rodent caging. Power and Lighting Power and Emergency Power The demand for electrical outlets in animal rooms has increased with the increased use of ventilated racks, data processing equipment, scales, research equipment, HEPA-filtered mass air displacement cabinets, and powered sanitation equipment in the animal room. The outlets should all have water-resistant covers. The location of the outlets needs to be planned carefully, especially if ventilated cages are to be used. Ground fault interrupters (GFIs) should be used for every circuit in areas of the facility where water will be routinely used, which is most of the facility. Emergency power should be adequate to maintain all essential services in the event of a main power failure. At a minimum, emergency power should include HVAC at 100% capacity, including chillers, any animal housing equipment that relies on power to maintain airflow (e.g., ventilated
84550.indb 177
11/11/10 6:24:51 PM
178
Handbook of Laboratory Animal Science
racks), all environmental control and monitoring systems, at least one light fixture per animal room, surgery lights and other safety lighting as required by code, the security system, the surgery room, and freezers. It sometimes proves to be most cost effective to put the entire animal facility on emergency power rather than providing separate circuits. Lighting Photoperiods are a critical component of maintaining the animal’s environment. Therefore, automatic control of lighting in windowless animal rooms is the norm. Most research facilities will require independent lighting control for each animal room. Conventional fluorescent lighting is standard for animal rooms. The ceiling fixtures may be recessed or surface mounted. The light fixtures should be water resistant and arranged to provide uniform lighting throughout the room, taking into consideration the arrangement of the cage racks or pens. A digital light control system located in a secure location remote from the animal rooms is best. It is also highly recommended to monitor and document light cycles by monitoring light intensity in the room so as not to rely on just documenting that the control system signaled for the lights to go on or off. A dark room light independently controlled with a timer switch at the room may be used to facilitate activities that must be conducted during the dark cycle. Providing white light during the dark cycle is not recommended because interruptions of the dark cycle do alter the animal’s circadian rhythm and may have an impact on experimental results (Dauchy et al. 1997; Hessler 2009c; Dysko et al. 2009). High light levels cause retinal damage in albino rodents. In recognition of this phototoxic effect, the Institute of Laboratory Animal Resources (ILAR) Guide (ILAR 1996) recommends that light levels in rooms housing albino animals be 325 lux (30 fc) at 1 m (3.3 ft) above the floor. This level is generally sufficient for animal care, and task lighting can be provided for performing procedures that require higher levels of light. A bilevel low/high (325/800 lux) lighting system may be considered with the intent of using the high level to facilitate working in the animal room; however, it must be noted that even brief periods of high light levels may result in retinal damage for albino animals. Retinal damage does not occur in animals with normally pigmented eyes at typical indoor lighting levels. Therefore, it is acceptable and even desirable to provide light levels of 800–1100 lux (75–100 fc) in animal rooms designed to house only dogs, nonhuman primates, or other animals that normally have pigmented eyes. Communications Essential communications design features include telephone lines strategically located throughout the facility to include most rooms but not animal rooms, Internet connections hard wired or wireless in most rooms (including animal rooms), and video cable lines in selected rooms (surgical and training rooms). See Hessler (2009c) for a more comprehensive description of special considerations for electrical features in research animal facilities. Plumbing Two decision points faced early in the process of planning a research animal facility concern sinks and floor drains in animal rooms. In particular, which animal rooms will require them, if any. In many cases there is not a right or wrong answer, just opinions, which vary widely. Sinks Sinks are desirable in most animal rooms, but are required (USDHHS 2007; Hessler et al. 1999b) only in ABSL-3 animal rooms. In addition to being useful for hand washing and miscellaneous uses
84550.indb 178
11/11/10 6:24:51 PM
Laboratory Animal Facilities and Equipment
179
Figure 8.16â•…A color version of this figure follows page 336. A stainless steel wall-mounted sink in an animal room. Its primary use is to facilitate room sanitation. Each animal room has a dedicated mop bucket and mop along with other sanitation implements seen in the figure to reduce the opportunities for spreading infectious agents between animal rooms.
that come up when working with animals in the room, the most beneficial use of sinks in an animal room is for dumping water from mop buckets. It is considered good sanitation practice for each animal room to be equipped with dedicated equipment for routine cleaning of the floor. For this reason, a stainless steel mop sink is desirable. The sink should be mounted on the wall to avoid impeding floor sanitation (Figures€8.5 and 8.16). Hands-off controls are desirable. A cold-water hose bib mounted on the wall under or near the mop sink at a height suitable for filling mop buckets is also useful. If there is uncertainty regarding the installation of sinks in the animal rooms, an option is to fit each room with plumbing for hooking up mobile sinks. Alternatively, several rooms in a suite of animal rooms could share a single sink as long as there is no concern about carrying infectious agents between rooms. Floor Drains Animal rooms intended for dry bedding cage systems do not require drains. Whether or not to include one is a matter of choice. There are advantages and disadvantages to both, with flexibility being the primary advantage to including them. Other than that, the disadvantages of having them may easily outweigh the advantages. Disadvantages include installation cost; confounding pest control, especially the control of cockroaches; the potential for sewage backing up into the room, slopping floors that can cause problems when trying to park racks with wheels; underutilized traps drying
84550.indb 179
11/11/10 6:24:52 PM
180
Handbook of Laboratory Animal Science
and thus allowing sewer gas to escape (capping the drain with airtight seals can alleviate this problem, but then the drains are not convenient to use); and taking up space, especially trough drains. In animal rooms designed for hose-down caging systems, the location of the floor drain and slopes of the floor are critical to efficient cleaning. Ideally, the drain should be at the low point of an open floor trough located against the side walls of the room so that the cages or pens back up to the drain trough but do not cover it. If floor pens are used, the trough should be uncovered and outside the pens, leaving a minimum 46 cm (18 in.) access aisle between the pens and the wall. The room floor should be sloped at a minimum of 1.6 cm per m ( 163 in. per foot) from a crown in the center of the room to the floor trough on each side of the room or from one side of the room to a trough on the opposite side if pens are to be located on one side of the room. The bottom of the trough should slope a minimum of 2 cm per m ( 14 in. per foot) toward a minimum 10 cm (4 in.) diameter drain or 15 cm (6 in.) diameter if the drain will service a large number of animals. The drain should have rim and/or trap flush fittings. In addition, there should be a water source at the high point of the trough controlled with the same ball-type valve that controls the flow of water to the trough and flush drain fittings. Animal Drinking Water Decisions, decisions: automatic watering or water bottles, potable house water or reverse osmosis (RO) water, acidification, chlorination, or no animal water treatment? All are important questions that are best answered in the planning stage of an animal facility because all involve special plumbing and equipment requirements. All options are currently used; this suggests that all are currently considered acceptable. There are opinions but no consensus as to which options are best or most cost effective. The answer may depend largely on individual situations. The issue of cost effectiveness is complex and clearly has no one best answer. One issue that is less complex for which a science-based rationale exists is that of using RO water. Given that controlling the research animal’s environment is a primary focus of laboratory animal science and that potable water content and quality vary considerably from one place to another and from time to time at the same location, it seems most rational to standardize water content and quality for all research animals worldwide. A simple and relatively inexpensive way to do this is to standardize using RO water for research animals. Using this same logic, water treatment should also probably be standardized. Since chlorination is not effective with water bottles used past a few days because it dissipates from the water, acidification appears to be a reasonable choice, but some other type of water treatment may be even better. This is just food for thought. Hoses Hose reels suspended from the ceiling or upper wall in areas where hoses are to be used routinely are highly recommended, along with an independent pressurized recirculating warm-water (~40°C; 104°F) plumbing system supplying water to all hoses in the facility. Safety Safety eyewash and shower stations are required where caustic chemicals may be used, including both sides of the cage sanitation area, near animal water-bottle-filling equipment with water treatment capabilities, near the reverse osmosis water production unit, and in most wet laboratories. Steam Service and Clean Steam In addition to the usual steam requirements for water heaters for domestic hot water and terminal hot water reheats for the HVAC system, steam service will be required for cage washers and
84550.indb 180
11/11/10 6:24:52 PM
Laboratory Animal Facilities and Equipment
181
autoclaves. The equipment manufacturers’ specifications for steam requirements should be documented during the planning stages. Clean steam will be required for autoclaves and for humidification. When steam services are available, steam-to-steam generators are often used for generating clean steam. Pharmaceutical-grade clean steam is not required. The use of potable water from which to generate the clean steam is acceptable; however, generating it from distilled or reverse osmosis water is best from the point of view of generator maintenance. The choice may depend on the “hardness” of the potable water. The reason for using clean steam for humidification and autoclaves is to avoid the potential confounding effects of chemical additives routinely used in steam boilers. Boiler additives are generally considered safe, with no known health effects at the levels present in air humidified with boiler steam. However, the degree to which the chemicals may or may not alter the research animal’s biological response to an experimental variable is impractical to document for the wide array of animal models that may be used in the facility. In addition, the seasonal variation in the level of chemical additives in the air, which is present in relatively large quantities in the winter and absent in the summer, is in itself an unnecessary environmental variable. For these reasons, chemically treated boiler steam is best avoided for humidification. For the same reasons, clean steam is also recommended for autoclaves (chamber but not necessarily for the jacket) that will be used for autoclaving animal husbandry equipment, such as cages, and supplies, such as feed and bedding. Bulk Detergent Delivery Detergents, acid, and neutralizing agents may be piped to cage sanitation areas from vats or barrels located at or near the receiving dock. See Dysko and colleagues (2009) for a more comprehensive description of special considerations for plumbing in research animal facilities. Security and Access Control The increasing use of destructive tactics by activists opposing the use of animals for biomedical research and safety testing combined with the great value of research resources invested in the animals dictates that all research animal facilities have sound security. This starts with hardening the perimeter by controlling access to the facility. Access points should be kept to a minimum and all that will be routinely used must be equipped with microprocessor-controlled security access devices. Those access/egress points that are not intended for routine use must be equipped with alarms to alert that they have been opened. Even though animal facility management makes the decision about who has access to the animal facility, because of its critical nature, this perimeter security system preferably is managed by a professional security specialist; this may be the institution’s own security service or a commercial security service. The use of a biometric identification system (e.g., thumb, palm, or retinal scan; voice recognition, etc.) is highly recommended. Closed-circuit TV monitoring and recording at all access points, routinely used or not, should also be considered. Controlling access to areas or rooms within the animal facility further enhances security; however, its primary benefit is as a management tool to protect animal health and the integrity of the research by controlling and monitoring access to animal housing rooms and isolated areas of the animal facility (e.g., barrier, biocontainment, primate, quarantine, etc.). No one should have access to an animal room or isolated area without justification and prior approval. Since the need to change access approvals for rooms and areas within the animal facility occurs frequently (often because of animal facility management decisions) and with short notice, it works best if this system is managed by animal facility personnel. Key lock systems are marginally manageable when a small number of people require access. When a large number of people in a high-turnover population require access, a key lock system is unmanageable and thus ineffective. Numeric coded locks for which each door has a unique code
84550.indb 181
11/11/10 6:24:52 PM
182
Handbook of Laboratory Animal Science
are also unmanageable and ineffective; they are unwieldy and inefficient for animal care staff that have to enter multiple areas and rooms each day and on weekends and holidays—possibly all areas and rooms. Effective access control can best be achieved with a microprocessor-controlled security system utilizing either a personal identification number and key pad or a card entry system. Equipment Cage Sanitation and Sterilization The cage washing area is one of the most important areas of a laboratory animal facility; unfortunately, it is an area that is often not planned carefully enough. Insufficient space and inadequate equipment are the most significant planning errors. The area needs to be large enough to hold the required space-demanding equipment as well as the dirty and clean cages in a configuration that facilitates work efficiency (Figure€8.3). It should have one or more pieces of cage sanitation equipment and may have an autoclave, bottle-filling equipment, and equipment for sanitizing automatic watering devices. The type of equipment required depends on the size of the facility in terms of cage rack capacity and types of cages, which, of course, is dependent on the species to be housed. During the planning phase, it is essential to evaluate the productivity metric to be certain that size and throughput capacity of the sanitation and sterilization equipment match the types and numbers of cages to be used in the animal facility. (See earlier sections of this chapter for architectural and HVAC considerations, and Klein, Kuntz, and Hessler, 2009, for another description of planning and designing a cage sanitation area.) Cage cleaning includes several steps to ensure that cages are freed from urine salts, feces, and vegetative microorganisms. This can be accomplished by washing the cages by hand, but not efficiently and not without the use of chemical disinfectants that are best avoided if possible. Mechanical cage washers sanitizing with high-temperature water accomplish the job more effectively, efficiently, and safely. Commercial mechanical cage washers, while in many respects similar to restaurant dishwashers and hospital cart washers, have design features specifically suited for washing cages. Cage washing cycles typically start with a prewash rinse to get rid of loosely attached items such as bedding material and feces. Depending on the washer, the rinse may be performed with cold tap water or with warm water recycled from the final rinse water. The second cycle is intended to wash cages free from fatty products using an alkaline (basic) detergent that is automatically dispensed into the wash water. The desirable water temperature in this cycle is 60–70°C (140–158°F). The last cycle is the final rinse with a water temperature of at least 82°C (180°F) to sanitize the equipment being washed and render it free of viruses and vegetative bacteria of concern. A fourth cycle involving an acid rinse may be interspersed between the detergent wash cycle and the final rinse cycle to neutralize the alkaline detergent that may result in hydrolyses of polycarbonate during autoclaving, especially if strong caustic sodium or potassium hydroxide alkaline detergents are not adequately rinsed from the cages. Hydrolysis of polycarbonate material during autoclaving is less likely to occur following washing with milder sodium bicarbonate alkaline detergents, thus eliminating the need for an acid rinse. Acid treatment of cages as an acid rinse cycle during mechanical washing or as a prewash treatment applied by hand can effectively reduce the buildup of urine salts on the cages. There are two basic types of mechanical washers: batch washers and continuous belt washers. Batch washers cover all the cycles within a single chamber. Continuous belt washers, also known as tunnel washers, transport materials to be washed on a belt through a tunnel divided into various rinse and wash sections. The type of washer selected depends on the facility size and species to be housed. It is common for facilities to have both. Larger facilities will require two or more of one or the other or both types of washers.
84550.indb 182
11/11/10 6:24:53 PM
Laboratory Animal Facilities and Equipment
183
Batch Washers Batch washers come in single-door and double-door pass-through models. Pass-through models are much preferred because they provide for the separation of clean and soiled sides of the cage sanitation area. They come in two basic types. The most common is the floor-loading model, often referred to as a “cage and rack washer” because entire racks of cages can be rolled into the washer. Typically, the washing chamber is sized wide enough for one cage rack, deep enough to hold one or two cages racks, and tall enough to hold the tallest mobile cage racks in the facility (nominal maximum rodent rack size = 0.6 m wide, 1.8 m long, 2 m high [2 ft wide, 6 ft long, 6.6 ft high]; dog and primate cages may be larger, e.g., 1 m wide [3.3 ft]). Cage and rack washers are best suited for washing large cages used for housing rabbits, dogs, nonhuman primates, etc. They are typically pit mounted to make the washer floor level with the room floor (Figure€8.17) and, while it is possible to ramp up to the washer floor, this presents a potentially dangerous ergonomic problem (Figure€8.18). With appropriate wash racks, they can
(a) Figure 8.17â•…A color version of this figure follows page 336. A series of figures illustrating the cage-washing equipment for an animal facility that has the capacity to house approximately 10,000 mouse cages. Not shown for the same facility are two floor-loading bulk autoclaves located on the clean side of the cage sanitation area that pass through to the sterilized cage storage area. (a) The soiled side of cage sanitation showing the cage and rack washer on the left and the tunnel washer on the right. (b) The clean side of cage sanitation showing the cage, the automatic bedding dispenser in front of the discharge end of the tunnel washer on the left, and the discharge side of the cage and rack washer on the right. Note the stainless steel canopy exhaust air hood over the door of the cage and rack washer. Note also the large duct between the bedding dispenser and the ceiling that is used to deliver bedding to the bedding dispenser automatically by a combination of a vacuum and auger transport system leading from a bulk bedding storage container located near the dock that is loaded in bulk from a delivery truck. The dispenser is also equipped with a vacuum to collect bedding dust generated by the bedding dispenser. (c) This is a close-up view of the bedding dispenser to illustrate the difference in height of the tunnel washer belt and the bedding dispenser belt. As the cages fall off the tunnel washer belt, they automatically turn over onto the dispenser belt so that the bedding can be dispensed into the cages as they pass through.
84550.indb 183
11/11/10 6:24:55 PM
184
Handbook of Laboratory Animal Science
(b)
(c) Figure 8.17â•…(Continued)
84550.indb 184
11/11/10 6:24:57 PM
Laboratory Animal Facilities and Equipment
185
Figure 8.18â•…A batch-type cage and rack washer that is not pit mounted, thus requiring a ramp to load rolling equipment into the washer. Pit mounting is much preferred in that it avoids the potentially dangerous ergonomic problems associated with the ramp. The wash rack alongside the washer is useful for holding small cages and other small equipment in the cage and rack washer.
also be used effectively for washing rodent shoebox-type cages (Figure€8.18). An optional feature on cage and rack washers is to have saddle tanks that store the water from either the alkaline detergent wash or acid rinse cycles (or both) so that it can be reused to conserve water and chemicals. A second type of batch washer, often called a cabinet washer, may have most of the features of the cage and rack washer except that it has a much smaller chamber—for example, 122 cm (48 in.) wide × 79 cm (31 in.) high × 86 cm long (34 in.)—with the bottom of the wash chamber approximately 91 cm (36 in.) off the floor (Figure€8.19). This type of washer is suitable for very small facilities that use only cages that can be safely lifted up into the chamber. Some larger facilities may have one for washing animal watering bottles, feed pans, and small cage parts. Continuous Belt Washers While not as versatile as cage and rack washers, continuous belt washers, also referred to as tunnel washers, are more efficient at washing and sanitizing solid-bottom shoebox cages (Figures€8.4 and 8.17a). Therefore, they are commonly used in facilities that have a large number of these types of rodent cages. This type of washer is also well suited for sanitizing cage pans, water bottles, and other small equipment. Cages and equipment to be cleaned and sanitized are placed on a conveyor belt that moves through a tunnel divided into sections (e.g., prerinse, detergent wash, rinse, and final rinse). Water is usually recirculated; the recirculating rinse water is used for prerinse, which is discarded, and the final rinse water flows into the recirculating rinse water to freshen it. Often, a dryer and bedding dispenser section are added to the tunnel washer (Figure€8.17b,c). Tunnel washer belt widths and lengths come in a variety of sizes; however, a typical tunnel washer may have a conveyor about waist high and 1.07 m (42 in.) wide, with a tunnel opening of 115 cm (45 in.) wide × 61 cm (24 in.) high and 4.6 m (15 ft) long. The width and length of the washer
84550.indb 185
11/11/10 6:24:58 PM
186
Handbook of Laboratory Animal Science
(a)
(b)
Figure 8.19â•…A color version of this figure follows page 336. The clean side of a cage sanitation for a facility with the capacity to house approximately 1,000 mouse cages. Note the stainless steel canopy exhaust hoods above the doors of the washer and autoclave. (a) A cabinet-style cage washer is on the left and a small autoclave (small relative to floor loading bulk autoclaves such as those illustrated in Figure 8.21) is on the right. Both have two doors so that equipment passes through from the soiled to clean side; however, when the autoclave is used for autoclaving cages, after having been washed, the cages are loaded and unloaded from the clean side. (b) The cabinet cage washer with the pull-down door opened and the wash rack pulled out onto the door. The doors on both sides operate the same to facilitate loading and unloading the washer. (c) The autoclave with the door opened and cart in front of it to facilitate unloading the autoclave. There is a similar cart on the soiled side.
84550.indb 186
11/11/10 6:25:01 PM
Laboratory Animal Facilities and Equipment
187
(c) Figure 8.19â•…(Continued)
determine the washing capacity. The wider the belt, the more cages it can hold per unit of length, and the longer the washing tunnel, the faster the conveyor belt can run and still provide an effective exposure time in each section. A dryer may add another 2.4 m (8 ft). With a dryer, the belt of the washer extends uninterrupted through the dryer. An automatic bedding dispenser may add an additional 3 m (10 ft). The belt of the bedding dispenser is independent of and below the level of the washer and dryer belt. Load and unload sections are other optional features that could add another meter (39 in.). With a washer, dryer, bedding dispenser, and load and unload extensions, a typical tunnel washer assembly may be 11 m (36 ft) long or longer. Ideally, a stainless steel roller conveyer extends 2–3 m (6.5–10 ft) past the dryer or bedding dispenser, whichever applies, to collect cages before they can be removed. Robotic Cage Washing The rapidly increasing use of mice in molecular biology research and the resulting large increase in the mouse cage census of many facilities has led to the introduction of robotics into the care of laboratory animals (Klein et al. 2009; Ruggiero 2001; Corey, Davey, and Faith 2001). To date, the use of robots has been limited to a relatively few facilities housing large numbers of mice. As the technology improves and decreases in price, the use of robots to assist with cage washing will likely increase. Robots may be used on both sides of the tunnel washer/dryer/bedding dispenser or just on one side or the other. On the soiled side, a robot picks soiled cages off a cart (Figure 8.20a), typically one to four cages at a time; dumps the bedding into a disposal unit (Figure 8.20b) that automatically transports and deposits the bedding in a disposal container outside the facility; and then places the cages on the tunnel washer belt (Figure€8.20c). At the clean end of the tunnel washer, a second robot picks the cages off the washer belt, places them under an automatic bedding dispenser, and then stacks them filled with bedding on a cart. Some robotic systems require using an indexing tunnel washer
84550.indb 187
11/11/10 6:25:02 PM
188
Handbook of Laboratory Animal Science
(a)
(b) Figure 8.20â•…A color version of this figure follows page 336. Robot processing cages four at a time on the soiled side of cage sanitation. (a) Picking up cages from a specially designed cage transport cart. (b) Dumping soiled bedding from the cage into a hopper from where it is transported by vacuum to a dumpster outside the building. (c) Placing the cages onto the tunnel washer belt.
84550.indb 188
11/11/10 6:25:04 PM
Laboratory Animal Facilities and Equipment
189
(c) Figure 8.20â•…(Continued)
and dryer that stops the belt while the robot is loading and unloading the belt on long extensions on both ends that hold, for example, 16–20 cages at a time. Because the belt does not run continuously, indexing tunnel washers may have automatic sliding doors that separate the various sections of the washer. Some robotic systems work with a standard continuous running belt. Sterilization Equipment The usual way to sterilize cages, bottles, lids, and other material necessary for laboratory animal care that can withstand high temperatures is to use an autoclave. An autoclave is a pressure vessel that uses a combination of above-atmospheric pressure and steam to generate temperatures above the atmospheric boiling point. A complete sterilization cycle for nonliquid materials typically includes several cycles of high vacuum; each vacuum cycle is followed by live steam injection to purge the chamber of air. The high temperature and pressure sterilization cycle, typically 120°C (250°F) for 20 min or 135°C (275°F) for 15 min, is followed by time for steam withdrawal, pressure equilibration, and cooling. High-vacuum autoclaves are far more effective and more practical than gravity autoclaves in the laboratory animal environment. In addition, “clean steam” generated from domestic, distilled, or RO water (not necessarily as high a quality as pharmaceutical-grade clean steam) should be used in the autoclave to avoid contaminating the cages, etc. with chemical additives (typically, filming amines used to coat the boiler and steam pipes to reduce corrosion) normally put in steam boilers. Amines from live steam potentially damage polycarbonate and add an avoidable variable to the animal’s environment. For effective sterilization, the material being autoclaved should not be wrapped, wrapped with materials that are air and steam permeable, or wrapped with the wrapping partially open to allow the vacuum cycle to extract the air and to prevent the creation of air pockets that impede sterilization. Autoclave chamber sizes used in animal facilities vary considerably from ones that will hold only a
84550.indb 189
11/11/10 6:25:06 PM
190
Handbook of Laboratory Animal Science
few cages to ones that will hold three, four, or even more racks of cages. When planned for use in an infectious containment facility or a rodent barrier facility where routine husbandry relies on the sterilization of cages, it is essential to calculate output capacity carefully, which is dependent on the size of the autoclave chamber and the autoclave cycle times. Pit-mounted, floor-loading bulk autoclaves large enough to hold one or more cage racks are commonly used in biocontainment and barrier facilities (Figure€8.21b). If it is not possible to pit mount or practical to ramp up to the autoclave floor, another option is to use a platform lift to raise the equipment to the level of the autoclave floor (Figure 8.21a). For productivity calculations, a cycle time of 1 h, 30 min should be used to determine the number of cycles possible in a typical workday. Some may choose to process plastic cages using a “pasteurization” cycle in which the sterilization cycle may be as short as 5 min. This shortened cycle will not uniformly sterilize, but will likely kill all microbes of serious concern while reducing damage to the plastic cages. Such heavy use of autoclaves increases maintenance downtime. Given the importance of the autoclave for maintaining routine animal care, a backup autoclave should be considered, preferably as two parallel, fully redundant units, although an autoclave elsewhere in the animal facility may suffice. Autoclaves used for supporting laboratories and surgical areas should be located in those areas. Autoclaves used for animal care are almost always required for providing some level of biocontainment and/or barrier. Most often, autoclaves used to support animal care have two doors so that materials can be passed through from one area to another. There are several options for locating autoclaves used for routine animal care. They may be located in the cage sanitation area, either between the soiled and clean sides or between the clean side and the “sterilized” equipment storage area. When supporting an infectious containment area (Figure 8.7) or a rodent barrier area of the animal facility, the ideal is to have a double-door, floor-loading autoclave as an integral part of the perimeter of these areas where cages, etc., can be autoclaved into or out of the area, thus eliminating the need to wrap the cages, etc., for transport between the respective animal housing area and the autoclave. Materials that may not withstand the high temperatures of autoclaving may be sterilized with gases such as paraformaldehyde, hydrogen peroxide, and chlorinedioxide. Irradiation is commonly used for sterilizing feed and bedding. Large chamber hydrogen peroxide and chlorinedioxide gas sterilizers along with dry heat sterilizers are increasingly being considered for sterilizing cages. One significant advantage, among others, is that they cause less damage to cages. Animal Watering In the earlier section on plumbing, the issue of water quality and treatment was covered. This section covers the equipment involved with animal watering. Choosing a watering strategy is a critical decision that needs to be made early in the facility planning process (Dysko et al. 2009; Lempken 1999; Novak 1999). Two methods are commonly used to provide water to laboratory animals: in bottles or pans attached to or in the cage or with automatic watering devices connected to a continuous supply of water. The bottles are typically glass or polycarbonate with holes in the bottle or lid, or equipped with stainless steel sipper tubes. Automatic watering for larger species such as dogs, pigs, rabbits, etc. is relatively problem free and has been widely accepted and used for many years; however, the use of automatic watering for rodents, especially mice, is more controversial. The following will focus primarily on watering for rodents housed in solid-bottom bedded shoebox cages since they make up the vast majority of animals used in research and watering for nonrodent species tends to be less of a problem. Water bottle and automatic watering systems have pros and cons. Water bottles are relatively trouble free and have a low start-up cost as compared with automatic watering; however, water bottles are labor intensive, needing to be exchanged with freshly sanitized bottles at least weekly and, if not acidified, at least two times a week. Automatic watering devices significantly reduce labor costs but
84550.indb 190
11/11/10 6:25:06 PM
Laboratory Animal Facilities and Equipment
191
(a)
(b) Figure 8.21â•…Double-door pass-through bulk autoclaves. The autoclave pictured in (a) has sliding doors and is not pit mounted, thus requiring a lift in order to load and unload carts and cage racks. The autoclave in (b) has hinged doors and is pit mounted, which greatly facilitates loading and unloading. A steel plate bridges the gap between the floor of the autoclave and the facility floor. This autoclave has the capacity to hold three of the type of mouse rack shown in the figure. Note the exhaust hood in the ceiling above the door.
84550.indb 191
11/11/10 6:25:10 PM
192
Handbook of Laboratory Animal Science
have a relatively high front-end cost, can and do leak and flood cages, and present a quality control challenge because of microbial growth in the low-pressure, low-flow water distribution system. In addition, sanitation of the automatic watering valves attached to the rack manifolds is a concern. Regarding leakage, water bottles also leak and, while the amount of water is limited to that in the bottle, it too can present serious problems for mice. The quality of automatic watering valves has greatly improved over the 40 or so years that they have been used, reducing but not eliminating the problem of cage flooding. Regarding sanitation of the watering valves, this concern is addressed by periodically autoclaving the entire rack with the stainless steel watering manifold, having quickdisconnect valves that can be removed and autoclaved, or by attaching the valves with quick-disconnects to the cage so that they are sterilized along with the cage. Regarding microbial contamination, the problem has been addressed by designing the distribution lines to avoid dead ends, flushing the distribution lines at least daily, treating the water (e.g., 1–3 ppm of chlorine even when RO water is used), and periodically sanitizing the system with hyperchlorinated water. This same microbial contamination problem also applies to the cage rack automatic watering manifold and is addressed in a similar manner by designing the manifold with a sigmoid configuration to achieve a single flow track through the manifold. Automatic flushing of the manifold is achieved by having two parallel distribution lines to each rack: one to supply water to the manifold and one to receive the water discharged from the manifold. A system of solenoid valves and a microprocessor control the automatic flushing. The bottom line is that both water bottles and automatic watering devices are acceptable and widely used methods of providing water to animals. The choice made will rely primarily on the degree of confidence one has in automatic watering and the issue of front-end equipment cost versus long-term maintenance labor costs. Equipment to support the use of water bottles includes the following (see Figure€8.22): wire baskets for handling multiple bottles at a time (e.g., 24 bottles), bottle basket carts for transporting water bottle baskets, a bottle filler manifold to fill all the bottles in a basket at the same time, a water treatment proportioner (acidification or chlorination), and equipment for sanitizing bottles. Various degrees of automation are available for filling bottles and treating water. Bottles can be washed and sanitized in tunnel washers, cabinet washers, and even cage and rack washers with the optional addition of a special rack for washing bottles (note that space needs to be provided for storing the bottle washing rack when it is not being used). A contemporary automatic watering system for laboratory animals designed to assure water quality includes the following features: RO water; automatic chlorination or acidification of RO water, ultraviolet sterilization of water, especially if a recirculation system is to be used; automatic flushing of the room distribution lines and rack manifolds; automatic monitoring and alarming of RO water quality and chlorine level or pH; water pressure throughout the system; and leak detection. In addition, maintenance equipment includes hyperchlorination and manual flush equipment on the soiled side of the cage sanitation area for sanitizing the rack manifolds and water coils with quick disconnects on both ends and a hyperchlorination unit for sanitizing the automatic watering distribution system. Caging Rodents Small rodents (e.g., rats, mice, hamsters, gerbils, and guinea pigs) are usually housed in solidbottom, shoebox-type cages with various types of bedding materials covering the cage bottom. Open stainless steel wire-bottom cages without bedding material are considered less desirable in terms of animal welfare but are occasionally used when required to achieve scientific objectives. Large guinea pigs are sometimes housed in rabbit cages, especially ones with perforated plastic floors. Rabbit cages are particularly adaptable to creating a highly enriched environment for housing rats (Figure€8.23).
84550.indb 192
11/11/10 6:25:11 PM
Laboratory Animal Facilities and Equipment
193
(a)
(b) Figure 8.22â•…A series of figures illustrating ways to handle water bottles for rodent cages to accommodate different needs. (a) Bottle baskets on a bottle basket transport cart. (b) A very basic water-bottlefilling manifold that folds down from a cabinet on the wall. (c) A water-bottle-filling manifold over a stainless steel roller conveyor leading from a tunnel washer dedicated to washing water bottles. (d) A water-bottle-filling station with a chemical proportioner to treat the water automatically with acid or chlorine.
84550.indb 193
11/11/10 6:25:13 PM
194
Handbook of Laboratory Animal Science
(c)
(d) Figure 8.22â•…(Continued)
84550.indb 194
11/11/10 6:25:15 PM
Laboratory Animal Facilities and Equipment
195
Figure 8.23â•…Rat group housed in a rabbit cage. The complexity of the environment has been increased with hay and a branch on which the rats can climb to a shelf.
Rodent shoebox-type cages are made of various types of plastic materials. The material most frequently used at this time is polycarbonate (Lexan® [GE] or Makrolon® [Baye]r). This material is transparent, rigid, durable, and sanitizable. Its softening temperature is 152–157°C (305–315°F), and it withstands autoclaving at 120°C (250°F) but eventually becomes brittle after multiple exposures to autoclave temperatures. A newer copolycarbonate (Apec® [Bayer]) better withstands higher temperatures; thus, it is often referred to as “high-temperature polycarbonate” and is reported to hold up better under repeated autoclaving. More recently, polysulfone (Radal® [Amoco]) was introduced as a cage material. It is a durable plastic reported to be able to handle thousands of repeated autoclavings while maintaining impact strength and transparency. Due to its brownish color, polysulfone has about 35% less light penetration than polycarbonate. This might not be a problem in many laboratory facilities; in fact, it may be beneficial because light intensity may be high to make caretaking easier without illuminating the animals too much. The sole drawback with polysulfone is the initial cost, which is high in comparison to the other materials, but it may be cost effective in the long run because of the high durability. Other types of plastic are also used for rodent caging. Polypropylene is a light, flexible material with high chemical inertia and thermal resistance up to 120°C. It may be translucent or opaque, depending on whether it is a copolymer or a reinforced copolymer. Polystyrene is rigid, with low impact and heat resistance. It is usually used to form disposable cages suitable for toxic or radioactive applications when decontamination of the cage is impractical or too dangerous for personnel. Disposable versus Nondisposable Cages When planning facilities for housing a small number of rodent cages, consideration could be given to using only disposable cages. This would involve comparing the major cost factors for each option over the anticipated life of the facility or some predetermined time period. Disposable cage cost considerations may include: • • • •
84550.indb 195
The cost of the disposable cages The cost of space for maintaining an inventory of disposable cages and storing soiled cages until disposed The cost of disposing of the cages The cost of labor for receiving, storing, and disposing of the clean and soiled cages
11/11/10 6:25:17 PM
196
Handbook of Laboratory Animal Science
Non-disposable cage costs may include: • • • • •
The cost of the space required to sanitize the cages The cost of the nondisposable cages, initial and replacement The cost of the equipment for sanitizing cages The cost of utilities and chemicals required to sanitize the cages The labor costs to sanitize the cages
These are just the major factors; there are many other minor and locality-related factors to be considered. Much emphasis has recently been given to environmental enrichment for all species, including rodents. The value of various types of enrichment for the well-being of rodents is controversial and will probably be a source of debate for years to come. Experiments examining different ways to increase the complexity of the primary environment for housing research animals are currently being performed (Figure€8.23). The challenge is to balance seemingly conflicting requirements to provide for animal well-being through cage enrichment with features required to provide for routine animal care and cage sanitation, to assure animal health, and to successfully achieve the research goals. Solid-bottom, shoebox-type cages may be covered with wire bar lids that leave the cage interior open to the room environment or with filter tops that provide a barrier between the cage microenvironment and the room macroenvironment that prevents the spread of airborne infections between cages. The most common type of filtered cage top is a rigid inverted shoebox with an air filter insert that covers the top of the cage and overlaps the sides to form the microbiologic equivalent of a petri dish. These are known as microisolation cages. Microisolation cages have proven to be effective at protecting rodents from microbial contamination especially when opened for any reason only inside a HEPA-filtered mass air displacement cabinet (Figure€ 8.24; Lipman 2009; Lipman, Newcomer, and Fox 1987; Dillehay, Lehner, and
Figure 8.24â•…A rack with static microisolator cages adjacent to a Class II, Type A biosafety cabinet with a microisolator cage inside the cabinet. The combination of the microisolation cage and cabinet makes up what is known as a “microisolation cage system.” The microisolation cages are only opened inside the cabinet, including for cage changes or research procedures.
84550.indb 196
11/11/10 6:25:20 PM
Laboratory Animal Facilities and Equipment
197
Huerkamp 1990; Borello et al. 2000; Whary et al. 2000; Otto and Talwani 2002). However, in the static air microenvironment of the cage, animals are subjected to elevated levels of ammonia, carbon dioxide, moisture, and heat (Lipman 1992, 2009; Corning and Lipman 1991; Hasenau, Baggs, and Kraus 1993). The longer the time between cage changes is, the higher the ammonia levels will be. There are many variables that affect the rate of ammonia production and ammonia levels in the cage (e.g., days between cage changes, relative humidity levels, and the type of bedding); in general, two changes per week is considered the minimum number of changes for a static microisolator cage housing the maximum capacity of mice. Ventilated microisolation caging was designed to better control the microenvironment (Lipman 1999, 2009; Novak and Sharpless 2001). Directly ventilating each microisolation cage with HEPA filtered air significantly slows the buildup of ammonia, etc. in the cage, thus decreasing the cage change frequency to once a week or even once every 2 or 3 weeks (Keller et al. 1989; Huerkamp, Dillehay, and Lehner 1994; Hasegawa et al. 1997; Reeb-Whittaker et al. 2001; Baumans et al. 2002). Ventilated cage racks are fitted with filter/fan units supplying HEPA-filtered air to each cage through a system of manifolds. Some ventilated racks also capture the air coming from each cage and HEPA filters it before dumping it back into the room (Figures€8.13 and 8.14) or, even better, directs the air from the cages directly into the room exhaust (Figures€8.12 and 8.15). Some ventilated microisolation cage racks that control the cage supply and exhaust air can selectively maintain the air pressure in the cage positively or negatively relative to the room air pressure. When exhausted directly into the room exhaust ducts, HEPA filtration is not necessary, but some filtration of discharge air is desirable to reduce the amount of dust dumped into the exhaust ducts and especially onto the heat recovery system. A significant “fringe benefit” of ventilated cage racks that HEPA-filtered air coming from the cages or directly exhaust the cage air from the room is the reduction of animal allergens in the workers’ environment because allergies to animal allergens are the most significant occupational hazard for personnel working with animals (Huerkamp et al. 2009; Reeb-Whittaker and Harrison 1999; Renström, Björing, and Höglund 2001). If the discharge air is directed into the room exhaust system, odors generated in the cage are also discharged. Another advantage of ventilated caging is that it allows for high-density housing. Typical mobile, double-sided, ventilated mouse cage racks commonly hold up to 140 cages (seven cages wide × 10 rows high × two sides); each cage has the capacity for up to five mice. Other large and smaller rack configurations are available, as are single-sided racks. Fixed racks (without wheels) may even be stacked higher, further increasing housing density but also increasing the ergonomic problems and possibly injuries associated with reaching the upper cages. Rabbits The traditional way to keep rabbits is to house them one to a cage to facilitate handling, preclude fighting, and minimize the risk for spread of infections. During the last 10 years, there has been movement away from fabricating rabbit cages with stainless steel toward warmer plastics, such as NORYL® (GE), which is opaque, lightweight, and nontoxic and withstands the sanitizing temperatures of cage washers. In-cage enrichment strategies have included a shelf in the cage, making it possible for the rabbits to utilize more of the cage volume than just the floor and to find a hiding or “burrowing” place underneath the shelf. Single housing is usually the sole alternative for housing male rabbits, which have a strong tendency to fight when housed together. Female rabbits housed together may also fight, but with careful management and observation to reduce fighting and the resulting fight wounds, it is possible to house them successfully in groups of two or more in cages or on the floor (Figure€8.25). Some rabbit cage racks are designed with pairs of cages separated with a removable divider. This allows doubling the floor area for housing two adult females together or for using it as a kindling cage and for housing a female and her nursing young. Pair housing females is especially beneficial when rabbits are kept for longer periods, such as those that are used for antibody production.
84550.indb 197
11/11/10 6:25:20 PM
198
Handbook of Laboratory Animal Science
Figure 8.25â•…Rabbit group housed on the room floor. Certain types of studies that involve holding rabbits for long periods of time, such as polyclonal antibody production protocols, allow for group housing of rabbits. Experience shows that most female rabbits readily adapt to being housed in groups, where they benefit from social interactions and exercise.
Canines, Swine, and Small Ungulates Housing dogs, pigs, sheep, and goats (referred to here as “large animals”) is lumped under a single housing category because they have similar housing requirements, including requiring relatively large cages or pens and a room that accommodates a hose-down sanitation system, and adapting well to group housing. However, when the research protocol requires housing them individually, this must be provided. This commonality is most convenient for biomedical research facilities that cannot be certain which species they will need to accommodate at any given time. A description of a generic room designed for hose-down sanitation is included in the plumbing/floor drain section of this chapter, including the slope of the floors, location of drains and drain troughs, and the location and type of water hoses. Within this generic room, any variety of portable pens with raised floors can be placed for housing any of the large animal species noted previously. The pens can be designed to group or individually house the animals by adding or removing pen dividers. Alternatively, instead of the pens, double-decked dog cages could be used in the room, as could most mobile primate cages or pens. Another useful and highly flexible large animal housing concept is large animal cubicles. This is a relatively new application of the animal cubicle design concept applied to large animal species requiring a hose-down sanitation system that is described in detail in Hessler and Britz 2009 and Hessler 1991c and 1993. It also divides a larger room into cubicles sized to house two or more large animals at a time, depending on the cubicle size, in fixed or mobile pens on raised floors with floors sloping toward a drain trough in back of the cubicle. Large animal cubicles allow for housing multiple species within a single room. These cubicles are also useful for housing poultry, cats, and other species. Facilities that are dedicated to a specific large animal species may be significantly different from the generic room described earlier. There are many options for housing these species, including a wide variety of indoor–outdoor pen configurations that are more applicable to rural settings than to a conventional urban research animal facility. For example, Figures€8.8 and 8.9 illustrate
84550.indb 198
11/11/10 6:25:21 PM
Laboratory Animal Facilities and Equipment
199
a facility for housing dogs that is divided into three parts: indoor pens with resting places, indoor cages where the dogs can be housed individually when fed, and outdoor pens for exercise. In many localities, dogs must be provided the space and opportunity to exercise daily. While species are typically separated for housing, pigs and dogs have proven to be compatible when sharing an outside exercise pen (Figure€8.10). See Casebolt (2009) for additional information regarding housing of large animal species. Nonhuman Primates There are many, highly varied indoor and outdoor options for housing nonhuman primates in cages and pens, many of which are described elsewhere (Bohm and Kreitlein 2009; Kelley and Hall 1995). Group or pair housing is most desirable for most species of nonhuman primates; however, single housing in stainless steel caging has been the more traditional way to keep large nonhuman primates such as macaques. Quad primate cage racks that have four housing compartments with removable partitions between the two top and two bottom compartments and between both the left and right top and bottom compartments are particularly versatile and useful not only for pair or group housing but also for giving the primates the opportunity to interact and become socialized before being housed together. Such caging is best accommodated in a room such as the generic room noted in the preceding section that provides appropriately sloped floors, drainage, and sources of water for cleaning. The main advantage for single caging of the larger primate species is safety, both for personnel and animals. These animals are strong, have sharp teeth, and may carry zoonotic diseases, including some that are deadly for humans (e.g., herpes B virus in macaques). Experimental procedures often require frequent handling of the animals. Individually housed animals can be more safely restrained and captured. In addition, the single animal per cage housing regime avoids the risk of severe harm that these large animals inflict on one another when fighting. In spite of these serious considerations, pair or group housing is preferable when possible because these are social, highly developed animals with needs for exercise and environmental complexity. For these reasons, individual caging is best limited to situations requiring individual housing based on scientific requirements and individual animal behavior. The most optimal housing may be to have two primates in a large cage with branches, toys, and bedding material into which food may be placed to stimulate foraging behavior. A transport cage may be attached to such cages, and the animals may be urged to go into the transport cage with some training. With two primates in one cage, it would be simple enough to get the primate wanted for a particular experiment without too much distress of nonhuman primate and personnel. It usually is possible to find two animals that like to be housed together. Larger groups of nonhuman primates may be formed even from animals that have no or little experience of group housing. Care must be taken to find a good structure of such a group, which needs a strong and reliable leader. Cats Cats, especially females, are also well suited to group housing (Figure€8.26). A regular animal room furnished with shelves, boxes, tunnels, branches, or other items where the animals may hide or climb is suitable for group housing of cats and is commonly used. For some types of studies, it is necessary to house cats individually (e.g., if they have been subjected to some kind of surgical procedure or if it is necessary to have male cats). In these cases, stainless steel cages with wood or plastic resting boards are commonly used.
84550.indb 199
11/11/10 6:25:21 PM
200
Handbook of Laboratory Animal Science
Figure 8.26â•…Group housing of cats. Most female cats readily adapt to group housing. The room is a regular laboratory animal facility animal room furnished with shelves and tunnels.
Equipment and Caging for Biological Control (Barrier and Containment) Biological Safety Cabinets Biological safety cabinets (BSCs) provide primary containment and protection when staff are working with infectious agents. The CDC-NIH publication “Biosafety in Microbiological and Biomedical Laboratories” (USDHHS 2007) includes a detailed description of the various types of BSCs along with installation requirements. There are three classes of biosafety cabinets: classes I, II, and III. Class I and II BSCs have inward airflow at velocities of 75–100 linear ft per min through an open front. Exhaust air from the cabinet passes through HEPA filters before being discharged into the room or into the laboratory exhaust system. Class II BSCs have the additional benefit of protecting objects in the cabinet from extraneous microbial contamination. Class I and II BSCs are suitable for working with up to BSL-3 infectious agents. Class III BSCs provide the highest degree of personnel and environmental protection from infectious aerosols, as well as protection from extraneous microbiological contaminants for the materials in the cabinet. They essentially are a totally enclosed, gas-tight, ventilated cabinet. All operations in the cabinet are performed through attached rubber sleeves with surgical-type gloves. Supply air is HEPA filtered, and exhaust air is filtered through two HEPA filters. Class III BSCs are suitable for working with infectious agents classified at the highest biosafety level, BSL-4. The class II BSC (Figure€8.24) is designed with inward airflow at a velocity to protect personnel (75–100 lfpm), HEPA-filtered downward vertical laminar airflow for product protection, and HEPA-filtered exhaust air for environmental protection. Class II BSCs are the type most commonly used in research animal facilities. They come in two types, A and B. Class II type A cabinets do not have to be vented, so they can be mobile. They are acceptable for use with low to moderate risk agents in the absence of volatile toxic chemicals and volatile radionuclides. They are not suitable for use with gas anesthetic agents unless exhausted via a thimble
84550.indb 200
11/11/10 6:25:22 PM
Laboratory Animal Facilities and Equipment
201
connection to the building exhaust ductwork. Class II type A BSCs are often used as rodent animal transfer stations (see the next section). Class II type B cabinets are hard-ducted to the building exhaust system and contain negative pressure plena. They come in three types: B1, B2, and B3, which are differentiated primarily by the percentage of air recirculated within the cabinet versus that directly exhausted into the building exhaust system, i.e., B1, 70% recirculated, 30% exhausted; B2, 30% recirculated, 70% exhausted; and B3, 100% exhausted. These features, plus a face velocity of 100 lfpm, allow work to be done with toxic volatile chemicals or radionuclides. Type B1 cabinets are suitable for working with anesthetic gases as well as infectious agents up to BSL-3. Of course, the higher the percent exhausted, the greater the control of volatile hazards is. Microisolation Caging System The microisolator caging system includes three parts: (1) a microisolation cage, static or ventilated; (2) a HEPA-filtered air cabinet, often referred to as animal transfer stations (ATSs) (Figure 8.12) to be used anytime the cages are opened, either to transfer the animals to a clean cage or to perform experimental procedures with them; and (3) rigorous operational sanitation procedures to be used immediately prior to and during the time when the cage is opened. Microisolation cages for rodents have already been described in this chapter as being highly effective in both barrier (keep in) and containment (keep out) environments. Static microisolation caging may be more suited for containment of high-level hazardous agents than ventilated microÂ� isolation cages unless the ventilated cages are tightly sealed. Regarding the ATS, class II type A biocontainment cabinets (Figure€8.24) are well suited for use as ATSs in barrier and containment situations since they are designed to protect the operator’s environment outside the cabinet from agents in the cabinet, as well as the “product” inside the cabinet—in this case, the mice in the opened cage—from contaminants outside the cabinet. Why use a biosafety cabinet in a barrier when there are no hazardous agents with which to be concerned ? Because it is assumed that there will be a disease “break” (e.g., mouse hepatitis virus). When this happens, the management objective is to contain the offending infectious agent until it can be detected and eliminated. There are ATSs available that reportedly do both, keeping agents from getting into or escaping from the cabinet, that are more user friendly, but that do not qualify to be classified as biocontainment cabinets. Which one is chosen will depend on how much emphasis one places on “containment” when working in a “barrier” setting. The rigorous operational sanitation procedures include wearing appropriate personal protective equipment (PPE) or, at a minimum, protective sleeves to cover the arms and the hands with gloves; opening one cage at a time in the cabinet except when both a soiled and a clean cage are required when transferring animals to a clean cage; and, between working with cages in the cabinet, using a fast-acting, high-level disinfectant on the cabinet work surface, the outside of the cage, the gloved hands, and any instrumentation that touches the animals. Isolators Isolators provide the most effective protection against the spread of infectious agents—whether used as a barrier to protect an animal in the isolator, even to the point of maintaining “germ-free” or gnotobiotic animals, or for containment to protect the macroenvironment (room) from biohazards in the isolator. They are of two basic types: flexible film isolators (Figure 8.27) and rigid wall isolators (Figure 8.28). The original germ-free isolators were rigid stainless steel cylinders, but most rigid isolators in use today are made with clear polycarbonate. Currently, most isolators are made from transparent polyvinyl-chloride (PVC) flexible plastic sheeting supported by a metal or plastic frame or only by positive pressure air, but rigid isolators made of stainless steel or polycarbonate are also used. Air, filtered either
84550.indb 201
11/11/10 6:25:22 PM
202
Handbook of Laboratory Animal Science
Figure 8.27â•…Flexible film isolator. The isolator in the photo has a large, square entrance port, which is easier to load than the more common circular ports. Many flexible film isolators do not have a rigid frame or a rigid port as shown in this figure.
with HEPA filters or other highly efficient filter media, is blown into the isolator, and when used for containment, the air coming from the isolator is also filtered. Isolators come in many sizes. In effect, the isolator is a room within a room that can be provided with sterile air, water, diet, cages, bottles, etc., thus minimizing the risk of contaminating animals with infectious agents from outside the isolator. Personnel work inside the isolators through portholes fitted with sleeves and
Figure 8.28â•…A rigid stainless steel isolator that was specially adapted to perform hysterectomies for rederivation of animals. The isolator is divided into two parts: one in which the hysterectomy is performed and one where the pups are resuscitated. The two parts are separated with a dip tank containing a disinfectant.
84550.indb 202
11/11/10 6:25:23 PM
Laboratory Animal Facilities and Equipment
203
gloves made of latex or other similar material. A well-established system involving equipment, ports, and procedures is used for moving sterile materials into the isolator and soiled materials and trash out of the isolator. Autoclaves Autoclaves were covered previously in this chapter. They are an essential component of an effective barrier or containment facility, required for sterilizing cages, cage parts, bedding, feed, and other supplies. In the case of barriers, cages and supplies are autoclaved into the barrier area. In the case of containment, they are autoclaved out of and, in many cases, also autoclaved into the containment area. If the autoclave is not an integral part of the barrier or biocontainment area, items to be autoclaved must be wrapped or otherwise protected or sealed to transport them safely between the autoclave and the barrier or containment area. This is less effective and significantly more labor intensive than having the autoclaves as an integral part of the barrier or containment perimeter.
Conclusion The emphasis in this chapter has been on research animal facilities required to support contemporary biomedical research, but it must be noted that sound management of the facilities is at least equally important. Control of some of the environmental factors noted in Figure€8.1 requires properly designed facilities, but all require sound management. The better the facility is designed to facilitate sound management, the lower the cost of animal care will be and the more likely environmental variables will be adequately controlled. It is theoretically possible for good management to overcome design features contrary to efficient management, but even the best management cannot overcome human nature, which dictates that if a routine task is difficult to do, this will guarantee that it will not be routinely done. A properly controlled environment for research animals means more reliable and reproducible research, thus reducing the number of animals required to achieve the research goals. Good facility design and sound management facilitate high-quality contemporary biomedical research, sound research economics, and, most importantly, humane care and use of laboratory animals. References ASHRAE (American Society of Heating, Refrigerating and Air Conditioning Engineers). 2007. Environmental Control for Animals and Plants. In 2007 ASHRAE Handbook: HVAC applications, chap. 22. Atlanta: ASHRAE. AVMA (American Veterinary Medical Association). 1993. Guidelines for animal surgery in research and teaching. American Journal of Veterinary Research 54:1544–1559. Baker, H. J., J. R. Lindsey, and S. H. Weisbroth. 1979. Housing to control research variables. In The laboratory rat: vol. I: Biology and diseases, ed. H. J. Baker, J. R. Lindsey, and S. H. Weisbroth, 169–192. New York: Academic Press. Barkley, W. E. 1997. Abilities and limitations of architectural and engineering features in controlling biohazards in animal facilities. In Symposium on Laboratory Animal Housing. Institute for Laboratory Animal Research, National Academy of Science, National Research Council, 158–163. Washington, D.C.: National Academy Press. Barkley, W. E., and J. H. Richardson. 1984. Control of biohazards associated with the use of experimental animals. In Laboratory animal medicine, ed. J. G. Fox, B. J. Cohen, and F. M. Loew, 595–602. Orlando, FL: Academic Press.
84550.indb 203
11/11/10 6:25:23 PM
204
Handbook of Laboratory Animal Science
Baumans, V., F. Schlingmann, M. Vonck, and H. A. Van Lith. 2002. Individually ventilated cages: Beneficial for mice and men? Contemporary Topics in Laboratory Animal Science 41:13–19. Bilecke, B. 2001. Integrating ventilated caging equipment with facility HVAC and monitoring systems. Lab Animal Fall: 42–47. Bohm, R. P., Jr., and E. S. Kreitlein. 2009. Facilities for nonhuman primates. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 289–312. San Diego, CA: Elsevier Inc. Borello, P., E. D’Amore, G. Panzini, V. Mauro, and L. R. Nello. 2000. Individually ventilated cages—Microbiological contaminant testing. Scandinavian Journal of Laboratory Animal Science 27:142–152. Brown, M. J. 1994. Aseptic surgery for rodents. In Rodents and rabbits: Current research issues, ed. S. M. Niemi, J. S. Venable, and H. N. Guttman, 67–72. Bethesda, MD: Scientist Center for Animal Welfare. Canadian Council on Animal Care (CCAC). 1993. Guide to the care and use of experimental animals, vol. 1, 2nd ed. Ottawa: CCAC. ———. 2003. Guidelines on laboratory animal facilities—Characteristics, design, and development. Ottawa: CCAC. Carlton, D. L. 2002. Affordable noise control in a laboratory animal facility. Lab Animal 31:47–48. Casebolt, D. B. 2009. Facilities for dogs, swine, sheep, goats and miscellaneous species. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 313–322. San Diego, CA: Elsevier Inc. Conti, P. A., and J. R. Hessler. 2009. Circulation. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 95–106. San Diego, CA: Elsevier. Corey, M., R. Davey, and R. Faith. 2001. Case study: Automated cage wash design at Baylor College of Medicine. Lab Animal Fall: 32–35. Corning, B. F., and N. S. Lipman. 1991. A comparison of rodent caging systems based on microenvironmental parameters. Laboratory Animal Science 41:498–503. Council of Europe, European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes, Council of Europe (Convention ETS 123). 1985 (adopted May 1999). Cunliffe-Beamer, T. L. 1993. Applying principles of aseptic surgery to rodents. AWIC Newsletter 4:2–6. Curry, G., H. C. Hughes, D. Loseby, and S. Reynolds. 1998. Advances in cubicle design using computational fluid dynamics as a design tool. Laboratory Animals 32:117–127. Dauchy, R. T. 2007. Animal facility dark-phase light contamination and circadian rhythm disruption. Techtalk 12:1–2. Dauchy, R. T., E. M. Dauchy, L. K. Davidson, J. A. Krause, D. T. Lynch, P. C. Terrell, R. P. Tirrell, L. A. Sauer, P. V. Riet, and D. E. Blask. 2007. Inhibition of fatty acid transport and proliferative activity in tissueisolated human squamous cell cancer xenografts perfused in situ with melatonin or eicosapentaenoic or conjugated linoleic acids. Comparative Medicine 27:377–382. Dauchy, R. T., L. A. Sauer, D. E. Blask, and G. M. Vaugnan. 1997. Light contamination during the dark phase in “photoperiodically controlled” animal rooms: Effect on tumor growth and metabolism in rats. Laboratory Animal Science 47:511–518. Dillehay, D. L., N. D. Lehner, and M. J. Huerkamp. 1990. The effectiveness of a microisolator cage system and sentinel mice for controlling and detecting MHV and Sendai virus infections. Laboratory Animal Science 40:367–370. Dolowy, W. C. 1961. Medical research laboratory of the University of Illinois. Proceedings of the Animal Care Panel 11:267–290. Dysko, R. C., M. J. Huerkamp, K. E. Yrjanainen, S. Smart, R. Curran, C. J. Maute, and W. D. Thompson. 2009. Plumbing: Special considerations. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 425–454. San Diego, CA: Elsevier Inc. Evans, M. E., and J. A. Lesnaw. 1999. Infection control in gene therapy. Infection Control and Hospital Epidemiology 20:568–576. Faith, R. E., M. A. Corey, and R. Nelan. 2009. Animal care and administration space. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 187–202. San Diego, CA: Elsevier Inc. Faith, R. E., and J. R. Hessler. 2006. Housing and environment. In The laboratory rat, 2nd ed., ed. M. A. Suckow, S. H. Weisbroth, and C. L. Franklin. San Diego, CA: Elsevier Inc. Faith, R. E., and M. J. Huerkamp. 2009. Environmental considerations for research animals. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 59–84. San Diego, CA: Elsevier Inc.
84550.indb 204
11/11/10 6:25:23 PM
Laboratory Animal Facilities and Equipment
205
Federation of Animal Science Societies (FASS). 1999. Guide for the care and use of agricultural animals in agricultural research and teaching, 1st rev. ed. Savoy, IL: FASS. Fullerton, F. R., D. I. Greenman, and D. C. Kendall. 1982. Effects of storage conditions on nutritional qualities of semipurified (AIJN-76) and natural ingredient (NIH-07) diets. Journal of Nutrition 12:567–573. Gordon, C. J. 1993. Temperature regulation in laboratory animals. New York: Cambridge University Press. Hasegawa, M., Y. Kurabayashi, T. Ishii, K. Yoshida, N. Eubayashi, N. Sato, and T. Kurosawa. 1997. Intracage air change rate on forced-air-ventilated microisolation system-environment within cages: Carbon dioxide and oxygen concentrations. Experimental Animals 46:251–257. Hasenau, J. J., R. B. Baggs, and A. L. Kraus. 1993. Microenvironments in microisolation cages using BALB/c and CD-1 mice. Contemporary Topics in Laboratory Animal Science 32:11–16. Hessler, J. R. 1991a. Single- versus dual-corridor systems: Advantages, disadvantages, limitations, and alternatives for effective contamination control. In Handbook of facilities planning, vol. 2, Laboratory animal facilities, ed. T. Ruys, 59–67. New York: Van Nostrand Reinhold. ———. 1991b. Facilities to support research. In Handbook of facilities planning, vol. 2, Laboratory animal facilities, ed. T. Ruys, 35–54. New York: Van Nostrand Reinhold. ———. 1991c. Animal cubicles. In Handbook of facilities planning, vol. 2, Laboratory animal facilities, ed. T. Ruys, 135–154. New York: Van Nostrand Reinhold. ———. 1993. Animal cubicles: Questions, answers, options, opinions. Lab Animals 22:21–32. ———. 1995. Methods of biocontainment. In Current issues and new frontiers in animal research, ed. K. A. L. Bayne, M. Greene, and E. D. Prentice, 61–68. Greenbelt, MD: Scientist Center for Animal Welfare. ———. 1999. The history of environmental improvements in laboratory animal science: Caging systems, equipment, and facility design. In Fifty years of laboratory animal science, ed. C. McPherson and S. Mattingly, 92–120. Memphis, TN: American Association for Laboratory Animal Science. ———. 2009a. Functional adjacencies. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 107–108. San Diego, CA: Elsevier Inc. ———. 2009b. Barrier housing for rodents. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 135–146. San Diego, CA: Elsevier Inc. ———. 2009c. Electrical: Special features. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 455–460. San Diego, CA: Elsevier Inc. Hessler, J. R., and W. R. Britz. 2009. Animal isolation cubicles. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 151–172. San Diego, CA: Elsevier Inc. Hessler, J. R., R. Broderson, and C. King. 1999a. Animal research facilities and equipment. In Anthology of biosafety 1. Perspectives on laboratory design, ed. J. Y. Richmond, 191–218. Mundelein, IL: American Biological Safety Association. ———. 1999b. Rodent quarantine: Facility design and equipment for small animal containment facilities. Lab Animal 28:34–40. Hessler, J. R., and D. P. Frasier. 2009. Heating, ventilation and air conditioning (HVAC): Special considerations. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 461–478. San Diego, CA: Elsevier Inc. Hessler, J. R., and H. Höglund. 2002. Laboratory animal facilities and equipment for conventional, barrier, and containment housing systems. In Handbook of laboratory animal science, vol. 1, Selection and handling of animals in biomedical research, 2nd ed., ed. J. Hau and G. Van Hoosier, 127–172. London: CRC Press. Hessler, J. R., and S. L. Leary. 2002. Design and management of animal facilities. In Laboratory animal medicine, ed. J. G. Fox, B. J. Cohen, and F. M. Loew, 909–953. Orlando, FL: Academic Press. Hessler, J. R., and N. D. M. Lehner, eds. 2009. Planning and designing research animal facilities. San Diego, CA: Elsevier Inc. Hessler J. R., and A. F. Moreland. 1984. Design and management of animal facilities. In Laboratory animal medicine, ed. J. G. Fox, B. J. Cohen, and F. M. Loew, 505–527. Orlando, FL: Academic Press. Howard, H., and Y. K. Foucher. 2009. Animal-use space. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 203–264. San Diego, CA: Elsevier Inc. Huerkamp, M. J., D. L. Dillehay, and N. D. M. Lehner. 1994. Comparative effects of forced air, individual cage ventilation or an absorbent bedding additive on mouse cage microenvironment. Contemporary Topics in Laboratory Animal Science 33:58–61.
84550.indb 205
11/11/10 6:25:24 PM
206
Handbook of Laboratory Animal Science
Huerkamp, M. J., M. A. Gladle, M. P. Mottet, and K. Forde. 2009. Ergonomic considerations and allergen management. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 115–128. San Diego, CA: Elsevier Inc. Hughes, C. H., S. Reynolds, and M. Rodrigues. 1996. Designing animal rooms to optimize airflow using computational fluid dynamics. Pharmaceutical Engineering March/April: 44–65. ILAR (Institute of Laboratory Animal Resources), Commission on Life Sciences, National Research Council, National Academy of Sciences. 1996. Guide for the care and use of laboratory animals. Washington, D.C.: National Academy Press. Jackson, C. W., J. Rehg, C. Rock, R. Henning, and S. Reynolds. 2002. Computational fluid dynamics optimizes ventilation in animal rooms. Lab Animal Fall: 50–53. Keller, L. S. F., W. J. White, M. T. Snider, and C. M. Lang. 1989. An evaluation of intracage ventilation in three animal caging systems. Laboratory Animal Science 39:237–241. Kelley, S. T., and A. H. Hall. 1995. Housing. In Nonhuman primates in biomedical research: Biology and management, ed. B. T. Bennett, C. R. Abee, and R. Henrickson, 193–209. San Diego, CA: Academic Press. Klein, H. J., M. J. Kuntz, and J. R. Hessler. 2009. Special fixed equipment for research animal facilities. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 409–424. San Diego, CA: Elsevier Inc. Leary, S. L., J. A. Majoros, and J. S. Tomson. 1998. Making cage wash facility design a priority. Laboratory Animals 27:28–31. Lehner, N. D. M., J. T. Crane, M. P. Mottet, and M. E. Fitzgerald. 2009. Biohazards: Safety practices, operations and containment facilities. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 347–364. San Diego, CA: Elsevier Inc. Lempken, B., 1999. Drinking water. In Handbook of facilities planning, vol. 2, Laboratory animal facilities, ed. T. Ruys, 174–180. New York: Van Nostrand Reinhold. Leverage, N., and C. R. Roberts. 2009. Finish decisions. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 397–408. San Diego, CA: Elsevier Inc. Lipman, N. S. 1992. Microenvironmental conditions in isolator cages: An important research variable. Laboratory Animal Science 21:23–27. ———. 1993. Strategies for architectural integration of ventilated caging systems. Contemporary Topics in Laboratory Animal Science 32:7–10. ———. 1999. Isolator rodent caging systems (state of the art): A critical view. Contemporary Topics in Laboratory Animal Science 38:1–17. ———. 2007. Design and management of research facilities for mice. In The mouse in biomedical research, 2nd ed., 271–319. London: Elsevier. ———. 2009. Rodent facilities and caging systems. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 265–288. San Diego, CA: Elsevier Inc. Lipman, N. S., C. E. Newcomer, and J. G. Fox. 1987. Rederivation of MHV and MEV antibody positive mice by cross-fostering and use of the microisolator caging system. Laboratory Animal Science 37:195–199. Lipman, N. S., and S. E. Perkins. 2002. Factors that may influence animal research. In Laboratory animal medicine, ed. J. G. Fox, B. J. Cohen, and F. M. Loew, 1143–1184. San Diego, CA: Academic Press. Memarzadeh, F. 1998. Ventilation design handbook on animal research facilities using static microisolators. In Animal facility ventilation handbook, vols. I and II. Bethesda, MD: NIH, Division of Engineering Services. Meyer, J. S, and J. E. Mollo-Christensen. 2009. Vivarium esthetics and visual design. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 109–114. San Diego, CA: Elsevier Inc. Mollo-Christensen, J. E. 2009. Doors and hardware: Practical choices. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 389–396. San Diego, CA: Elsevier Inc. National Research Council, National Academy of Sciences. 1997. Occupational health and safety in the care and use of research animals. Washington, D.C.: National Academy Press Novak, G. 1999. Selecting an appropriate watering system for your facility. Lab Animal 28:43–46. Novak, G. R., and L. C. Sharpless. 2001. Selecting an individually ventilated caging system. Laboratory Animals Fall: 36–41. Otto, G., and R. J. Tolwani. 2002. Use of microisolator caging in a risk-based mouse import and quarantine program: A retrospective study. Contemporary Topics in Laboratory Animal Science 41:20–27. Reeb-Whitaker, C. K., and D. J. Harrison. 1999. Practical management strategies for laboratory animal allergy. Lab Animal 28:25–30.
84550.indb 206
11/11/10 6:25:24 PM
Laboratory Animal Facilities and Equipment
207
Reeb-Whittaker, C. K., B. Paigen, W. G. Beamer, R. T. Bronson, G. A. Churchill, I. B. Schweitzer, and D. D. Myers. 2001. The impact of reduced frequency of cage changes on the health of mice housed in ventilated cages. Lab Anim 35:58–73. Renström, A. G., G. Björing, and U. Höglund, U. 2001. Evaluation of individually ventilated cage systems for laboratory rodents: Occupational health aspects. Lab Anim 35:42–50. Reynolds, S. D. 2009. Using computational fluid dynamics (CFD) in laboratory animal facilities. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, 479–488. San Diego, CA: Elsevier Inc. Richmond, J. Y. 1991. Hazard reduction in animal research facilities. Lab Animals 20: 23–29. ———. 1996. Animal biosafety levels 1–4: An overview. In Proceedings of the Fourth National Symposium on Biosafety: Working Safely with Research Animals, ed. J. Y. Richmond, 5–8. Atlanta, GA: Centers for Disease Control. Roe, P. 2002. Cage processing and waste management: A cost-analysis and decision-making exercise. Lab Animal 31:43–46. Ruggiero, R. F. 2001. Considerations for an automated cage-processing system. Lab Animal Fall: 28–31. Ruys, T. 1988. Isolation cubicles: Space and cost analysis. Lab Animal 17:25–23. ———. 1991. Handbook of facility planning: vol. 2, laboratory animal facilities. New York: Van Nostrand Reinhold. Shalev, M., ed. 2001. Facility design and planning. Laboratory Animals (entire fall issue). USDHHS, PHS, Center for Disease Control and Prevention, and the National Institutes of Health. 2007. Biosafety in microbiological and biomedical laboratories, 5th ed. Washington, D.C.: U.S. Government Printing Office. USPH National Institutes of Health. 2009. Biomedical and animal research facilities design policies and guidelines. Available on the NIH web site at http://orf.od.nih.gov/PoliciesAndGuidelines/ BiomedicalandAnimalResearchFacilities DesignPoliciesandGuidelines/ Webber, D. J., and A. R. William. 1999. Gene therapy: A new challenge for infection control. Infection Control and Hospital Epidemiology 20:530–532. Whary, M. T., J. H. Cline, A. E. King, C. A. Corcoran, S. Xu, and J. G. Fox. 2000. Containment of Helicobacter hepaticus by use of husbandry. Comparative Medicine 50:78–81. White, W. J. 1996. Special containment devices for research animals. In Proceedings of the Fourth National Symposium on Biosafety: Working Safely with Research Animals, ed. J. Y. Richmond, 109–112. Atlanta, GA: Centers for Disease Control.
84550.indb 207
11/11/10 6:25:24 PM
84550.indb 208
11/11/10 6:25:24 PM
Chapter 9
Laboratory Animal Genetics and Genetic Quality Control
Michael F. W. Festing and Cathleen Lutz Contents The Development of Laboratory Animal Genetics........................................................................ 210 The Major Classes of Genetic Stocks............................................................................................. 212 Genetically Undefined Stocks.................................................................................................... 212 Outbred Stocks...................................................................................................................... 212 Genetically Heterogeneous Stocks, Segregating Hybrids, and Advanced Intercross€Lines................................................................................................................ 217 Outbred Selected Stocks....................................................................................................... 218 Partially Genetically Defined................................................................................................ 218 The Isogenic Family of Strains and Their Derivatives.............................................................. 218 Inbred Strains........................................................................................................................ 219 F1 Hybrids............................................................................................................................. 225 Coisogenic and Congenic Strains.......................................................................................... 226 Recombinant Inbred (RI) Strains.......................................................................................... 227 Recombinant Congenic (RC) Strains.................................................................................... 229 Consomic or “Chromosome Substitution” Strains................................................................ 230 Genetically Altered (GA) Animals (Mutants, Polymorphisms, Transgenes, and Targeted Mutations Including “Knockout” and “Knockin” Mutations)............................... 231 Breeding Transgenes and Mutations Produced by Gene Targeting (Knockouts)................. 237 Genetic Quality Control.................................................................................................................. 238 Aims........................................................................................................................................... 238 Technical Methods..................................................................................................................... 238 Genetic Monitoring of Isogenic Strains..................................................................................... 238 Authentication of Newly Established Strains........................................................................ 239 Existing Colonies.................................................................................................................. 239 Troubleshooting..................................................................................................................... 239 Genetic Monitoring of Outbred Stocks......................................................................................240 Comparisons of Different Colonies.......................................................................................240 References.......................................................................................................................................240 Appendix 9.1: Genetic Nomenclature.............................................................................................244 Outbred Stocks...........................................................................................................................244 209
84550.indb 209
11/11/10 6:25:24 PM
210
Handbook of Laboratory Animal Science
Inbred Strains.............................................................................................................................244 Coisogenic, Congenic, and Segregating Inbred Strains............................................................. 245 Recombinant Inbred Strains......................................................................................................246 Mixed Inbred Lines....................................................................................................................246 Recombinant Congenic Strains..................................................................................................246 Consomic or Chromosome Substitution Strains........................................................................246 Mutants, Polymorphisms, and Genetically Modified Loci........................................................ 247 Genetically Modified Loci......................................................................................................... 247 Transgenes.................................................................................................................................. 247 Targeted Mutations....................................................................................................................248 Appendix 9.2: Web Resources for Laboratory Animal Genetics...................................................248
The Development of Laboratory Animal Genetics The remarkable advances in the science of genetics over the past few decades are having a strong impact on laboratory animal science. These advances are due to the development of a range of molecular techniques that make it possible to sequence the whole genome and mutate specific genes in order to study their function. Progress has also been facilitated by the classical period of mouse genetics that started soon after the rediscovery of Mendel’s work in 1900 and continued until about the 1980s. This laid a firm foundation on which the new molecular methods could be based. A large proportion of “laboratory animal genetics” is in fact “mouse genetics” (Malakoff 2000). This does not imply that the genetics of other species is unimportant, but simply that, for many technical reasons, including small body weight, high reproductive performance, small space requirements, and the availability of a wide range of strains and mutants, the mouse has been used more extensively than any other mammalian species. In 1900, following the rediscovery of Mendel’s paper setting out the laws of inheritance of discrete characters in garden peas, the validity of these laws was soon confirmed by Cuenot (Silver 1995) and others using coat color in pet mice as the unit characters. Since that time, visible mouse mutants have often been preserved, even if the mutation appeared to be bizarre and had no obvious biomedical significance. The development of inbred strains of mice by C. C. Little, rats by Whilhemina Dunning (both in 1909), and guinea-pigs by G. M. Rommel (later taken over by Sewall Wright) made pure-breeding lines available for the first time, a major scientific advance (Grüneberg 1952). Several of these strains, as well as others developed by investigators in this early period, are still available and widely used in biomedical research. Inbreeding, usually due to many generations of brother × sister mating, has the effect of fixing the genotype within a strain, while maximizing the differences between strains. Some of these early strains were selected for a high incidence of various types of cancer, such as mammary tumors in C3H and DBA, leukemia in strain AKR, and lung tumors in strain A, and these strains were widely used in cancer research. As early as 1903, it was found that tumors could often be transplanted without rejection within the strain in which they originated, but were usually rejected when transplanted into a different strain (Klein 1975). These early studies were conducted using Japanese waltzing mice from fancy mouse breeders that had apparently accidentally become inbred. Subsequent studies by Little and Tyzzer (1916) showed that tumor rejection was dependent on a number of genes with a dominant mode of inheritance. George Snell, at the Jackson Laboratory in Maine, continued these studies and identified some of the gene loci responsible for this rejection by backcrossing the ability to resist tumor grafts from one strain into a strain where the grafts would normally be accepted (called the
84550.indb 210
11/11/10 6:25:25 PM
Laboratory Animal Genetics and Genetic Quality Control
211
inbred partner). His so-called “congenic-resistant” strains were usually found to differ from the inbred partner at a single genetic locus (Snell and Stimpfling 1966). In most cases, this was what is now known as the major histocompatibility complex (MHC), also designated the H2 locus or complex in mice. Not only did Snell identify this important locus, but his work also promoted the use of backcrossing of mutations to an inbred strain as a method of fixing the genotype to provide stable material for further study. Snell was awarded the Nobel Prize for medicine in 1980 for this work. He and his generation of transplantation immunologists developed several hundred of these congenic-resistant strains, which are still widely used in research involving the immune system. Congenic strains developed by backcrossing mutants and, more recently, transgenes to an inbred genetic background are now widely used. Another significant advance was the development of sets of “recombinant inbred strains” by Donald Bailey (1971). He crossed two standard inbred strains and then brother × sister mated the offspring for 20 generations so as to produce a set of new inbred strains in which the genes from the parental strains had recombined. His first set consisted of seven strains derived from a cross between inbred strains BALB/c and C57BL/6By. Larger sets of RI strains were later developed by B. A. Taylor (1976). They initially were used extensively to determine whether a given phenotypic (i.e., observed) difference between strains could be attributed to a single genetic locus and, if so, whether it was linked to any known genetic markers. Their use is discussed in more detail later. RI strains are quite good for resolving the genetics of characters controlled by one or two loci, but have some limitations for studying many characters with a polygenic mode of inheritance (i.e., where the phenotype depends on the joint action of several gene loci, each with a small effect as well as nongenetic factors such as environmental effects and accidents of development). This led Demant (Demant and Hart 1986) to develop sets of “recombinant congenic” strains specifically for this task by crossing two inbred strains then backcrossing for about two generations to one of the parental strains, followed by brother × sister inbreeding to produce about 20 new “RC” strains. These differ from an inbred partner by about 12.5% of those loci originally polymorphic between a donor and an inbred partner strain, assuming two backcross generations. This illustrates how specific strains can be developed as useful tools in biomedical research. An important feature of this classical period was the development of flexible and adaptable nomenclature rules for inbred strains, genetic loci, alleles, mutants, and chromosomes. These rules are administered by international nomenclature committees for mice and rats that attempt to prevent the same strain, mutant, etc. from being named differently by different investigators. Conversely, the committees also ensure that different strains or mutants do not end up with the same name. For rats, three competing nomenclature systems for the rat MHC were initially established. However, after considerable effort by the rat nomenclature committees, they have been reconciled into a new system. The maintenance of ever increasing numbers of mouse and rat strains is expensive in terms of space and scientific resources. This was alleviated, toward the end of this classical period, by the development of methods for freezing mouse embryos (Whittingham, Leibo, and Mazur 1972). Mouse embryos can be maintained in liquid nitrogen for many years, resulting in considerable savings on maintenance costs and space. The development of associated methods of handling preimplantation embryos has been critical to the progress made with transgenic strains. The development of molecular techniques from the 1980s, largely driven by the Human Genome Project, has had a number of important consequences. The cloning and sequencing of DNA taken together with the development of methods for handling early embryos, led to the development of methods for producing transgenic strains following the injection of foreign DNA into the pronucleus of early embryos (Gordon et al. 1980). Later, embryonic stem (ES) cell lines were developed from cultured early embryos (Evans and Kaufman 1981). These could be maintained and manipulated as cell
84550.indb 211
11/11/10 6:25:25 PM
212
Handbook of Laboratory Animal Science
cultures, allowing gene targeting by homologous recombination in order to develop “knockout” mice in which a specific gene was inactivated (Thomas and Capecchi 1987). This has proven to be a powerful tool for discovering the functions of many genes and has resulted in a remarkable proliferation of new strains. The 2007 Nobel Prize for medicine was awarded to Evans, Capecchi, and Smithies for developing embryonic stem cells and techniques to “knock out” or inactivate specific genes. Microsatellites are short, repetitive DNA sequences with unique flanking sequences. These are widely distributed throughout the mouse and rat genomes and provide a large number of genetic markers that have been used for genetic mapping and genetic quality control. These markers can also be used to map so-called quantitative trait loci (QTLs): loci controlling the inheritance of many complex characters, including susceptibility to numerous toxic agents, diseases (e.g., cancer, diabetes), and various aspects of behavior. Once these genes have been mapped to a general chromosomal location, they must then be identified. This usually requires extensive backcrossing programs using the methods employed by Snell to develop his congenic resistant strains. Unfortunately, such backcrossing procedures can take 2 or 3 years, but by using an array of microsatellite markers, “speed congenics” (see later discussion) can be developed in about half the time, although there are some organization, testing, and reagent costs. Keeping track of the vast amount of genetic information now being generated on gene and genome sequences, polymorphic genetic markers, genetic maps, new strains, and phenotypes would have been impossible without the parallel development of informatics, a separate discipline as of the mid-1990s. The Internet and World Wide Web have had a substantial impact on the way that science is conducted and communicated. Many Web sites offer relevant genetics information and resources at locations throughout the world. Some of these are listed in Appendix 9.2 at the end of this chapter. The Major Classes of Genetic Stocks Table€ 9.1 classifies the different genetic types of laboratory mice and rats into a genetically undefined group, a partially isogenic group, and an isogenic family. The properties of these three main classes of stock are discussed next. Most of the larger species of laboratory animals, such as primates, dogs, and cats, are only readily available as outbred stocks (outbred colonies are called stocks and inbred ones are called strains, although the term “strain” may also be used collectively to cover both strains and stocks). However, a few colonies carry mutations on an outbred background or have the genotype defined at some loci, such as the major histocompatibility complex. There are also some mutant stocks, inbred strains or selected stocks of dogs, rabbits, guinea pigs, chickens, and hamsters used in biomedical research. However, with all these species, the characteristics, research uses, and maintenance are essentially the same as those for mice and rats. Genetically Undefined Stocks Outbred Stocks An outbred stock is a breeding group of genetically heterogeneous animals, usually maintained as a closed colony without the introduction of animals from another stock or strain. The degree of genetic variation depends on the previous history of the colony. Colonies founded with a small number of pairs and/or that are maintained with relatively few breeding animals per generation will tend to lose genetic variation. In some cases, the colony may become moderately or highly inbred. Outbred stocks are available for all species of laboratory animals.
84550.indb 212
11/11/10 6:25:25 PM
Laboratory Animal Genetics and Genetic Quality Control
213
Table€9.1╅The Major Classes of Genetic Stocks Genetically Undefined Outbred stocks Genetically heterogeneous stocks Outbred selected stocks Segregating hybrids Partially Genetically Defined Mutants on an outbred background Transgenes on an outbred background Inbred strains in development Advanced intercross lines Genetically Defined Inbred strains Congenic strains Recombinant inbred strains Recombinant congenic strains Consomic (chromosome substitution) strains F1 hybrids Segregating inbred strains Clones Monozygous twins
Research Uses Outbred stocks continue to be widely used, even though in those species where there is a choice (mainly mice and rats), there is a compelling case for preferring isogenic strains (Festing 1995, 1999). For other species, there is generally no practical alternative. Outbred animals have some advantages. They are cheaper to buy or breed than isogenic strains, partly because they are more prolific and partly because they can be bred on a large scale with relatively little waste. As a result, they tend to be more readily available in groups of a defined weight or age range. Disadvantages include that each animal is genetically unique; thus, there is no information on the genotypes of individuals unless each is specifically genotyped. Phenotypic variation of outbred stocks is usually greater than that seen for isogenic strains because individuals differ due to both genetic and nongenetic factors. This means that a larger number of outbred than inbred animals are typically needed to achieve a given level of statistical precision. Also, stocks are subject to genetic change as a result of inbreeding and/or directional selection, leading to changes in gene frequency. Over a period of many generations, outbred mice and rats have become larger than their isogenic counterparts because small animals tend to be culled as “runts.” Inbred strains cannot be altered by such selection but outbred stocks can change rapidly over a period of a few generations, meaning that background data on the phenotypic characteristics of the stock may quickly lose their validity (Papaioannou and Festing 1980). In particular, there is good evidence that colonies of rats with the same name (such as “Sprague–Dawley”) from different breeders are often genetically distinct (Ito et al. 2007). This can cause difficulties when an attempt is made to replicate another laboratory’s work because nominally identical animals may, in fact, be genetically distinct. Unfortunately, genetic quality control in outbred stocks is difficult for several reasons: • There are no standards specifying what genes a stock such as a Wistar rat should have, so it is not possible to determine whether an individual rat is or is not a Wistar.
84550.indb 213
11/11/10 6:25:25 PM
214
Handbook of Laboratory Animal Science
• Although it is possible to identify a large number of genetic markers (genes or noncoding DNA) in an outbred stock, this requires large sample sizes because some genes or markers will be rare and only present in a few individuals. • Outbred stocks are quite labile so they will change over time; however, the rate of change depends on complex interactions between colony size, natural and artificial selection, and colony structure. • Colonies of white rats or white mice can sometimes get mixed up.
For these reasons and because customers rarely demand any genetic quality control of the outbred stocks they order, few breeders do any genetic quality control of outbred stocks. Outbred stocks are mainly used in general research for studies using species where there is no isogenic alternative, studies where genotype is judged to be of little importance, in noncritical studies, and in cases where the research worker is unaware of the advantages of isogenic strains. Outbred animals, which typically breed well, are also sometimes used as foster mothers. They are widely used in toxicological testing, though their use in such studies has been questioned (Kacew and Festing 1996; Festing 1997). When animals of a broad range of phenotypes are wanted, it is often better to use small numbers of several isogenic strains rather than outbred animals of a single stock (Festing 1997). However, outbred stocks are suitable for within-animal experiments, such as crossover experimental designs (i.e., where treatments are applied sequentially to the same animal; see Chapter 13, this volume) or those comparing left and right sides, such as an experiment in which one eye is used as a control for some treatment to the other eye. In this case, precision is not affected by differences between animals. They may also be suitable for experiments requiring sources of live tissue and for the maintenance of parasites, assuming host–parasite relationships are not of interest. They continue to be widely used in neuroscience, though this may be due more to ignorance of the advantages of isogenic animals rather than any good scientific argument for their continued use. A discussion of the relative merits of isogenic and nonisogenic stocks in aging research, which is relevant to many other fields of research, is given in Miller and colleagues (1999) and Festing (1999). The nomenclature of outbred stocks is given in Appendix 9.1. Breeding and Maintenance The aim in maintaining an outbred colony is usually to prevent genetic change. Alternatively, the aim may be to change the colony in some way; this can usually be accomplished by selective breeding. In any colony, genetic change occurs as a result of change in the frequency of genes (or, more correctly, alleles) at the various polymorphic genetic loci in the animals in the colony. For example, in a colony of 50 individuals, there will be a total of 100 genes at any given locus, since each individual has one gene from each of its two parents. Suppose that the frequency of alleles of type A is 90%. The frequency of alleles of type a will then account for the remaining 10% of alleles. Genetic changes occur if these frequencies change as a result of chance (genetic drift), mutation, or selection. The so-called Hardy–Weinberg law (Falconer 1981) states that in a large population with random mating and in the absence of selection, if the frequency of the A gene is p and that of the a gene is 1-p, then the frequency of individuals of type AA, Aa, and aa will be given by the equation p2 AA, 2p(1-p)Aa, and (1-p)2 aa, respectively. Therefore, if the frequency of type A is 0.9 and of a is 0.1, then there will be, on average, 81% AA, 18% Aa, and 1% aa animals in the colony. Within an outbred stock, individual animals will often differ at many thousands of loci. If the frequencies of individual alleles change, then the phenotype of the colony may also change. The four main causes of genetic change in an outbred stock are discussed in the following subsections. Genetic Contamination or Immigration. An outcross is sometimes done deliberately if the breeding performance of a colony declines. However, the introduction of new genetic material may
84550.indb 214
11/11/10 6:25:25 PM
Laboratory Animal Genetics and Genetic Quality Control
215
alter the characteristics of the colony. Genetic contamination may also occur by accident if the colony is maintained in physical proximity to another strain or stock. This may not be detected in the absence of routine genetic quality control. Mutation. Mutations occur at the rate of about 1 in 100,000 to 1 in 1,000,000 per locus. Some of these will become established within the colony, depending on chance and whether or not they are deleterious or advantageous. Recessive genes may be maintained within the colony for many generations, even if they have large phenotypic effects. For example, the Rowett athymic nude rat mutation apparently occurred at some time prior to 1955 in a colony of outbred rats at the Rowett Research Institute in Scotland. At that time, hairless athymic rats were discovered, but they did not survive. However, in 1978, two heterozygotes were apparently mated by chance, producing some nude young that were successfully reared (Festing et al. 1978). Directional Selection. If the stock is selected for a specific phenotype, such as large body weight, high reproductive performance, or enhanced immune response to some antigen, then the frequency of alleles associated with these characters will change, thereby altering the characteristics of the colony. The rate of change will depend on the strength of selection and the extent to which the character is inherited. Many characters are correlated, so a change in body weight may lead to a change in other characters, such as organ weights, life span, breeding performance, and tumor incidence. Natural selection may also alter some characters. If there is an infection in the colony, animals that are most resistant will tend to produce more offspring, increasing the frequency of “resistance” genes. As another example, if husbandry or environment changes, then animals that thrive under the new conditions will tend to leave more offspring. Thus, in order to prevent genetic change, the colony should be maintained as far as possible without deliberate selection of animals with a particular phenotype such as large body weight or good breeding performance. Any change in environment may also alter the colony because animals best adapted to the new conditions will tend to leave more offspring. Breeding stock should be selected strictly at random, without taking into account any of their phenotypic characteristics (obviously, abnormal animals would usually be excluded). If major deleterious (and unwanted) genes, such as blood clotting factors in dogs, are present in the colony, then steps may be taken to eliminate both affected animals and carriers using progeny testing methods described in many genetics textbooks (Falconer 1981; Hutt 1979). When single loci have been cloned and sequenced, it will often be possible to develop a polymerase chain reaction (PCR)based method for identifying carriers (Venta et al. 2000), thus making it easier to eliminate them from the breeding stock. Random Drift. Genetic segregation generates new combinations of genes in an essentially random manner, and in a large outbred stock colony, the frequency of an allele at any given genetic locus should remain constant according to the Hardy–Weinberg law (Falconer 1981). However, in smaller populations, gene frequency can change simply as a result of chance selection of breeding animals with certain alleles at any given genetic locus. Once an allele becomes fixed in the colony, with all animals genetically identical at that locus, it will no longer be able to change. The rate of change due to random genetic drift depends on the level of inbreeding, which in turn depends on the size of the breeding colony and on the amount of genetic variation already present. The coefficient of inbreeding (F) is the probability that the two alleles at a locus are identical by descent (i.e., that they are copies of the same gene at some previous period in the animal’s ancestry). F ranges from zero in a colony with no inbreeding to 1.0 or 100% in a fully inbred one. The inbreeding per generation, assuming random mating, is given by the formula
∆F = 1/8Nm + 1/8Nf
where ∆F is the increase in inbreeding per generation and Nm and Nf are the numbers of males and females present in the breeding colony that actually or potentially can leave offspring in the next generation.
84550.indb 215
11/11/10 6:25:25 PM
216
Handbook of Laboratory Animal Science
Note that the rate of inbreeding depends largely on the number of animals of the less numerous sex. For example, if the colony has four breeding males, then the inbreeding with random mating will be 1/32 = 3.1% just from the male side, no matter how many females there are. This is important when considering colonies of larger animals such as dogs, cats, and some species of primates, where a few stud males can be used with large numbers of females. In such circumstances, in order to reduce the rate of inbreeding, it is often necessary to utilize more males than would otherwise be needed. Inbreeding only reduces existing heterozygosity each generation. Over a period of time, the total inbreeding in the colony will be
∆Ft = ∆F + (1 – ∆F)Ft–1
where ∆Ft is the inbreeding at generation t and ∆F is the inbreeding for the current generation. If there is a genetic “bottleneck” with a reduced number of breeding animals in one or a few generations, this will lead to an increase in inbreeding that cannot be “undone,” even if the colony is subsequently expanded. If the size of the colony varies in each generation, the rate of inbreeding will also be affected. Formulae taking this into account are given by Falconer (1981). Bottlenecks are common when a small number of breeding animals are used to start a new breeding colony, or when a colony is “cleaned up” or “rederived” by hysterectomy or embryo transfer following an outbreak of disease. Care must be taken to ensure that enough breeding animals are used. As a general rule, if the aim is to maintain the colony for long periods, it is recommended that inbreeding levels should not be greater than about 1% per generation. Table€9.2 shows the inbreeding per generation from colonies maintained with various numbers of breeding individuals, and Figure€9.1 shows the coefficient of inbreeding over a period of 30 generations in colonies of various sizes. Inbreeding over a period of time, such as 5 years, can be reduced by having a long interval between generations. With mice and rats, it may be possible to produce only two generations per year by saving breeding stock from older females and avoiding breeding from first litters. With larger animals such as dogs and cats, there is even more scope for reducing inbreeding over time by increasing the intergeneration interval. A “maximum avoidance of inbreeding” system that approximately reduces the rate of inbreeding by half can also be used with small colonies. Using random mating, on average, each breeding pair will contribute one breeding male and one breeding female to the next generation. However, some Table€9.2â•…Inbreeding Coefficient (F%) in an Outbred Stock with Various Numbers of Breeding Males and Females No. of Males 4 4 4 8 8 16 16 32 32 100 100
84550.indb 216
No. of Females 4 8 24 8 16 16 32 32 64 100 200
Inbreeding per Generation
Inbreeding after 10 Generations
6.25 4.69 3.65 3.13 2.34 1.56 1.17 0.78 0.59 0.25 0.19
47.5 38.1 31.0 27.2 21.1 14.6 11.1 7.5 5.7 2.5 1.9
11/11/10 6:25:26 PM
Laboratory Animal Genetics and Genetic Quality Control
Coefficient of Inbreeding, F
1.0
217
1 Pair
0.8 5 Pairs
0.6
10 Pairs
0.4
20 Pairs
0.2 0.0
40 Pairs 80 Pairs 0
5
10
15
20
25
30
Generation Figure 9.1â•…Coefficient of inbreeding over 30 generations, with various numbers of breeding pairs.
animals will contribute more and some less due to sampling variation. If each breeding female and each breeding male can be made to contribute exactly one female or male, respectively, to the next generation, then this will decrease the rate of inbreeding by one-half relative to random mating. Various rotational breeding schemes aim to ensure that this happens. However, in practice, there is little difference between them (Nomura and Yonezawa 1996). All rely on ensuring equal representation of the current generation of breeding animals in the next generation. It may not be worthwhile to use a maximum avoidance of inbreeding system if the level of inbreeding is already well below the recommended 1% per generation. Halving something that is already very small may not be worthwhile. Brother × sister matings are also usually avoided, although in the long run, such matings do not increase the overall rate of inbreeding because, in the next generation, the offspring of brother × sister matings will be mated to unrelated animals and the inbreeding will be reversed. In summary, if the aim is to prevent genetic change, outbred stocks should be maintained as large, randomly mated colonies with less than 1% inbreeding per generation. Breeding individuals should be selected at random and without consideration of phenotype to avoid directional selection. When it is not possible to maintain sufficiently large colonies, maximum avoidance of inbreeding should be used by ensuring that each breeding female contributes as close to one female and each breeding male contributes as close to one male as possible to the next generation of breeding stock. Genetically Heterogeneous Stocks, Segregating Hybrids, and Advanced Intercross Lines Genetically heterogeneous (GH) stocks have been synthesized by crossing several inbred strains and maintaining the offspring as an outbred stock. Such stocks have been used as a base population for selection experiments and in a range of behavioral studies (McClearn, Wilson, and Meredith 1970; McClearn and Hofer 1999; Heller et al. 1998). They have also been used for mapping QTLs. Classically, this has been done using crosses between inbred strains, but many genes on the same chromosome remain linked; there have been too few generations for crossing over to have broken up the linkage groups. However, in an outbred stock, over a period of many generations, there will have been considerable crossing over among loci on the same chromosome, thereby substantially reducing linkage disequilibrium. The fragments of chromosome containing the QTLs and marker
84550.indb 217
11/11/10 6:25:27 PM
218
Handbook of Laboratory Animal Science
genes will be much shorter. This means that each QTL can be mapped more accurately than is possible using F2 hybrids (Mott et al. 2000; Valdar et al. 2006). Segregating hybrids or advanced intercross lines (AILs) (Darvasi and Soller 1995) are outbred stocks produced by crossing two or a few inbred strains, followed by random or semirandom mating for several generations. These stocks should be maintained in the same way as outbred stocks, with large population sizes and maximum avoidance of inbreeding procedures, when necessary, in order to minimize the loss of heterozygosity. Outbred Selected Stocks Genetic selection with the maintenance of genetic heterogeneity has been used by several investigators as a means of producing new animal models. Examples include the Biozzi mice selected for high and low immunological response to sheep red blood cells (Feingold et al. 1976; Liu et al. 1993), SENCAR mice selected for susceptibility to skin carcinogenesis (Updyke et al. 1989), the Dahl rats selectively bred for high or low blood pressure on a high-salt diet (Rapp 1982), and several stocks of mice selected for various aspects of response to alcohol (Collins and Marks 1992). This method of selection, with the avoidance of inbreeding, is more efficient and likely to result in a more substantial change in phenotype than the alternative method of selection with inbreeding. In some cases, the original selected stock continues to be used in research; in others, one or more inbred strains have been developed from the outbred selected stock. It is important for breeders to know whether a particular colony is of the original outbred stock—in which case it should be maintained in the same way as other outbred stocks (described previously)—or whether it is an inbred strain, in which case it should be maintained using methods described later. Partially Genetically Defined Mutants may occur within an outbred stock and, with larger animals, transgenic stocks will also usually be developed in the outbred stock. Similarly, in rodents, outcrossing in order to obtain better breeding performance is quite common in the development of genetically modified animals. In these situations, the breeder needs to consider how the stock is to be maintained in the future. With large animals or species where inbred strains are not available, there is little option but to maintain the mutation/transgene on the outbred background, taking into account the possible need to avoid genetic drift and inbreeding and any associated reductions in breeding performance. A choice is possible with rats and mice. The stock can be maintained as a segregating outbred stock, the colony could be inbred with heterozygosity being maintained, or, preferably, the mutation/ transgene could be backcrossed to an existing inbred strain. The expression of most mutants and transgenes is influenced by the genetic background and, if heterogeneous, the expression of the mutation will also be variable. Moreover, if the genetic background drifts or if outcrossing to another strain is used to boost breeding performance, then the expression of the mutation may vary over time. For this reason, most geneticists would recommend that the mutation/transgene be transferred to an inbred or possibly hybrid background by backcrossing, as described later in the development of a congenic strain (Silva et al. 1997). Once a congenic strain has been developed, its characteristics will be fixed and it can be maintained in small numbers without problems related to genetic drift. The Isogenic Family of Strains and Their Derivatives This family of strains includes inbred, segregating inbred, congenic, recombinant inbred, recombinant congenic, consomic, and F1 hybrid strains, as well as monozygous twins and genetic clones. These latter two classes are not discussed here. The most important feature of this class of strains is
84550.indb 218
11/11/10 6:25:27 PM
Laboratory Animal Genetics and Genetic Quality Control
219
that at least two individuals (usually, many more) are genetically identical or nearly identical. All of these strains, with the exception of twins and clones, are based on inbred strains as discussed next. Inbred Strains Inbred strains are produced by at least 20 generations of brother × sister mating or the genetic equivalent, with all individuals being traceable to a single breeding pair in the 20th or a subsequent generation in order to eliminate sublines. Parent × offspring mating is one alternative, provided the mating is always to the younger of the two parents (several generations of mating female offspring to the same male is not the same thing, genetically). The effect of this mating scheme is to increase homozygosity within the strain to more than 98% (Figure€ 9.1), though the approach to complete homozygosity is asymptotic and, in theory, a strain never becomes fully inbred. Inbreeding also increases the total genetic and phenotypic variance, but all this variation is seen as differences between the substrains with the within-substrain genetic variation reducing to near zero. Thus, if an outbred stock is inbred and all sublines are kept, the total phenotypic variation is substantially greater than in the original outbred colony. (The nomenclature of inbred strains is given in Appendix 9.1.) Properties of Inbred Strains Inbred strains have been described as “immortal clones of genetically identical individuals.” Thus, they have many useful properties that make them the animal of choice for many types of research. The main properties of these strains are discussed in the following subsections: Isogenicity. All animals within an inbred strain are virtually genetically identical. One consequence is that only a single individual needs to be genotyped at any locus in order to type the whole strain. Over a period of time, a catalogue or “genetic profile” of the alleles carried by each inbred strain can be generated. The DNA of several inbred strains of mice has now been fully sequenced (Frazer et al. 2007). This information can be used in planning and interpreting experiments and in mapping genes of interest. Isogenicity also implies that a single male and female taken from the colony should have all the alleles present in that colony. Hence, a daughter colony founded on a single breeding pair will, for most practical purposes, be genetically identical to the parent colony, at least until the colonies begin to diverge as a result of the accumulation of new mutations (see later discussion). Isogenic individuals will also be histocompatible, so skin, cell, and organ grafts exchanged between samesex members of the same strain should not be immunologically rejected. Homozygosity. Inbred strains are defined in terms of homozygosity. By the end of 20 generations of full sib mating, the chance that any two alleles at a given locus are identical by descent (i.e., are copies of the same allele in a previous generation) is more than 98%. The most important practical consequence of this is that there should be no genetic segregation within the strain; all genes will be expressed under appropriate conditions and there will be no hidden recessive genes to cause confusion in breeding experiments. Homozygosity ensures that parents and offspring are genetically identical; this is what makes the strain immortal. Phenotypic uniformity. Because there is no genetic variation within an inbred strain, the phenotype for highly inherited characters tends to be more uniform. The only variation between individuals will be due to nongenetic causes. One consequence is that, other things being equal, fewer inbred animals will be needed in an experiment to achieve a given level of statistical precision than if outbred animals had been used. In some cases, the reduction can be substantial. For example, in one study (Jay 1955), the standard deviation of sleeping time under barbiturate anesthetic, averaged across 185 mice of five inbred strains, was 3.2 minutes. However, it was 13.5 minutes averaged across 91 mice of two outbred stocks. If an experiment were to be set up to see whether some
84550.indb 219
11/11/10 6:25:27 PM
220
Handbook of Laboratory Animal Science
treatment altered sleeping time by 4 minutes or more, assuming a power of 90% and a significance level of 0.05, this could be done with 14 inbred mice per group, but it would require more than 10 times that number of outbred mice to achieve the same objective. While this may be an extreme example, it emphasizes the great importance of controlling the genetic variation if the aim is to minimize the numbers of animals used in research. Long-term stability. In an outbred stock, change in allele frequency and therefore in phenotype can be caused by directional selection, genetic drift due to inbreeding, and new mutations, assuming that genetic contamination is avoided. An inbred strain is already fully inbred, so further inbreeding will have no effect. Because there is no genetic variation within the strain, directional selection will be ineffective in changing the genotype and phenotype. Thus, the phenotype of an inbred strain will only change as a result of the fixation of new mutations or as a result of environmental changes (which can often be of great importance). New mutations are relatively rare, and only a quarter of them will normally be fixed with continued full sib mating. Therefore, inbred strains tend to stay genetically constant for long periods of time. However, some sublines of most of the more widely used strains have been separated for many generations, providing substantial time for mutations to have occurred and become fixed. Many of these mutations will be “quiet,” showing no obvious phenotype except in unusual circumstances (Stevens et al. 2007). It is important for investigators to state clearly the origin of the animals they use with correct, internationally recognized nomenclature. Even the small amount of genetic drift due to new mutations can be eliminated by preserving frozen embryos (Whittingham et al. 1972). In a few cases, directional selection may apparently alter strain characteristics, but this may be the result of associated microflora rather than through the genetics of the inbred strain (or the strain may have become genetically contaminated). The increased obesity sometimes found in inbred strains over a period of time is almost certainly due to changes in husbandry rather than to genetic change. The long-term stability of inbred strains has important consequences. If the environment is kept constant, phenotypic profiles of strain characteristics can be developed. These can include life span, types of spontaneous disease, immune functions, susceptibility to microorganisms, and biochemical and physiological characteristics. This information can be built up by many research workers using genetically identical animals over a long period of time. Individuality. Each strain is genetically unique and has unique phenotypic characteristics. Sometimes these characteristics are useful in research; at other times, they may preclude the use of a particular strain for a given research project. Many strains of mice were developed for use in cancer research and have high levels of a particular type of tumor (http://tumor.informatics.jax. org). For example, strain AKR gets leukemia, SJL gets reticulum cell sarcoma, and C3H gets mammary tumors (provided individuals carry the mammary tumor virus). However, other strains, such as C57BL/6, have relatively low levels of cancer and tend to be relatively resistant to carcinogens. Similarly, the F344 rat strain gets a high incidence of testicular tumors. Differences in virtually any phenotypic characteristic are likely to be found among a group of independent strains. Many aspects of behavior, such as open-field activity, wheel running, aggression, and learning ability differ between strains, and responses to bacteria, viruses, and toxic chemicals often vary dramatically between different strains. These and many more strain characteristics are listed in the mouse phenome database (http://phenome.jax.org). These strain differences have enormous importance because strains can be chosen that have characteristics that make them useful for any given type of research. In extreme cases, whole areas of research have been opened up based on the characteristics of a single mouse strain. For example, monoclonal antibody technology is all based on the plasmacytoma tumors induced in BALB/c mice by intraperitoneal injection of mineral oil (Roberts et al. 2002; Potter 1972). The technology for gene targeting to produce knockout mice by homologous recombination (in which a single defined
84550.indb 220
11/11/10 6:25:28 PM
Laboratory Animal Genetics and Genetic Quality Control
221
gene is inactivated) is based on the establishment of ES cell lines from strain 129 mice (Evans and Kaufman 1981). The starting point for an investigation using the mouse or rat should usually be a small study comparing different inbred strains. Identifiability. The isogenic property of inbred strains makes it possible to build up a genetic profile of each strain. This can then be used for genetic quality control purposes. Thus, if some white rats are supposed to be strain F344, this can be tested from a small sample of DNA using one of the methods discussed below (see “Genetic Quality Control” section). However, if the rats are supposed to be outbred Wistars, no known method of genetic quality control will distinguish Wistars from Sprague–Dawleys, or Wistars from different colonies. Sensitivity. As a broad generalization, inbred strains tend to be more sensitive to environmental influences than outbred stocks or F1 hybrids (see later discussion). This is disadvantageous because it means that extra care is needed to ensure that inbred strains have the optimum environment to minimize variability. However, it is advantageous because inbred strains will also tend to be more sensitive to experimental treatments than other classes of stock. International distribution. The isogenic property of inbred strains means that long-term, stable daughter colonies that are very similar to the parent colony can be set up using only a single breeding pair. Thus, work can be repeated using similar animals throughout the world. In contrast, although outbred stocks, such as Wistar rats, are available all over the world, genetic drift and directional selection mean that each colony is, to some unknown extent, genetically different. Research Uses of Inbred Strains Inbred strains are or should be the animal of choice for general research using mice or rats. They are the nearest thing to a pure reagent that is possible when using laboratory animals. As noted previously, a strain represents a single genotype that can be repeated by the thousands. The strain remains genetically constant for long periods; considerable background information on the genetic and phenotypic characteristics of the more common strains is available. The phenotype tends to be uniform, and genetic quality control is relatively easy using a wide range of DNA genetic markers. If the project requires the screening of animals with a wide range of potential genetic susceptibility to a toxic or pharmaceutical agent or microorganism, then small numbers of animals of several strains can be used. This is much more effective than using an outbred stock, which may be phenotypically quite uniform yet does not necessarily have a wide range of susceptibility genotypes. There are compelling arguments for using this approach (Festing 1990, 1995, 1999; Kacew, Ruben, and McConnell 1995). Inbred strains are also of increasing importance as the “genetic background” for mutants and genetic modifications. This is discussed in more detail in the “Congenic Strains” section. By 2000, at least 17 Nobel Prizes had been awarded for work that either required inbred strains or would have been much more difficult without them (Festing and Fisher 2000); by 2008, a total of 22 such prizes had been awarded. There are well over 400 “straight” inbred strains of mice and about 200 inbred strains of rats, excluding the congenic, recombinant, etc. strains discussed later. Each of the congenic, recombinant, etc. strains are inbred strains in their own right. Details concerning the origin and some phenotypic characteristics of these strains are available on the Jackson Laboratory Web site and in the rat genome database (see Appendix 9.2 for Web resources). Extensive genealogies of inbred mouse strains have also been developed (Beck et al. 2000), and dendrograms have been published showing the genetic similarities among inbred strains of mice (Petkov et al. 2004) and rats (Thomas et al. 2003). However, approximately 80% of research using inbred strains is done using only about 10–15 of the most common strains (www.isogenic.info). The estimated “top 10” most widely used strains of mice and rats are listed in Tables€9.3a and 9.3b.
84550.indb 221
11/11/10 6:25:28 PM
222
Handbook of Laboratory Animal Science
Table€9.3a╅ Inbred Strains of Mice and Frequency of Citations in the PubMed Bibliographic Database MeSH Heading Yes Yesa Yesa Yes Yes Yes Yes Yes Yesa Yesa No No Yes Yes Yes No Yes No
Mice, Inbred Strains BALB/c C57BL/6 C57BL/10 C3H ICR CBA A/J NOD DBA/1 DBA/2 MOLD 129 HRS MRL NZB NIH AKR SJL
No. of Times Cited 20,421 10,279 6,251 3,764 3,502 2,167 1,771 1,722 1,199 1,178 877 619 479 411 294 291 290 200
Percent of Inbreds 36.4 18.3 11.1 6.7 6.2 3.9 3.2 3.1 2.1 2.1 1.6 1.1 0.9 0.7 0.5 0.5 0.5 0.4
Notes: Based on a PubMed search of the number of papers using each strain for 2001–2005. Strain BALB/c is clearly the most popular strain, followed by the two major substrains of C57BL. Together, these accounted for approximately two-thirds of all times that a strain was used. a PubMed does not distinguish between sublines.
There are two main reasons for choosing a strain, such as C57BL/6 mice or F344 rats, for a particular research project: • The strain may be chosen because it is regarded as a good general-purpose strain that can be used to replace the use of an outbred stock, such as Swiss mice or Wistar rats. In this case, a strain would be chosen that has no known characteristics that would preclude its use for the project. Strains with a high incidence of a specific disease and those that were highly aggressive or had some known immunological defects, for example, might not be suitable for the project. When a suitable strain is chosen for a new project, several available strains should be screened to identify the ones that show an appropriate response to the treatment of interest. The project could then be continued with that strain, with occasional studies of other strains to confirm that the results were not highly strain specific. • A specific strain may be chosen because it has characteristics that make it useful for a particular type of research. For example, C57BL/6 mice maintained on a high-fat diet develop atherosclerosis, making them a useful model of that condition in humans (Li et al. 2008; Paigen et al. 1990). C57BL/6 mice also like sweet tastes and alcohol and are highly active in an open field. A searchable database of these sorts of phenotypic characteristics of mouse and rat strains is available at the mouse phenome database (see Appendix 9.2).
Inbred strains are sometimes used in studying the inheritance of complex or polygenic characters (i.e., those with a complex mode of inheritance depending upon several genes with nongenetic or environmental influences). To perform these analyses, strains that differ for the character of interest are crossed and then intercrossed to produce F2 hybrids. A search for any association between genetic markers that differ between the parental strains and the expression of the character of interest is then performed. For example, strain A/J is susceptible to the induction of lung tumors by carcinogens and C57BL/6 mice are resistant. In the F2 hybrids, an association between a marker
84550.indb 222
11/11/10 6:25:28 PM
Laboratory Animal Genetics and Genetic Quality Control
223
Table€9.3b╅ Inbred Strains of Rats Strain F344 LEW SHR WKY BN Dahl SS and SR WF ACI BB OLETF WAG BUF PVG LEC OM COP BDIX BDII BH Total inbred strains
No. of Citations 4,611 3,227 2,086 1,821 989 318 309 224 209 171 95 90 79 66 16 15 12 8 7
MESH Heading
Percent of All Rats
Percent of Inbred Rats
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes No Yes No No No No No
3.8 2.7 1.7 1.5 0.8 0.3 0.3 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0
32.1 22.5 14.5 12.7 6.9 2.2 2.2 1.6 1.5 1.2 0.7 0.6 0.6 0.5 0.1 0.1 0.1 0.1 0.0
11.9
100.0
14,353
Notes: This is based on a PubMed search of the number of papers using the designated strain or stock for 2001–2005. Note that inbred strains were used on only 11.9% of occasions, with the vast majority of work being done using outbred Sprague–Dawley or Wistar rats. Only 12 inbred rat strains have a PubMed MeSH heading.
(Kras2) on chromosome 6 and susceptibility to lung tumors has been found, implying that a susceptibility gene is linked to (or identical to) Kras2 (Festing, Yang, and Malkinson 1994). Breeding and Maintenance of Inbred Strains The usual aim in maintaining a colony of laboratory animals is to prevent genetic change so that information on the characteristics of the strain can be accumulated and used in the planning and interpretation of future experiments. Once they are fully inbred, the only way that inbred strains can change (assuming no genetic contamination due to mating with another strain) is as a result of the accumulation of new mutations. A breeding program should be designed to minimize the chance that a mutation will become fixed in the colony. It should also be economical and flexible in response to changes in demand. Unlike outbred stocks, inbred strains can be maintained with very small numbers of breeding animals. Inbred mice and rats are usually maintained as permanently mated monogamous pairs, and the minimum-sized breeding colony is one that just prevents the colony from dying out, given that breeding performance can be somewhat uncertain for some strains. When large numbers of animals are needed for research purposes, an appropriate breeding scheme is to maintain a “stem line” or “foundation” colony of between 10 and 30 breeding pairs, with a multiplication colony of sufficient size to provide all the required experimental animals as a result of about four to six generations of breeding. The multiplication colony is used only to produce experimental animals and does not contribute to the long-term survival of the strain. This breeding scheme is shown diagrammatically in Figure€9.2.
84550.indb 223
11/11/10 6:25:28 PM
224
Handbook of Laboratory Animal Science
The traffic light system for breeding inbred strains White label FS
Green label
FS Yellow label FS
FS
Red label
Figure 9.2â•…The “traffic light” system for controlling the foundation stock (FS) and multiplication colonies of an inbred strain. The FS colony is self-perpetuating, and any surplus stock can be transferred to the multiplication colony, in breeding cages with a white label. Their offspring can be used for a further three generations of multiplication, but the offspring of the red label colony are not used for breeding.
The stem line colony should be maintained by brother × sister mating (usually as pairs) and should be kept as physically separated from all other animals of the same species as possible. Some breeders maintain such colonies in isolators. A bank of frozen embryos can be maintained (if possible) as a backup to prevent loss of the strain due to poor breeding or disease. Genetic quality control methods (see later discussion) should be used to authenticate the strain and to monitor it over a period of time. Detailed records on each breeding pair should be maintained, as well as a pedigree chart showing the relationships between the pairs. These can be paper- or computer-based records. Generally, all breeding pairs should be descended from a single breeding pair approximately five to seven generations back. Although directional selection within an inbred strain should have no effect, it is advisable to select for good breeding performance to try to prevent diminished breeding as a result of the accumulation of new deleterious mutations. Often, such mutations will have a very small effect and the best method for selection is based on the average performance of the various sublines in the colony. For example, an index of productivity of each breeding pair, such as the number of young weaned per pair per week, can be recorded on the pedigree chart. Replacement breeding stock would then be chosen according to the average productivity of the different substrains, although a substantial rise in productivity could indicate genetic contamination with consequent hybrid vigor. A simplified example is given in Figure€9.3. The multiplication colony should also be as physically separated from other colonies of the same species as possible to prevent genetic contamination via a nonstrain mating. If the aim of the colony is to identify any new mutations that may occur, sib matings can be performed to reveal any recessive mutations. Alternatively, if the aim is simply to produce large numbers of experimental animals, the colony can be randomly mated and trios or other more economical mating systems can be used. Random mating for a few generations should have no adverse impact on the genetic quality of the animals. A few new mutations may accumulate, but in most cases, these are unlikely to have any impact on research. However, as noted before, a maximum of about four to six generations of such mating should be used. A practical scheme for the multiplication colony is the use of the “traffic light” scheme (Lane-Petter and Pearson 1972) in which surplus offspring of the stem line go into breeding cages with a white label and their offspring go into cages with a green label; their offspring go into cages
84550.indb 224
11/11/10 6:25:29 PM
Laboratory Animal Genetics and Genetic Quality Control
225
Choice of subline in the foundation stock (FS) colony 0.8
0.6
0.9
1.0
0.3
0.8
0.5
0.2
0.8 0.9 0.2 0.0
0.5 0.5 0.7 0.8 0.8
Mean subline 1 = (0.8+0.6+0.5+0.9+....+0.2+0.0)/8 = 0.59 Mean subline 2 = (1.0+0.8+0.2+0.3+.......+0.8+0.8)/10 = 0.56 Conclusion: no significant difference, but probably best to select line 1 Figure 9.3â•…Preprinted pedigree chart showing the output of individual breeding pairs (young weaned/female/ week) and a scheme for selecting against any decline in breeding performance by comparing the average performance of each subline.
with an amber label and their offspring go into breeding cages with a red label. No offspring from red-labeled cages are used for breeding. Four generations of multiplication means that even a poorly breeding strain can produce large numbers of experimental animals, but if required, several more generations can be bred. F1 Hybrids These are the first-generation cross between two inbred strains and being isogenic; they have many of the more useful properties of the parental inbred strains. F1 hybrids usually exhibit hybrid vigor; that is, they are both more vigorous than inbred strains and less influenced by adverse environmental conditions. As a result, they tend to be slightly more uniform than inbred strains (Klempt et al. 2006; Festing 1976). However, they may also be less responsive to experimental treatments, so the increased uniformity does not necessarily mean that fewer animals are needed to attain a given level of statistical precision in an experiment. F1 hybrids are heterozygous at all of the loci at which the parental strains differ and will therefore not breed true. Mating two F1 hybrid animals together yields an F2 hybrid—a genetically segregating generation. F2 hybrids are widely used for genetic mapping studies, but are of less value as general research animals (see Appendix 9.1 for designations of F1 hybrids). Research Uses of F1 Hybrids F1 hybrids have one great advantage over pure inbred strains: They are much more robust. F1 hybrids tend to live longer, have fewer of the idiosyncrasies of the parental strains, and are less sensitive to adverse environmental conditions than inbred strains. They are particularly valuable as foster mothers in the production of transgenic strains. However, they should be used with caution in experiments involving breeding because their offspring will be F2 hybrids, which are genetically segregating for the gene loci that differ in the parental strains. F1 hybrids also sometimes possess useful characteristics not normally found in the parental strains. For example, the NZBNZWF1 is widely studied as a model of autoimmune systemic lupus erythematosus, which is not found in the parental strains (Dubois et al. 1966; Tao et al. 2008).
84550.indb 225
11/11/10 6:25:30 PM
226
Handbook of Laboratory Animal Science
F1 hybrids can be of some value in immunological studies because they will accept tissue, organ, tumor, or cell grafts from either of the parental strains without immunological rejection. One possible disadvantage of F1 hybrids is that they tend to be intermediate in phenotype, somewhere between the two parental strains. Thus, if an experiment requires several animals differing in their phenotypes, it may be better to use several inbred strains rather than several F1 hybrids. Coisogenic and Congenic Strains Definition and Development A pair of strains is said to be coisogenic if they differ at only a single genetic locus as a result of a mutation within one branch of the strain. Such strains are useful because the effects of the mutation can be studied without the complication of genetic segregation in the genetic background. Unfortunately, coisogenic strains cannot be produced to order because they depend on an uncontrollable mutation occurring within the inbred strain. However, targeted mutations, produced by homologous recombination using ES cells usually having the 129 inbred strain genotype, are coisogenic with strain 129 unless they are outcrossed to another strain. ES cells from strain C57BL/6 and other more widely used strains are now becoming available. Thus, someone who has produced a targeted mutant (knockout) mouse should give careful consideration to choice of genetic background. If the knockout was made using 129 ES cells, then backcrossing to another strain should be considered because strain 129 breeds poorly and is not well characterized. A pair of strains is said to be congenic if they approximate the coisogenic state as a result of backcrossing a gene (or, more strictly, an allele at a particular locus) (known as the differential allele) to an inbred strain. Several methods of backcrossing can be used, depending on the mode of inheritance and method of identification of the gene. Basically, a donor strain is mated with the chosen background strain (often C57BL/6, but any strain may also be used depending on the aim of the study) to produce an F1 hybrid, designated N1. These animals are then mated to the inbred strain to produce the first backcross generation, designated N2. These animals do not all carry the gene of interest, so carriers of the gene must be identified. Many transgenes can be identified from a sample of DNA; in this case, animals that carry the transgene or mutation would be selected for further backcrossing. If the gene can only be identified by its phenotypic effect and if it is recessive, then it will be necessary to breed some of the N2 generation together in order to produce some homozygous mutant animals to continue with the backcrossing program (the N is not incremented in this case because it does not count as a backcross generation). These procedures are repeated—ideally until at least the N10 generation has been reached. At this stage, the strain carrying the mutation is said to be congenic with the background strain, and if the differential allele can be made homozygous, it can be maintained just like an ordinary inbred strain (as discussed previously) without further backcrossing. Unfortunately, backcrossing procedures take a long time. It is difficult to produce even four generations of mice per year, so developing a congenic strain can often take as long as 3 years. Modern research requires rapid results, tempting people not to go through the whole process before using the animals. However, after five generations of backcrossing (N5), the coefficient of inbreeding will be 97%—not much less than that of a newly inbred strain after 20 generations of brother × sister mating. Therefore, semicongenic strains are often used at this stage, although it is advisable to continue backcrossing when possible. Marker-assisted or “speed congenics” (Markel et al. 1997) can be used from the offspring of the N2 generation onward. Approximately 20 males that have been tested and are known to carry the gene of interest are genotyped at about 80 or more microsatellite or single-nucleotide
84550.indb 226
11/11/10 6:25:30 PM
Laboratory Animal Genetics and Genetic Quality Control
227
polymorphism (SNP) genetic markers. The male that has the most alleles of the background strain summed across these loci is then chosen for further backcrossing. He will be mated to several females in order to produce about 80 offspring. On average, half of these will be females (not used) and half of the males will not carry the desired gene; this leaves about 20 offspring to be tested for the next generation, from which the best male is again selected. Such a breeding program will decrease by one-half the amount of time that it takes to produce a congenic strain, but also is expensive and needs careful organization. Nomenclature of coisogenic and congenic strains is given in Appendix 9.1. Research Uses of Congenic Strains Congenic strains have been used extensively by immunogeneticists to isolate the genes responsible for the rejection of allografts and subsequently to study the genetics of the major histocompatibility complex in the absence of the additional complexity and “noise” created by segregation of the background (Klein 1975). Many sets of congenic strains that differ at the MHC are available, and these have been used, for example, to study the effects of this complex locus on response to microorganisms (the Leishmania parasite) (Bradley 1982) and carcinogens (urethane) (Miyashita and Moriwaki 1987). Such strains can be used without the necessity of genotyping individual animals. Many spontaneous and induced mutations, knockouts, and transgenes have been backcrossed to an inbred strain. In many cases, a C57BL/6J genetic background has been chosen. This has the advantage that not only can a mutation be compared with wild-type (i.e., the nonmutant type), but it can also be compared with other mutations on the same background. However, the expression of a gene may depend on the genetic background, so a mutation is sometimes backcrossed to several inbred strains to study the range of expression due to genetic modifiers. In some cases, genetic modifiers have been mapped (Cormier et al. 2000). Controlling such modifiers may provide a therapeutic approach to controlling the expression of some mutants in animals and, eventually, in human medicine (Nadeau 2001). Maintenance of Congenic Strains Once a strain is fully congenic, it can be maintained in the same way as any other inbred strain, although, if the genetic alteration is deleterious, then the strain may need to be maintained with forced segregation at the altered locus. For example, female nude mice do not breed well, so it is common to maintain a strain congenic for the nude mutation (Foxn1nu) by mating homozygous males to heterozygous females. Recombinant Inbred (RI) Strains Definition and Development Sets of recombinant inbred (RI) strains are produced by crossing two standard inbred strains (such as C57BL/6 and DBA/2) to produce an F1 hybrid, followed by brother × sister matings of the F1s for at least 20 generations while maintaining several separate lines. Ideally, at least 20 new lines are developed, though smaller sets can still be useful. In these strains, the genes for which the parental strains differ have been recombined in each of the resulting RI strains. At any one genetic locus at which the parental strains differ (such as the major histocompatibility complex), about half the strains should resemble each of the parental strains. Several sets of RI strains developed from
84550.indb 227
11/11/10 6:25:31 PM
228
Handbook of Laboratory Animal Science
different pairs of parental strains (Taylor 1996) are available and listed on the JAX Web site. The nomenclature of RI strains is given in Appendix 9.1. Research Uses of RI Strains Sets of RI strains are useful for investigating characters that differ between the two parental strains. If the parental strains differ for some phenotype, such as the activity of a particular enzyme or susceptibility to a pharmaceutical or toxic agent, it is often of interest to find out whether the difference is due to a single genetic locus and, if so, the chromosomal location of the gene. This can be done very efficiently and quickly using RI strains without having to do any cross breeding, assuming that a set of RI strains is already available. The procedure for examining these two questions is to screen the parental strains of several sets of RI strains to see if they differ for the character of interest. When pairs of strains are found to be different, small numbers of animals of each of the corresponding set of RI strains are phenotyped, and the distribution of strain means is studied. If the means fall into two distinct groups, each of which resembles one of the parental strains, then that is evidence that the phenotype is controlled by a single genetic locus (the “new gene”). In contrast, if there is a continuum of phenotypes covering many intermediate values, then this would suggest that the character has a more complex mode of inheritance. If the RI strains fall into two groups, then the pattern of similarities to other gene loci already mapped may indicate the chromosome location of the new gene. As an example, strains C57BL/6 and DBA/2 differ in the activity of the Mod1 enzyme. The BXD set of 26 RI strains was developed from a cross between these two strains. Suppose all 26 strains are typed and that, in about half of them, the enzyme activity resembles strain C57BL/6 and in others it resembles DBA/2; all strains can unambiguously be classified into these two groups, with no intermediate strains. This suggests that a single gene determines enzyme activity. The resemblance of each RI strain to one of the parental strains can be indicated by a B (like C57BL/6) or a D (like DBA/2), and a pattern of response such as DBBDD BBDBB, etc. can be built up. This is called the “strain distribution pattern” (or SDP) of the new gene. Two genes that are tightly linked on the same chromosome will tend to have similar or even identical SDPs because the loci will rarely have recombined during inbreeding. In contrast, the SDPs of two unlinked genes will not be identical. It turns out that the activity of the Mod1 gene and the polymorphic D9Mit10 (a microsatellite locus) have identical SDPs of DBBDD BBDBB BDDBB DDBBB BBBDB B, so the two loci are closely linked. D9Mit10 has been mapped to a particular location on mouse chromosome 9. Thus, Mod1 must be close to D9Mit10 on chromosome 9 (in fact, Mod1 was mapped before D9Mit10, but this example illustrates the method). Most of the large sets of RI strains have been typed at many loci of known chromosomal map location. If a single gene is indicated as a result of phenotyping a set of RI strains, then there is a high chance that it can be mapped simply by looking for matches in the SDPs, as explained before. Sets of RI strains can also sometimes be used to map QTLs involved in the inheritance of polygenic characters. For example, there is a significant association between the mean number of lung tumors and genotype at the Kras2 locus in the AXB and BXA set of RI strains (Malkinson 1991), although the strains do not fall into two distinct groups. There is a similar association with the H2 complex, so, in this case, two quantitative trait loci have been associated with particular chromosomes. However, few sets of RI strains are sufficiently numerous to allow the resolution of more than two loci. Maintenance of RI Strains Each RI strain is an inbred strain is its own right and should be maintained as an inbred strain using methods described previously.
84550.indb 228
11/11/10 6:25:31 PM
Laboratory Animal Genetics and Genetic Quality Control
229
Recombinant Congenic (RC) Strains Definition and Development Sets of RC strains are developed in a very similar way to the development of RI strains, except that instead of inbreeding from the F2 hybrids, a few (generally about three) generations of backcrossing to one of the parental strains is done first, followed by sib mating of several independent lines for the equivalent of 20 generations (Demant and Hart 1986). Each backcross generation is regarded as being equivalent to two generations of sib mating. If the F1 hybrid is designated generation N1, then backcrossing to N3 is typically used so that each strain has, on average, a 12.5% contribution from the donor strain and 87.5% contribution from the background strain. The set is regarded as fully inbred at N3F14. Other levels of backcrossing may be used and will be indicated in the description of each RC set. The nomenclature of RC strains is given in Appendix 9.1. Research Uses of RC Strains Many characters of biomedical interest have a complex “polygenic” mode of inheritance, depending on several genes (each possibly on a different chromosome) and on environmental influences. The genes are known as QTLs. Type I diabetes is such a character (Cordell and Todd 1995). Susceptibility is inherited, but no single gene is implicated and there are environmental influences. Any pair of inbred strains will normally differ at several quantitative trait loci for each character of interest (diabetes, body weight, cancer susceptibility, etc.). The strategy is to try to separate these out so that individual loci can be identified and studied on a common inbred background. RC strains have been developed for this purpose. Assume, for example, that a set of RC strains is already available, and that the character of interest is the response to a toxic agent. The donor and recipient strains will be phenotyped. If they differ in response, then the whole set of RC strains (possibly 10–20 of them) is phenotyped (i.e., their response is tested) with several animals of each RC strain being tested. The aim is to find one or more strains that are intermediate between the donor and recipient strain in the hope that it or they differ from the recipient strain by just a few genes or, ideally, by only a single gene. When such strains are found, then crosses can be made between them and the recipient strain to find out if they differ at only a single locus. If so, then that QTL will have been isolated on a genetic background that only differs from the recipient strain at about one-eighth of all loci. With further breeding and backcrossing, it should then be possible to isolate and identify the individual QTL. All this may seem a bit complicated; however, it is not easy to identify QTLs, and this provides an approach to doing so. RC strains have been used in a wide range of studies, including analyses of susceptibility to cancer (Tripodis and Demant 2001), pathogens (Fortin et al. 2001), and bone strength (Yershov et al. 2001). However, until now, their use has been limited to a few laboratories that have made the considerable investment to develop and genotype some useful sets. The main disadvantage of using this approach is that RC strains usually can only be used for studying characters that differ between the two parental strains. Thus, if the phenotype of interest is not seen in either of the parental strains, it is unlikely (though not impossible) that a particular set of RC strains will be of use in investigating that character. Maintenance of RC Strains Once a set of RC strains is fully inbred, each strain is an inbred strain in its own right and should be maintained in the same way as any other inbred strain.
84550.indb 229
11/11/10 6:25:31 PM
230
Handbook of Laboratory Animal Science
Consomic or “Chromosome Substitution” Strains Definition, Development, and Research Uses Consomic strains are developed by backcrossing (introducing) a whole chromosome from one donor inbred strain into a recipient “background” strain. They are used for studying the genetics of complex traits and in the eventual identification of QTLs (Nadeau et al. 2000). A minimum of 10 backcrosses is used; ideally, all the background strain chromosomes (except the one being introduced) should be present and uncontaminated with donor chromosomal fragments. The whole donor (introduced) chromosome should be present as well. Strains consomic for autosomal chromosomes are developed using about five or six polymorphic genetic markers on the chosen chromosome that are known to differ between the two strains. Offspring are then only chosen for backcrossing if they have all markers of the donor type for the chosen chromosome (i.e., if the chromosome has not recombined). A full set of consomic strains for the mouse would involve 21 strains, 19 of them representing the background strain with each of the 19 autosomes substituted and the other two representing the background strain with either the X or Y chromosome from the donor strain. However, even a few of these strains can be used to answer questions about whether or not loci on a particular chromosome influence a particular phenotype. For example, a number of consomic rat strains with chromosomes from BN strain rats substituted on an SS background have been developed. The background strain exhibits high blood pressure when placed on a high-salt diet. These strains are now being characterized for 203 heart, lung, vascular, and blood function phenotypes. For example, strain SS.BN-13, which has the BN chromosome 13 substituted, has lower mean arterial pressure than the SS strain, demonstrating that there is a salt-related blood pressure gene on this chromosome (Cowley et al. 2001). Consomic rat strains have also been used to map genes for susceptibility to mammary cancer following treatment with a carcinogen (Adamovic et al. 2008). In mice, these strains have been used to study susceptibility to germ-cell tumors in 129 strain mice, using a chromosome 19 substitution strain (Martin et al. 1999). Sets of consomic strains have now been developed with chromosomes from Mus musculus musculus placed on a C57BL/6 genetic background (mostly M. m. domesticus) (Gregorova et al. 2008), providing a wide range of genetic diversity among the strains. The development of Y chromosome consomic strains is relatively easy. A donor strain male is crossed with a recipient strain and male offspring are again backcrossed to the recipient strain. This is repeated for 10 generations, always using background strain females. Such strains have been used to study the phenotypic effects of genes on the Y chromosome (Kren et al. 2001). Consomic strains provide a powerful set of genetic tools for studying quantitative trait loci. They provide the opportunity of mapping loci with quite small phenotypic effects because sample size can be as large as needed to detect any effect thought to be of biological significance (Nadeau et al. 2000). One note of caution: QTLs may interact with one another, so a difference seen on one genetic background may not be observed on a different background.
Maintenance of Consomic Strains Once backcrossing is completed, each consomic strain is an inbred strain in its own right and should be maintained in the same way as an inbred strain, as discussed before. The nomenclature of consomic strains is given in Appendix 9.1.
84550.indb 230
11/11/10 6:25:31 PM
Laboratory Animal Genetics and Genetic Quality Control
231
Genetically Altered (GA) Animals (Mutants, Polymorphisms, Transgenes, and Targeted Mutations Including “Knockout” and “Knockin” Mutations) Several thousand mutants, polymorphisms, transgenes, and targeted mutations are now used in biomedical research. The numbers will increase substantially over the next few years as targeted mutations and mutants produced by chemical mutagenesis and gene trapping continue to be produced. These have a Mendelian mode of inheritance, though there may be variation in gene expression and viability that upset normal Mendelian ratios. The nomenclature of mutants, polymorphisms, and genetically altered animals is given in Appendix 9.1. Spontaneous Mutations Spontaneous mutations may result in the alteration of a single base pair that may, in turn, affect a coding or regulatory region or be genetically silent. Alternatively, a deletion or insertion of a larger tract of DNA may occur. In some cases, such as the dilute coat color allele in DBA/2 mice, the mutation has been caused by the insertion of a retrovirus within the gene (Jenkins et al. 1981). Large chromosomal rearrangements, such as translocations and inversions, are also classified as mutations. Spontaneous mutation rates at five coat-color loci ranged from 3.0 × 10 –3 to 3.9 × 10 –6 (Schlager and Dickie 1967), although rates at other loci appear to be at the lower end of this range. Most of the coat-color variants of laboratory mice have resulted from spontaneous mutations that were preserved, often by mouse fanciers. More recently, when abnormal individuals have been found and breeding has shown that this is due to a mutation, they have been preserved because of their potential biomedical interest. Probably the most widely used of these are the athymic nude mouse (Foxn1nu) and rat (Foxn1rnu), the obese (Lprob) and diabetic mouse (Leprdb), the fatty rat (Leprfa), the scid (severe combined immune deficiency) mouse (Prkdcscid), and the hairless (hr) mouse, although there are several hundred others. Table€9.4 gives examples of some mutations used in biomedical research. Note that when a mutation occurs, it is usually given a descriptive name such as “nude” and a gene symbol such as nu. However, when the locus is identified, the temporary symbol becomes an allele of that locus (in this example, Foxn1nu). Prior to genetic engineering, spontaneous mutations were the main source of mouse models for disease. They were easily identified in large commercial production colonies such as the Jackson Laboratory in the United States and the Medical Research Council unit in Harwell, UK. Unlike many of the next-generation knockouts, these spontaneous mutations were not complete loss-of-function alleles and provided unique insight to gene function, based on point mutations that resulted in subtle changes in gene function. Despite the overwhelming number of genetically engineered mice, spontaneous mutations continue to play a unique role in the study of gene function. So-called “quiet” mutations, with no obvious phenotype, can also occur within inbred strains. These are often found by accident when the strain happens to be used in an unrelated investigation (Stevens et al. 2007). For example, the lipopolysaccharide mutation Tlr4 (previously Lps) was found in strain C3H/He mice several years after it had apparently occurred, when an abnormal response was found following challenge with bacterial lipopolysaccharide. Polymorphisms Polymorphisms are genetic alterations that have little or no obvious phenotypic effect, although in some cases they are maintained in a population because heterozygotes have some survival advantage. Once they occur, they may be retained indefinitely, either segregating in an outbred stock or fixed in different inbred strains. These are widely used as genetic markers in applied genetics research. For example, one of the first polymorphisms to be discovered was in the hemoglobin beta
84550.indb 231
11/11/10 6:25:31 PM
232
Handbook of Laboratory Animal Science
Table€9.4╅ Examples of Old Mutants That Have Become Redesignated as Alleles of an Identified Gene Allele Name
Description
Former Symbol
Brown
Brown coat color
B
Beige
Beige coat color
Bg
Albino Dominant cataract
Albino coat color Eye cataracts
C CatFr
Diabetes Dwarf
Diabetes Dwarf
Jimpy
Tremors and convulsions Dwarfism
Little
Lipopolysaccharide response Retinal degeneration 1
Resistant to bacterial lipopolysaccharide Progressive blindness
Db Dw
Jp Lit
Lpsd Rd1
Gene Name
New Symbol
Tyrosinase-related protein Lysosomal trafficking regulator Tyrosinase Major intrinsic protein of eye lens fiber Leptin receptor Pituitary specific transcription factor 1 Myelin proteolipid protein Growth hormone releasing hormone receptor Toll-like receptor-4
Tyrp1b
Phosphodiesterase 6B, cGMP (rod receptor), beta polypeptide
Lystâ•›bg
Tyrâ•›c MipCat-Fr
Leprâ•›db Pit1dw
Plpjp Ghrhrâ•›lit
Tlr4Lps-d Pde6brd1
chain gene, which can be detected electrophoretically as either a single or a double band on a gel. “Biochemical” polymorphisms of this type were widely used in research before more convenient DNA-based markers became available. Mini- and Microsatellites Mini- and microsatellites are repetitive sections of noncoding (usually) DNA that are often polymorphic in the number of repeats. The polymorphisms in minisatellites, which may be several kilobases long, have been used to produce DNA fingerprints in which virtually all humans except monozygous twins differ (Jeffreys, Wilson, and Thein 1985). These are also present in laboratory species; for example, each inbred mouse strain has a different DNA fingerprint. However, all individuals are identical within a fully inbred strain. Microsatellites are shorter, noncoding simple DNA sequences of perhaps 50–200 repeats, each of which has unique sequence flanking DNA. There are thousands of these in the genome, and they are widely used as genetic markers for genetic mapping and genetic quality control. Although microsatellites are noncoding, they have been implicated in human diseases, such as Huntington’s disease, where microsatellites in an intron have expanded, leading to malfunctioning of the Huntington gene (Imarisio et al. 2008). Single Nucleotide Polymorphisms Single nucleotide polymorphisms (SNPs, pronounced “snips”) are, as their name suggests, single base variants at a given locus in a particular species. They can occur anywhere in the
84550.indb 232
11/11/10 6:25:31 PM
Laboratory Animal Genetics and Genetic Quality Control
233
genome, including coding, regulatory, and noncoding regions. They are common in the mouse. When the full DNA sequences of 15 inbred mouse strains were determined, a total of over 8,000,000 SNPs was discovered (Frazer et al. 2007). An average rate of one SNP per 1,060 base pairs has been quoted for the variation among common laboratory mouse strains, and one SNP per 200 base pairs for variation between laboratory strains and M. m. castaneous (using inbred strain CAST/Ei), which separated from M. m. domesticus approximately one million years ago (Lindblad-Toh et al. 2000). Single nucleotide polymorphisms are used as genetic markers in genetic mapping and for testing the integrity and identity of inbred strains; they can also be used to assess the genetic similarity between strains. For example, they have been used to show the genetic similarity among 55 inbred mouse strains, including some wild strains (Tsang et al. 2005). In a study of rat inbred strains, there was substantial variation among sublines of the same inbred strain, suggesting a considerable amount of residual heterozygosity at the time at which the sublines were separated. Mutation was thought to have been an unlikely cause (Smits et al. 2004). SNPs can also be used as an aid to identify QTLs by associating individual SNPs with phenotypic effects in segregating populations (Smits et al. 2004). Many QTLs may be the result of SNPs altering gene function, either because they occur in a regulatory region of DNA or because they slightly alter one of the resulting polypeptides. Mutagenesis Programs In the last decade, several mutagenesis methods have been developed in the mouse that aim to mutate a significant proportion of the estimated 25,000 genes in the mouse. Two approaches are being used: • Phenotype-driven methods involve inducing mutations identified by a phenotype, with the mutated locus subsequently being mapped and identified by positional cloning. Methods include chemical mutagenesis, usually with ENU (ethyl-nitroso-urea) and gene trapping discussed later. These methods are suitable for large-scale programs, but no specific genes are targeted. • In gene-driven methods, a defined gene is mutated, then any phenotypic effects are studied. This involves targeted mutagenesis using embryonic stem cells and homologous recombination. ENU mutagenesis can also be used to mutate specific genes now that gene sequencing methods are highly automated. These techniques, discussed later, will increasingly be used to mutate genes that have not been mutated by gene trapping and classical ENU mutagenesis.
Phenotype-Driven Mutagenesis Several large-scale mutagenesis programs in the mouse use ENU as the mutagen. Briefly, the technique involves treating male mice with ENU and mating them with untreated females about 6 weeks later. The F1 offspring can then be screened for any dominant mutations. The mice are typically examined for any obvious abnormalities and are then put though a screening test appropriate to the desired type of mutation. Recessive mutations can only be identified by additional rounds of breeding. The F1 animals are mated and the resulting offspring are mated back to their parents (usually daughters mated with fathers), and these offspring are again screened for possible mutations. A large number of standardized screens have been developed (Brown et al. 2005). Once a mutation has been confirmed by further breeding, it must be mapped and identified—a laborious process and one of the drawbacks of this approach. One advantage of ENU mutagenesis is that it usually only changes a single base pair, rather than knocking out the whole gene. This means that several alleles can be produced at each locus that differ in their expression and that there is a smaller chance that the mutation will be lethal.
84550.indb 233
11/11/10 6:25:31 PM
234
Handbook of Laboratory Animal Science
ENU mutagenesis can also be used in a gene-driven approach. Sperm of male offspring of mutagenized mice are preserved and DNA of the mice is prepared. An investigator looking for mutations at a particular locus will then sequence the chosen gene in the preserved DNA to see if any of the male mice were mutated at this locus. Any mutations found can then be recovered from the frozen sperm (Acevedo-Arozena et al. 2008) by using it to inseminate females. The resulting offspring should be carrying the mutation. Genotype-Driven Mutagenesis: Transgenic Strains Transgenic strains were first developed in the 1980s by injecting foreign DNA into the male pronucleus of a one-cell embryo. This then became incorporated into the host chromosomes at a “random” location. An early example was a transgenic mouse strain expressing rat growth hormone under the control of a metallothionine promoter (Palmiter et al. 1982). When treated with copper sulfate in the water to induce activity of the transgene, the animals grew much faster than normal, although they also developed a range of other abnormalities. Another example, the “oncomouse,” is a transgenic mouse carrying an activated v-Ha-ras oncogene under the control of the mouse mammary tumor virus promoter (Colombo et al. 2001). It gets a high incidence of mammary tumors and is a potential model for the development of anticancer drugs. However, its development was also controversial because it was patented by Harvard University and thus not made freely available as had been the tradition up until that time. The method is still used. Injected embryos are placed in a foster mother and allowed to come to term. The offspring can then be tested to see whether the foreign DNA has become incorporated and, in appropriate cases, whether it is expressed. The technique can be used to overexpress a gene, to replace a faulty gene, or to express a gene from another species. Many copies of the DNA construct in a head–tail array are usually incorporated. If the construct is a functional gene with promoter sequences, then it will usually be expressed. This technique has been used in farm animals as a method of producing large quantities of a foreign protein in milk, for instance. Care must be taken when using transgenic mice to monitor copy number by qPCR and phenotype because methylation can cause trangenes to lose expression. Gene Trapping Gene trapping is a high-throughput method of randomly mutating genes expressed in mouse embryonic stem cells by insertional mutagenesis. Briefly, a DNA construct consists of a splice acceptor, selectable and reporter markers, and a polyadenylation signal at the 3′ end that causes the mRNA to be nonfunctional. When inserted into an expressed gene, the construct is transcribed from the promoter of the gene where the gene trap inserted. A range of constructs are available with additional sequences that aid further manipulation of the trapped gene. The construct is transfected into the cells, usually by electroporation. If it is incorporated into a gene expressed in the ES cells, it is activated by that gene’s promoter. The DNA surrounding the construct can be sequenced and the mutated gene identified. Although the trapped gene is usually inactivated, there may just be a loss of function, depending on where in the trapped gene the construct has become integrated. The International Gene Trap Consortium (IGTC) currently has a bank of over 45,000 well-characterized mutated ES cell lines comprising about 40% of known mouse genes; these are available to investigators free of charge (Nord et al. 2006). Its Web site (www.genetrap.org) allows investigators to search for mutated genes or sequences, but investigators must culture the ES cells and insert them into blastocysts in order to recover live mice.
84550.indb 234
11/11/10 6:25:32 PM
Laboratory Animal Genetics and Genetic Quality Control
235
Targeted Mutations (“Knockout” and “Knockin” Alterations) Knockout mice are produced by inactivating a defined gene using homologous recombination. Very briefly, an inactive copy of the gene of interest can be synthesized in vitro, usually by splicing a gene for resistance to neomycin into one of the exons. This provides a positive selectable marker. A thymidine kinase gene giving a toxic product when treated with gancyclovir is usually spliced at one end of the construct as a negative selectable marker. The construct is then incorporated into ES cell cultures by electroporation, microinjection, or incorporating it into a virus and infecting the cells. The DNA construct recognizes the homologous sequence in the ES cell DNA and, in rare cases, homologous recombination occurs with the construct (complete with the neomycin gene) replacing the ES cell gene. In these cases, the thymidine kinase gene is usually lost, yielding cells where this has occurred that are resistant to both neomycin and gancyclovir. The modified ES cells are then injected into blastocyst stage embryos, usually from an unrelated strain, and returned to a foster mother. Some of the resulting offspring are chimeras composed of cells from the host blastocyst and the ES cells. Coat color markers are usually used to allow chimeric offspring to be easily identified. In some cases, the gonads of the chimeric animals will also be chimeric, permitting offspring to be produced whose genomes contain the knocked-out gene. An improvement to gene knockout technology is the engineering of conditional mutations, using the Cre/lox technology. In designing a conditional allele, Lox P sites from the bacteriophage P1 are engineered into constructs to flank the DNA sequence of the gene. The enzyme cre recombinase is a site-specific DNA recombinase that is then used to catalyze the recombination of DNA between these Lox P sites. Mice carrying a “floxed” allele (i.e., with a LoxP site at each end) can then be mated to a variety of engineered transgenic cre lines, where different promoters drive expression of cre recombinase only in certain organs or tissues or at specific times (determined by the investigator), resulting in time- or tissue-specific manipulation of genes. Complete gene knockout mice can still be derived from conditional alleles by simply mating a mouse with the floxed allele to a mouse that carries cre driven by a promoter expressed in early embryonic development that then promotes recombination in the germ line of the resulting progeny. Conditional mutagenesis was critical in the study of gene function, where complete knockouts resulted in embryonic lethality. When working with conditional alleles, it is important to use cre lines whose patterns of expression have been well characterized. The “knockin” technique can be used to replace an existing mouse gene with one from another species or to overexpress a gene in a more defined way than would be possible using microinjection. The disadvantage of simple microinjection is that a tandem array of genes is often incorporated into some undefined region of a chromosome. One advantage of “knocking in” a sequence to a specific locus is that the expression is more likely to reflect the patterns of the endogenous promoter, since all of the upstream regulatory sequences surrounding that promoter are present. However, with fidelity of expression pattern, one often sacrifices the high expression levels achieved with a microinjected transgene. The choice of engineering will depend on the experimental model one wishes to produce. Targeted mutagenesis and the application of genetic engineering using ES cell technology have been extremely successful in mice. However, the lack of germ line competent ES cell lines has prevented many genome-specific manipulations in rats. RNA interference (RNAi) is a mechanism in which double-stranded RNAs with a specific base sequence inhibit corresponding gene expression by inducing degradation of its mRNA; it has provided a promising advance for rat genetics. Using traditional transgenesis or lentiviral vectors, one can use RNAi to produce transgenic animals expressing short hairpin RNA (shRNA) to “knock down” specific gene expression. An overview of these techniques is given by Silver (1995), and more detailed descriptions of methods of producing transgenic animals of various species have been published by Maclean (1994).
84550.indb 235
11/11/10 6:25:32 PM
236
Handbook of Laboratory Animal Science
Genetic engineering techniques are advancing rapidly at present, particularly in the development of techniques that make it easier to control gene expression in particular tissues and at particular times. Genetic Background of Mutants and Transgenes The phenotype resulting from either a spontaneous or induced mutation is virtually always subject to modification by other genes. A mutation that causes death or severe disability on one genetic background may be viable and have mild to nonexistent effects on another. When a mutant or transgene is on a genetically segregating background, it is nearly always good research practice to backcross to one or more inbred backgrounds before starting detailed investigation of the phenotype (Linder 2001). If this is not done, then the phenotypic characteristics may drift quite rapidly as a result of inbreeding, genetic drift, and natural selection, unless the colony is maintained as described for outbred stocks in an earlier section. This may mean that results from studies done soon after the genetic modification has been done may be difficult to replicate at a later date. Breeding and Maintenance of Mutant and Transgenic Strains The breeding methods needed to maintain these strains will depend on the mode of inheritance of the mutant, including whether it is dominant, codominant, or recessive; whether it has a distinct phenotype; and whether all classes of stock are viable and fertile. The genetic background (inbred or outbred) also needs to be taken into account. It will also depend on whether there is a DNA-based method of identifying the gene of interest. A brief description of some of the more common situations is given next. However, more complex breeding systems may be needed to produce animals with a desired combination of mutant genes or when identification of genotype is difficult. Genetic mapping studies may involve quite complex breeding schemes, but due to space limitations, they cannot be discussed here. Both Sexes Fully Viable and Fertile. Mutants in which both sexes are fully viable and fertile can be maintained according to the genetic background. If this is inbred, then the methods for maintaining inbred strains should be used. If the mutant/transgene is on a heterogeneous, outbred background and there are no plans to backcross to an inbred strain (as might be the case with large animals) and if the aim is to maintain the colony for some time, then it should be maintained using the methods for an outbred stock. In order to prevent inbreeding and genetic drift, at least 25 breeding pairs should be used each generation with random mating or about 13 pairs if a maximum avoidance of inbreeding system is to be used. With large animals, as already noted, it may be necessary to maintain more males than necessary for reproductive purposes in order to minimize inbreeding. Smaller colonies can be maintained for short periods, although substantial genetic drift and change in expression of the mutation may result. A Recessive Mutation with One Sex Infertile. When one sex is infertile, the mutant is usually maintained by mating a homozygous animal of the viable sex with a heterozygous animal of the other sex. This is the common situation with nude mice, where the females often display poor breeding performance (depending on genetic background), but males are fully fertile. With this system, half the offspring will be of the desired mutant phenotype and the other half will carry the gene and the appropriate sex can be used for further breeding. Matings need to take into account the genetic background—either inbred or outbred—and should be maintained as such. A Dominant Mutation with Homozygous Lethality. Some dominant genes, such as the yellow allele Ay at the agouti locus, are lethal in the homozygous state. In this case, matings between heterozygous
84550.indb 236
11/11/10 6:25:32 PM
Laboratory Animal Genetics and Genetic Quality Control
237
mutant animals and wild-type ones are performed, with approximately half of the offspring displaying the mutant phenotype. Again, account needs to be taken of the genetic background. Recessive Mutation with Both Sexes Infertile. Recessive mutations resulting in both sexes being infertile are quite common for mutations with serious phenotypic effects, such as the obese and diabetic mutations in mice. If two heterozygotes are mated, on average a fourth of the offspring will be mutant, half will be heterozygous carriers, and a quarter will be homozygous wild type; the latter two classes will be phenotypically indistinguishable. Only the heterozygotes will be useful for further breeding. The classical way of dealing with this situation is to set up test matings in order to identify heterozygotes. Known heterozygous animals can be crossed with individuals of unknown genotype. A quarter of the offspring are expected to be homozygous mutants if the test animal is heterozygous. If no mutant progeny are produced among the first 10 offspring, then the animal is presumed not to be a heterozygote. Several animals must be tested in this way in order to identify enough carriers to maintain the colony and produce experimental material. Unfortunately, this is time consuming and inefficient. Another alternative is to perform random matings among the offspring of heterozygous matings. Two-thirds of the animals are, on average, heterozygous, so 4/9 (2/3 × 2/3) of the matings will be between heterozygotes, but only one-fourth of the offspring of such matings will be homozygous mutants. Again, if no mutants are produced in about the first 10 offspring, it will be assumed that the mating is not between two heterozygotes, permitting culling of the animals. Again, though relatively simple, this is inefficient. If the mutant is on an inbred genetic background, an alternative method is to graft ovaries from a homozygous mutant animal into a wild-type female of the same inbred strain. This animal is then mated to a wild-type male to produce offspring known to be heterozygous for the mutation and suitable for breeding. When a mutant locus has been sequenced and the nature of the mutation identified, it is often possible to develop a PCR-based method of genotyping individuals. All that is needed is a small sample of DNA and the laboratory capability of running the appropriate tests. Such a method has been developed for the diabetes (Horvat and Bunger 1999) and obese (Namae et al. 1998) mutations in mice and will be available for many additional mutants now that the full DNA sequence of the mouse is available. Breeding Transgenes and Mutations Produced by Gene Targeting (Knockouts) Carriers of a transgene can be of three types: hemizygotes, with one copy of the transgene but no normal host allele; heterozygotes, in the case of a gene targeted mutation, with one mutant and one normal wild-type allele; or homozygotes with two copies of the transgene. Identification of carriers usually presents no problems because a PCR-based test will usually be available to identify them. Thus, all that is needed is a sample of DNA, typically obtained from the tail tip, an ear punch, a hair bulb sample, or saliva (Irwin et al. 1996). Once carriers have been identified, matings can be established to continue to maintain the transgene, to backcross it to an inbred background, or to produce homozygous animals (if the transgene is in a hemizygous or heterozygous state). Problems may arise in differentiating between animals that are homozygous (two copies of the transgene) and those that are hemizygous or heterozygous (one copy of the transgene). This will require a quantitative PCR (McPherson, Quirke, and Taylor 1991) or Southern blot analysis, or a progeny test. The latter involves mating the animal (assumed to be fertile and viable) to some wild-type animals and then testing the progeny for the presence of the transgene. If all the progeny carry a copy of the transgene (8–10 progeny are typically tested), then the animal of interest can be assumed to be homozygous.
84550.indb 237
11/11/10 6:25:32 PM
238
Handbook of Laboratory Animal Science
Genetic Quality Control Aims The aim of genetic quality control programs is to detect genetic contamination of one strain through an inadvertent mating with another strain. Currently, it is not possible to monitor strains for new mutations except by observing the phenotype. As a result, mutations that affect invisible characters, such as minor changes in immune response, physiology, or susceptibility to infectious organisms, may go undetected for many generations (Stevens et al. 2007). Technical Methods Historically, several methods of genetic quality control have been employed, ranging from the use of biochemical polymorphisms to the use of quantitative characters, such as the shape of the skeleton and breeding performance (Nomura, Esaki, and Tomita 1984; Hedrich 1990). Biochemical polymorphisms were often technically difficult to determine and somewhat limited in their distribution among strains of mice and rats. Skeletal morphology and breeding performance have the disadvantage of giving a statistical result, rather than a clear-cut positive or negative answer. Breeding performance should still be routinely monitored for husbandry purposes and any changes investigated. Any sudden increase in breeding performance in an inbred strain may represent hybrid vigor as a function of genetic contamination. The development of a large number of microsatellite and other DNA-based genetic markers, such as SNPs, has now changed the situation. However, changes in phenotype noticed by animal technicians or scientific users of the animals continue to be an important means to identify genetic contamination. The great advantage of DNA-based methods is that only a small sample of tissue that can be stored indefinitely in the deep freeze is needed. Should an experiment give unexpected results, the genotypes of the mice or rats can then be investigated. When colonies of genetically altered mice are maintained, allele-specific PCR-based assays can be designed to test for the presence of the allele. Even in a homozygous state, breeders should be routinely confirmed for the genotype. The main difficulties with genetic quality control are in determining the number of genetic loci to test, the appropriate sample size, and appropriate sampling frequency. For routine monitoring, the sample size depends primarily on the presumed extent of potential genetic contamination. A high level of contamination (above 20%) can be detected with small sample sizes, but it is virtually impossible to detect a small number of wrong matings in a colony of 1,000 breeding cages. Table€9.5 shows the sample size required to detect different levels of genetic contamination at a specified level of probability. It should be clear that the best approach is to do everything possible to avoid contamination in the first place. Genetic Monitoring of Isogenic Strains Inbred strains are relatively easy to monitor because all individuals should be identical at all loci, apart from rarely occurring recent mutations. However, at present, there are no agreed-upon standards on the number of loci to test, appropriate sample sizes, and appropriate sampling frequency. Generally, the effort expended in monitoring each strain should depend on the chance of genetic contamination, while accounting for the importance of the colony and the potential damage to research or reputation as a result of a contamination. Danger arises when colonies are first established because there are no complete assurances that the animals are what they are supposed to be. Additionally, maintaining animals of several strains in the same animal room will clearly increase the probability of a wrong
84550.indb 238
11/11/10 6:25:32 PM
Laboratory Animal Genetics and Genetic Quality Control
239
Table€9.5╅Sample Size Needed to Give a 95% Chance of Detecting Genetic Contamination for Given Levels of Contaminated Animals in the Colony Percent Contamination 2 5 10 15 20 30 40 70
Sample Size (S) 148 58 28 18 13 8 6 2
Note: Formula used is S = log(p)/log(U), where p is the chance of missing the contamination (in this case, p = .05) and U is the proportion of animals in the colony that are uncontaminated.
mating. Staff need to be well trained and should be strongly encouraged to report anything that they consider to be unusual, without fear of being blamed if something has gone wrong. Authentication of Newly Established Strains Ideally, newly established colonies should be tested as rapidly as possible to ensure that they are of the correct genotype. With mice, control DNA from most strains is available from the Jackson Laboratory (www.jax.org). For rats, samples of DNA may be obtained from colleagues or known holders of the strains. If the colony is to be established from a small breeding nucleus, it may be possible to test all animals. In this case, the main aim is to test the authenticity of the strain, though the possibility of contamination by one or more nonstrain animals is also a possibility. A sample of 5–10 animals should be adequate at this stage, and they should be tested at at least 10 microsatellite loci or a critical set of SNPs. Existing Colonies Breeding colonies of inbred strains will often be divided up into a stem line and an expansion colony (see earlier discussion). Ideally, the stem line colony will be physically separated from other colonies. If not, it should at least be kept with strains of a different coat color. With good physical separation and a small colony size, the chance of contamination is low, reducing the frequency with which the colony needs to be monitored. If the colony is maintained in an isolator with no other strains, then it will hardly need routine monitoring after initial authentication. Expansion colonies may be large and at risk from other colonies in the same building. The colony might be monitored two to four times per year, with sample sizes of about 10 or more animals from as many different cages as possible, using a set of markers that preclude contamination by all other strains in the building. Even so, it is unlikely that a low level of genetic contamination, possibly in only one section of a large colony, can be detected. Troubleshooting In practice, genetic contamination is often first noticed by the users of the animals as a function of unexpected results. If DNA samples can be obtained from the abnormal animals (scientists
84550.indb 239
11/11/10 6:25:32 PM
240
Handbook of Laboratory Animal Science
should keep frozen tissue samples of their animals), then there is a very good chance that contamination will be obvious from a few genetic markers. Genetic Monitoring of Outbred Stocks Outbred stocks are subject to genetic drift as well as the possibility of genetic contamination, and any genetic monitoring scheme should aim to detect both. In contrast to inbred strains, there can be no authentication stage because there is no “correct” genotype for an outbred stock. Outbred animals with names like Wistar rats and Swiss mice from different colonies differ, and there is no international or national standard to identify the “correct” genotype. One further complication is that the outbred stock may turn out to be quite inbred (i.e., homozygous), if it has been maintained as a closed colony for many years. In this case, it may be quite difficult to find segregating markers suitable for monitoring genetic drift. As with inbred strains, the preferred technical methods for monitoring outbred stocks involve the use of DNA-based genetic markers. However, the use of microsatellite markers is not as easy with outbred stocks because there may be several different alleles present at each locus. If there are small differences between alleles, then classification of individual bands can be difficult. But it should be possible to identify heterozygous animals with two bands. Recently, large numbers of SNP markers have been used to study outbred stocks (Aldinger et al. 2009). These have been able to identify genetic differences between samples of the same outbred stock from the same breeder, but from different rooms. There is no recent theoretical work on sample sizes, choice of markers, or frequency of monitoring in outbred stocks. It is also possible that the availability of large numbers of SNP markers with highly automated apparatus may solve the technical problems but open up entirely new questions. For example, if mice of the same stock designation and breeder but from different rooms are genetically different, it seems highly likely that each colony of an outbred stock may be genetically unique. (In fact, genetic theory predicts this. When daughter colonies are set up, only a sample of genes is passed from the parent colony, so the two colonies are immediately different due to the “founder effect.”) Genetic monitoring might be useful for detecting genetic contamination of an outbred stock as a result of an inadvertent cross with another stock. But is this really worthwhile? If investigators want to have repeatable results, they would be much better off using inbred or F1 hybrid strains. Comparisons of Different Colonies Breeders often want to maintain the same outbred stock at different locations and need a method to test the extent to which the colonies have drifted apart. The best approach at this stage would appear to be to commission a laboratory with the capability of detecting SNPs to sample both colonies, possibly using 20–30 animals from each colony. The main problem would be interpreting the results. How different would they have to be to be considered “different”? References Acevedo-Arozena, A., S. Wells, P. Potter, M. Kelly, R. D. Cox, and S. D. Brown. 2008. ENU mutagenesis, a way forward to understand gene function. Annual Review of Genomics and Human Genetics 9:49–69. Adamovic, T., D. McAllister, J. J. Rowe, T. Wang, H. J. Jacob, and S. L. Sugg. 2008. Genetic mapping of mammary tumor traits to rat chromosome 10 using a novel panel of consomic rats. Cancer Genetics and Cytogenetics 186:41. Aldinger, K. A., G. Sokoloff, D. M. Rosenberg, A. A. Palmer, and K. J. Millen. 2009. Genetic variation and population substructure in outbred CD-1 mice: Implications for genome-wide association studies. PLoS One 4:e4729.
84550.indb 240
11/11/10 6:25:33 PM
Laboratory Animal Genetics and Genetic Quality Control
241
Bailey, D. W. 1971. Recombinant inbred strains, an aid to finding identity, linkage, and function of histocompatibility and other genes. Transplantation 11:325. Beck, J. A., S. Lloyd, M. Hafezparast, M. Lennon-Pierce, J. T. Eppig, M. F. W. Festing, and E. M. C. Fisher. 2000. Genealogies of mouse inbred strains. Nature Genetics 24:23. Bradley, D. J. 1982. Models of complex host–parasite relationships: Murine leishmaniasis. In Animal models in parasitology, ed. D. G. Owen, 69–80. London: Macmillan Press. Brown, S. D., P. Chambon, and M. H. de Angelis. 2005. EMPReSS: Standardized phenotype screens for functional annotation of the mouse genome. Nature Genetics 37:1155. Collins, A. C., and M. J. Marks. 1992. Genetic studies of nicotine and nicotine/alcohol reactivity in humans and animals. In Genetically defined animal models of neurobehavioral dysfunctions, ed. P. Driscoll, 146–173. Boston: Birkhäuser. Colombo, L. L., M. C. Lopez, G. Chen, and R. R. Watson. 2001. In vitro response of v-Ha-ras transgenic mouse lymphocytes after in vivo treatment with alcohol. Immunopharmacology and Immunotoxicology 23:597. Cordell, H. J., and J. A. Todd. 1995. Multifactorial inheritance in type I diabetes. Trends in Genetics 11:499. Cormier, R. T., A. Bilger, A. J. Lillich, R. B. Halberg, K. H. Hong, K. Gould, N. Bornstein, E. S. Lander, and W. F. Dove. 2000. The Mon1AKR intestinal tumor resistance region consists of Pla2g2a and a locu distal to D4Mit64. Oncogene 19:3182. Cowley, A. W., R. J. Roman, M. L. Kaldunski, P. Dumas, J. G. Dickhout, A. S. Green, and H. J. Jacob. 2001. Brown Norway chromosome 13 confers protection from high salt to consomic Dahl S rat. Hypertension 37:456. Darvasi, A., and M. Soller. 1995. Advanced intercross lines: An experimental population for fine mapping. Genetics 141:1199. Demant, P., and A. A. M. Hart. 1986. Recombinant congenic strains—A new tool for analyzing genetic traits determined by more than one gene. Immunogenetics 24:416. Dubois, E. L., R. E. Horowitz, H. B. Demopaulos, and R. Teplitz. 1966. NZB/NZW mice as a model of systemic lupus erythematosus. Journal of the American Medical Association 195:285. Evans, M. F., and M. H. Kaufman. 1981. Establishment in culture of pluripotential cells from mouse embryos. Nature 292:154. Falconer, D. S. 1981. Introduction to quantitative genetics. London: Longman. Feingold, N., J. Feingold, D. Mouton, Y. Bouthillier, C. Stiffel, and G. Biozzi. 1976. Polygenic regulation of antibody synthesis to sheep erythrocytes in the mouse: A genetic analysis. European Journal of Immunology 6:43. Festing, M. F. W. 1976. Phenotypic variability of inbred and outbred mice. Nature 263:230. ———. 1990. Contemporary issues in toxicology: Use of genetically heterogeneous rats and mice in toxicological research: A personal perspective. Toxicology and Applied Pharmacology 102:197. ———. 1995. Use of a multistrain assay could improve the NTP carcinogenesis bioassay program. Environmental Health Perspectives 103:44. ———. 1997. Fat rats and carcinogen screening. Nature 388:321. ———. 1999. Warning: The use of genetically heterogeneous mice may seriously damage your research. Neurobiology of Aging 20:237. Festing, M. F. W., and E. M. C. Fisher. 2000. Mighty mice. Nature 404:815. Festing, M. F. W., K. Kondo, R. Loosli, S. M. Poiley, and A. Spiegel. 1972. International standardized nomenclature for outbred stocks of laboratory animals. ICLA Bulletin 30:4. Festing, M. F. W., D. May, T. A. Connors, D. P. Lovell, and S. Sparrow. 1978. An athymic mutation in the rat. Nature 274:365. Festing, M. F. W., A. Yang, and A. M. Malkinson. 1994. At least four genes and sex are associated with susceptibility to urethane-induced pulmonary adenomas in mice. Genetical Research 64:99. Fortin, A., L. R. Caradon, M. Tam, E. Skamene, M. M. Stevenson, and P. Gross. 2001. Identification of a new malaria susceptibility locus (Char4) in recombinant congenic strains of mice. Proceedings of the National Academy of Sciences 98:10793. Frazer, K. A., E. Eskin, H. M. Kang, M. A. Bogue, D. A. Hinds, E. J. Beilharz, R. V. Gupta, et al. 2007. A sequence-based variation map of 8.27 million SNPs in inbred mouse strains. Nature 448:1050. Gordon, J. W., G. A. Scangos, D. J. Plotkin, J. A. Barbosa, and F. H. Ruddle. 1980. Genetic transformation of mouse embryos by microinjection of purified DNA. Proceedings of the National Academy of Sciences USA 77:7380.
84550.indb 241
11/11/10 6:25:33 PM
242
Handbook of Laboratory Animal Science
Gregorova, S., P. Divina, R. Storchova, Z. Trachtulec, V. Fotopulosova, K. L. Svenson, L. R. Donahue, B. Paigen, and J. Forejt. 2008. Mouse consomic strains: exploiting genetic divergence between Mus m. musculus and Mus m. domesticus subspecies. Genome Research 18:509. Grüneberg, H. 1952. The genetics of the mouse. The Hague: Nijhoff. Hedrich, H. J. 1990. Genetic monitoring of inbred strains of rats. Stuttgart: Gustav Fischer Verlag. Heller, D. A., F. M. Ahern, J. T. Stout, and G. E. McClearn. 1998. Mortality and biomarkers of aging in heterogeneous stock (HS) mice. Journal of Gerontology A: Biological Sciences and Medical Sciences 53:B217–B230. Horvat, S., and L. Bunger. 1999. Polymerase chain reaction-restriction fragment length polymorphism (PCRRFLP) assay for the mouse leptin receptor (Lep(db)) mutation. Laboratory Animals 33:380. Hutt, F. B. 1979. Genetics for dog breeders. San Francisco: W. H. Freeman and Company. Imarisio, S., J. Carmichael, V. Korolchuk, C. W. Chen, S. Saiki, C. Rose, G. Krishna, et al. 2008. Huntington’s disease: From pathology and genetics to potential therapies. Biochemical Journal 412:191. Irwin, M. H., R. J. Moffatt, and C. A. Pinkert. 1996. Identification of transgenic mice by PCR analysis of saliva. Nature Biotechnology 14:1146. Ito, T., M. Takahashi, K. Sudo, and Y. Sugiyama. 2007. Marked strain differences in the pharmacokinetics of an alpha4beta1 integrin antagonist, 4-[1-[3-Chloro-4-[N-(2-methylphenyl)-ureido]phenylacetyl]-(4S)-fluoro-(2S)-pyrrolidine-2-yl]-methoxybenzoic acid (D01-4582), in Sprague–Dawley rats are associated with albumin genetic polymorphism. Journal of Pharmacology and Experimental Therapeutics 320:124. Jay, G. E. 1955. Variation in response of various mouse strains to hexobarbitol (Evpal). Proceedings of the Society of Experimental Biology and Medicine 90:378. Jeffreys, A. J., V. Wilson, and S. W. Thein. 1985. Hypervariable “minisatellite” regions in human DNA. Nature 314:67. Jenkins, N. A., N. G. Copeland, B. A. Taylor, and B. K. Lee. 1981. Dilute coat color mutation of DBA/2J mice is associated with site of integration of an ecotropic MuLV genome. Nature 293:370. Kacew, S., and M. F. W. Festing. 1996. Role of rat strain in the differential sensitivity to pharmaceutical agents and naturally occurring substances. Journal of Toxicology and Environmental Health 47:1. Kacew, S., Z. Ruben, and R. F. McConnell. 1995. Strain as a determinant factor in the differential responsiveness of rats to chemicals. Toxicologic Pathology 23:701. Klein, J. 1975. Biology of the mouse histocompatibility-2 complex. Berlin: Springer–Verlag. Klempt, M., B. Rathkolb, E. Fuchs, M. H. de Angelis, E. Wolf, and B. Aigner. 2006. Genotype-specific environmental impact on the variance of blood values in inbred and F1 hybrid mice. Mammalian Genome 17:93. Kren, V., N. Qi, D. Krenova, V. Zidek, M. Sladka, M. Jachymova, B. Mikova, et al. 2001. Y-chromosome transfer induces change in blood pressure and blood lipids in SHR. Hypertension 37:1147. Lane-Petter, W., and A. E. G. Pearson. 1972. The laboratory animal—Principles and practice. London: Academic Press. Li, Y., T. R. Gilbert, A. H. Matsumoto, and W. Shi. 2008. Effect of aging on fatty streak formation in a dietinduced mouse model of atherosclerosis. Journal of Vascular Research 45:205. Lindblad-Toh, K., E. Winchester, M. J. Daly, D. G. Wang, J. N. Hirschhorn, J. P. Laviolette, K. Ardlie, et al. 2000. Large-scale discovery and genotyping of single-nucleotide polymorphisms in the mouse. Nature Genetics 24:381. Linder, C. C. 2001. The influence of genetic background on spontaneous and genetically engineered mouse models of complex diseases. Lab Animals (NY) 30:34. Little, C. C., and E. E. Tyzzer. 1916. Further studies on inheritance of susceptibility to a transplantable tumor of Japanese waltzing mice. Journal of Medical Research 33:393. Liu, G. Y., D. Baker, S. Fairchild, F. Figueroa, R. Quartey-Papafio, M. Tone, D. Healey, A. Cooke, J. L. Turk, and D. C. Wraith. 1993. Complete characterization of the expressed immune response genes in Biozzi AB/H mice: Structural and functional identity between AB/H and NOD A region molecules. Immunogenetics 37:296. Maclean, N. 1994. Animals with novel genes. Cambridge, England: Cambridge University Press. Malakoff, D. 2000. The rise of the mouse, biomedicine’s model mammal. Science 288:248. Malkinson, A. M. 1991. Genetic studies on lung tumor susceptibility and histogenesis in mice. Environmental Health Perspectives 93:149. Markel, P., P. Shu, C. Ebeling, G. A. Carlson, D. L. Nagle, J. S. Smutko, and K. J. Moore. 1997. Theoretical and empirical issues for marker-assisted breeding of congenic mouse strains. Nature Genetics 17:280.
84550.indb 242
11/11/10 6:25:33 PM
Laboratory Animal Genetics and Genetic Quality Control
243
Martin, A., G. B. Collin, Y. Asada, D. Varnum, and J. H. Nadeau. 1999. Susceptibility to testicular germ-cell tumors in a 129.MOLF-Chr 19 chromosome substitution strain. Nature Genetics 23:237. McClearn, G. E., and S. M. Hofer. 1999. Genes as gerontological variables: Genetically heterogeneous stocks and complex systems. Neurobiology of Aging 20:147. McClearn, G. E., J. R. Wilson, and W. Meredith. 1970. The use of isogenic and heterogenic mouse stocks in behavioral research. In Contribution to behavior genetic analysis. The mouse as a prototype, ed. G. Lindzey, 3–32. New York: Appleton-Century-Crofts. McPherson, M. J., P. Quirke, and G. R. Taylor. 1991. PCR—A practical approach. New York: IRL Press at Oxford University Press. Miller, R. A., S. Austad, D. Burke, C. Chrisp, R. Dysco, A. Galecki, A. Jackson, and V. Monnier. 1999. Exotic mice as models for aging research: Polemic and prospectus. Neurobiology of Aging 20:217. Miyashita, N., and K. Moriwaki. 1987. H-2 controlled genetic susceptibility to pulmonary adenomas induced by urethan and 4-nitroquinoline 1-oxide in A/Wy congenic strains. Japanese Journal of Cancer Research 78:494. Mott, R., C. J. Talbot, M. G. Turri, A. C. Collins, and J. Flint. 2000. From the cover: A method for fine mapping quantitative trait loci in outbred animal stocks. Proceedings of the National Academy of Sciences USA 97:12649. Nadeau, J. 2001. Modifier genes in mice and humans. Nature Reviews Genetics 2:165. Nadeau, J. H., J. B. Singer, A. Martin, and E. S. Lander. 2000. Analyzing complex genetic traits with chromosome substitution strains. Nature Genetics 24:221. Namae, M., Y. Mori, K. Yasuda, T. Kadowaki, Y. Kanazawa, and K. Komeda. 1998. New method for genotyping the mouse Lep(ob) mutation, using polymerase chain reaction assay. Laboratory Animal Science 48:103. Nomura, T., K. Esaki, and T. Tomita. 1984. ICLAS Manual for genetic monitoring of inbred mice. Tokyo: University of Tokyo Press. Nomura, T., and K. Yonezawa. 1996. A comparison of four systems of group mating for avoiding inbreeding. Genetics Selection and Evolution 28:141. Nord, A. S., P. J. Chang, B. R. Conklin, A. V. Cox, C. A. Harper, G. G. Hicks, C. C. Huang, et al. 2006. The International Gene Trap Consortium Web site: A portal to all publicly available gene trap cell lines in mouse. Nucleic Acids Research 34:D642. Paigen, B., P. A. Holmes, E. K. Novak, and R. Swank. 1990. Analysis of atherosclerosis susceptibility in mice with genetic defects in platelet function. Atherosclerosis 10:648. Palmiter, R. D., R. L. Brinster, R. E. Hammer, M. E. Trumbauer, M. G. Rosenfeld, N. C. Birnberg, and R. M. Evans. 1982. Dramatic growth of mice that develop from eggs microinjected with metallothionine-growth hormone fusion genes. Nature 300:611. Papaioannou, V. E., and M. F. W. Festing. 1980. Genetic drift in a stock of laboratory mice. Laboratory Animals 14:11. Petkov, P. M., Y. Ding, M. A. Cassell, W. Zhang, G. Wagner, E. E. Sargent, S. Asquith, et al. 2004. An efficient SNP system for mouse genome scanning and elucidating strain relationships. Genome Research 14:1806. Potter, M. 1972. Immunoglobulin-producing tumors and myeloma proteins of mice. Physiological Reviews 52:631. Rapp, J. R. 1982. Dahl salt-susceptible and salt-resistant rats. A review. Hypertension 4:753. Roberts, I., I. Kwan, P. Evans, and S. Haig. 2002. Does animal experimentation inform human healthcare? Observations from a systematic review of international animal experiments on fluid resuscitation. British Medical Journal 324:474. Schlager, G., and M. M. Dickie. 1967. Spontaneous mutations and mutation rates in the house mouse. Genetics 57:319. Silva, A. J., M. E. Simpson, J. S. Takahashi, H. P. Lipp, S. Nakanishi, J. M. Wehner, K. P. Giese, et al. 1997. Mutant mice and neuroscience: recommendations concerning genetic background. Neuron 19:755. Silver, L. M. 1995. Mouse genetics. New York: Oxford University Press. Smits, B. M., B. F. Van Zutphen, R. H. Plasterk, and E. Cuppen. 2004. Genetic variation in coding regions between and within commonly used inbred rat strains. Genome Research 14:1285. Snell, G. D., and J. H. Stimpfling. 1966. Genetics of tissue transplantation. In Biology of the laboratory mouse, ed. E. L. Green, 457–491. New York: McGraw–Hill. Stevens, J. C., G. T. Banks, M. F. Festing, and E. M. Fisher. 2007. Quiet mutations in inbred strains of mice. Trends in Molecular Medicine 13:512.
84550.indb 243
11/11/10 6:25:33 PM
244
Handbook of Laboratory Animal Science
Tao, X., F. Fan, V. Hoffmann, C. Y. Gao, N. S. Longo, P. Zerfas, and P. E. Lipsky. 2008. Effective therapy for nephritis in (NZB × NZW)F(1) mice with triptolide and tripdiolide, the principal active components of the Chinese herbal remedy Tripterygium wilfordii Hook F. Arthritis and Rheumatism 58:1774. Taylor, B. A. 1976. Development of recombinant inbred lines of mice. Behavior Genetics 6:118. ———. 1996. Recombinant inbred strains. In Genetic variants and strains of the laboratory mouse, 2nd ed., ed. M. F. Lyon, 1597–1659. New York: Oxford University Press. Thomas, K. R., and M. R. Capecchi. 1987. Site-directed mutagenesis by gene targeting in mouse embryoderived stem cells. Cell 51:503. Thomas, M. A., C. F. Chen, M. I. Jensen-Seaman, P. J. Tonellato, and S. N. Twigger. 2003. Phylogenetics of rat inbred strains. Mammalian Genome 14:61. Tripodis, N., and P. Demant. 2001. Three-dimensional patterns of lung tumor growth: Association with tumor heterogeneity. Experimental Lung Research 27:521. Tsang, S., Z. Sun, B. Luke, C. Stewart, N. Lum, M. Gregory, X. Wu, M. Subleski, N. A. Jenkins, N. G. Copeland, and D. J. Munroe. 2005. A comprehensive SNP-based genetic analysis of inbred mouse strains. Mammalian Genome 16:476. Updyke, L., H. L. Yoon, A. Chuthaputti, R. W. Pfeiffer, and G. K. W. Yim. 1989. Induction of interleukin-1 and tumor necrosis factor by 12-O-tetradecanoylphorbol-13-acetate in phorbol ester-sensitive (SENCAR) and resistant (B6C3F1) mice. Carcinogenesis 10:1107. Valdar, W., L. C. Solberg, D. Gauguier, S. Burnett P. Klenerman, W. O. Cookson, M. S. Taylor, J. N. Rawlins, R. Mott, and J. Flint. 2006. Genome-wide genetic association of complex traits in heterogeneous stock mice. Nature Genetics 38:879. Venta, P. J., J. Li, V. Yuzbasiyan-Gurkan, G. J. Brewer, and W. D. Schall. 2000. Mutation causing von Willbrand’s disease in Scottish terriers. Journal of Veterinary Internal Medicine 14:10. Whittingham, D. G., S. P. Leibo, and P. Mazur. 1972. Survival of mouse embryos frozen to –196 degrees and –269 degrees C. Science 178:411. Yershov, Y., T. H. Baldini, S. Villagomez, T. Young, M. L. Martin, R. S. Bockman, M. G. Peterson, and R. D. Blank. 2001. Bone strength and related traits in HcB/Dem recombinant congenic mice. Journal of Bone and Mineral Research 16:992.
Appendix 9.1: Genetic Nomenclature Outbred Stocks Official rules for the nomenclature of outbred stocks were published by ICLA (the International Council on Laboratory Animals, now ICLAS, the International Council on Laboratory Animal Science) in 1972 (Festing et al.). They have not been revised since then. Stocks should be closed colonies for at least four generations and should be maintained with less than 1% inbreeding per generation (see later discussion). They should be designated by a code consisting of uppercase letters or letters and numbers, starting with a letter. When a stock is already known by a designation that included other characters, it should retain its existing designation. This code should be preceded by a laboratory code, which is the same as the one used for inbred strains (see later discussion), with the code and strain designation being separated by a colon (e.g., Hsd:WIST, a stock of Wistar rats designated WIST and maintained by Harlan Sprague–Dawley). In practice, the use of such standard designation is voluntary because no international organization currently takes much interest in the genetics and nomenclature of outbred stocks. Inbred Strains The rules governing genetic nomenclature for mice and rats are formulated and maintained by international committees made up of research workers with an interest in the genetics of these two
84550.indb 244
11/11/10 6:25:33 PM
Laboratory Animal Genetics and Genetic Quality Control
245
species. The rules are subject to constant revision as knowledge of genetics increases and as different types of animals become available. For example, transgenic strains were not envisioned at the time the rules were originally formulated. Full details of mouse genetic nomenclature are given on the Jackson Laboratory Web site (www.informatics.jax.org) and of rats and mice on the rat genome database Web site (www.rgd.mcw.edu) Briefly, inbred strains of mice and rats are designated by a code consisting of uppercase letters (e.g., SJL, LEW) or (less preferably) letters and numbers (e.g., C57BL, F344), except for strains already known by some other designation at the time that the rules were formulated. Short symbols are preferred, and duplicate designations must be avoided. Strains do not have names, only designations. Many people using rats seem to want to name them, sometimes with names that are quite inaccurate. For example, the DA strain of rats is sometimes called the “dark agouti” strain, though it is not dark, and the “D” in the designation actually stands for the D blood group (now an obsolete nomenclature), not for the word “dark”. A name is confusing because it does not conveniently fit with the rest of the genetic nomenclature, particularly with the need to designate substrains. Fortunately, users of inbred mice seem to accept designation codes rather than names. When necessary, the level of inbreeding can be designated by an “F” followed by the number of generations of full sib mating—for example, (F87), or (F?+50) in the case of a strain with unknown inbreeding followed by 50 known generations of sib mating. Substrains can arise when branches of an inbred strain have been separated after 20 generations, but before 40 generations of inbreeding, or when branches have been maintained separately for 100 or more generations from their common ancestor. In both cases, some divergence is expected. Substrains are also formed when genetic differences between branches have been found. Substrain is indicated by a slash and a number, a laboratory registration code, or a combination of the two. For example, FL/1Re and FL/2Re are two substrains of strain FL, both established in the laboratory with code Re (see later discussion). Some exceptions, such as the “c” in BALB/c, have been permitted for strains already known using a different designation. Laboratory registration codes are administered by the Institute of Laboratory Animal Research (ILAR) and codes can be registered directly on its Web site (http://dels.nas.edu/ilar_n/ilarhome/). They are used, as in the preceding, to indicate substrain differences and to remind people that genetically identical animals raised in different environments, such as found in different laboratories, may differ phenotypically due to nongenetic causes. Coisogenic, Congenic, and Segregating Inbred Strains Coisogenic strains, in which a mutation arose, was backcrossed, or was engineered onto an inbred background, have the designation of the background strain, followed by a hyphen, and then the symbol of the differential allele, shown in italics (the nomenclature of genes is discussed later—for example, C57BL/6J-Lepob . Congenic strains are produced by repeated backcrosses to an inbred strain, with selection for a particular marker from the donor strain. Congenic nomenclature is denoted using the background strain designation (which is often abbreviated), a period, the donor strain designation, a dash, and the allele designation (e.g., B10.129-H12b). B10 is an abbreviation for C57BL/10, and a full list of such abbreviations is given in the mouse genome and the rat genome databases. If the donor strain is not inbred or the genetic difference is complex, the symbol “Cg” should be used to denote the donor strain—for example, B6.Cg-PhexHyp/J, the hypophosphatemia mutation that arose in a C57BL/6J stock bearing the quaking mutation. The quaking mutation arose in the DBA/2J strain and was crossed twice to the C3H strain before being crossed onto C57BL/6J. The complexity of the donor strain warrants the Cg designation. The use of Cg indicates that alleles in
84550.indb 245
11/11/10 6:25:33 PM
246
Handbook of Laboratory Animal Science
the strain name came from more than one source. Parentheses may also be used to show that an inbred or congenic strain may have a minor contribution from one other strain—for example, C.129P(B6)-Il2tm1Hor, a targeted mutation created in a 129 ES cell line and transferred from a B6;129P mixed background to BALB/c. Note that, at present, gene symbols are changing quite frequently, often to reflect gene function. Also, as the genes responsible for mutant phenotypes are identified, the designation of the coisogenic and congenic strains also needs to change. Recombinant Inbred Strains RI strains are known by an abbreviation of the two parental strains, with the female parent given first, separated by an uppercase X, followed by numbers for the individual strains. For example, AXB2 is a strain derived from a cross between a strain A female and a strain C57BL/6 male, identified as strain 2 of this set. If the male parental strain already ends in a number, a hyphen may be used to distinguish the individual strains. Mixed Inbred Lines Inbred strains that are derived from only two parental strains can be designated using uppercase abbreviations for the two strains, separated by a semicolon (e.g., B6;D2-a Es1e/J). The order in which the strains are listed has meaning. The convention followed is the same as that used when an F1 hybrid designation is constructed; the abbreviation of the strain from which the female originated in the first cross precedes the semicolon and the abbreviation from which the male originated follows the semicolon. In the case of strains derived from ES cells, the designation preceding the semicolon should be the host and the strain following the semicolon the donor—for example, B6;129S4-Nos1tm1Plh /J, where ES cells from 129S4 were used in the gene targeting and then crossed to B6. Because the semicolon designations may be used for a mixed stock before it is fully inbred, these stocks should not be assumed to be inbred unless accompanied by an inbreeding generation number (e.g., >F20). Recombinant Congenic Strains Recombinant congenic strains are formed by crossing two inbred strains, followed by a few (usually two) backcrosses of the hybrids to one of the parental strains, the recipient strain. The individual RC strains are designated by uppercase abbreviations of the strain names, with the recipient (i.e., background) strain designated first, separated by a lowercase “c” with numbers to indicate the individual lines. For example, one of the CcS set of RC strains with BALB/c as the recipient strain and STS as the donor strain could be designated CcS6 if it was the sixth of such a set of strains, where C is the single letter designation for BALB/c Consomic or Chromosome Substitution Strains Consomic strains are produced by repeated backcrossing of a whole chromosome onto an inbred strain for at least 10 generations. Consomic strains of mice are officially designated by RECIPIENT STRAIN-CHROMOSOMEDONOR STRAIN. For example, C57BL/6J-YAKR is a consomic strain with the C57BL/6 autosomes, but the Y chromosome from strain AKR. Similarly, C57BL/6J-Chr 6A/J/NaJ is a consomic strain where the genome is C57BL/6J except for chromosome 6, which comes from the donor strain A/J.
84550.indb 246
11/11/10 6:25:33 PM
Laboratory Animal Genetics and Genetic Quality Control
247
Mutants, Polymorphisms, and Genetically Modified Loci The nomenclature of mutants is reasonably straightforward. However, there are some complications because gene symbols are changing frequently as a result of genetic advances. Thus, mutants such as obese were given a symbol (in this case, ob) that indicated both the allele and the locus. The wild type at the ob locus was then designated +ob or ob+, and when the context was clear, just +. However, when the gene was mapped and cloned, it was found to code for a protein (leptin), which was given the locus symbol Lep, so the obese allele has now been renamed Lepob. Now the wild type is designated Lep+. Many mutants have undergone such changes in their designations, with old symbols such as c for the albino locus now being redesignated Tyrc . The nude mutation has been redesignated Foxn1nu because it is a mutation at a locus first described in Drosophila that causes a forked head in that insect. As far as possible, the same symbols are used for orthologous genes in mice, rats, and humans. Details of the gene nomenclature rules for mice are given on the Jackson Laboratory Web site (www.informatics.jax.org). Briefly, names for genes, loci, and alleles should be brief and, if possible, descriptive (e.g., “obese” or “congenital hydrocephalus”). Genes are functional units, whereas a locus can be any distinct DNA sequence. Symbols for genes should be short abbreviations of one to four letters, starting with the same letter as the name; the first letter of the symbol is capitalized and the remaining letters are lowercase (note that gene symbols in human genes consist of all capital letters). Arabic numbers can be included as part of the name when necessary, but the first symbol should always be a letter. Roman numbers and Greek letters should not be used. Hyphens are used only for clarity, such as when two numbers need to be separated. In published articles, gene symbols are given in italics. Loci defined by anonymous DNA probes are given a symbol starting with a D followed by the chromosome assignment (i.e., the numbers 1–19; X or Y), a laboratory registration code (see preceding for nomenclature of inbred mouse strains), and a unique serial number. The laboratory methods for detection of the locus also need to be specified. Alleles are usually designated by a superscript. When this is not possible (e.g., when superscripts are not accepted electronically), the symbol can be enclosed in chevrons (e.g., Gpi1a or Gpi1
). There are a range of additional rules relating to things like pseudogenes, supergene families or complexes, retroviruses, and special classes of genes and gene complexes, such as biochemical variants, lymphocyte antigens, histocompatibility loci, etc. Genetically Modified Loci The nomenclature of genetically modified alleles has the potential to evolve rapidly as different genetic engineering techniques become standard practice. A balance must be struck in grouping alleles to one generalized nomenclature standard and breaking out those alleles that warrant a new nomenclature designation. Although nomenclature should be informative, it is not designed to replace primary literature or databases for source information on the techniques used to design an allele. Transgenes Transgenic constructs that are introduced into the genome through pronuclear injection are designated by a general formula: Tg(xxx-yyy)###/ZZZ, where Tg denotes the transgene, xxx the promoter, yyy the gene being driven by the promoter, ### the founder line, and ZZZ the ILAR lab code. An example of transgene nomenclature as part of the overall strain designation follows. In this example, the transgenic construct contains the mouse prion protein promoter (Prnp) driving the
84550.indb 247
11/11/10 6:25:34 PM
248
Handbook of Laboratory Animal Science
human alpha-synuclein gene (SNCA). The A53T mutation designation preceded by the asterisks indicates that this is a mutant allele and is helpful in distinguishing it from the wild-type transgenes that are often simultaneously made as controls for disease models. 0XWDWLRQ
3URPRWHU
,/$5/DEFRGH
B6;C3-Tg(Prnp-SNCA*A53T)83Vle/J Genetic background
Founder line
Holder, Jackson Lab
Human synuclein gene
Targeted Mutations Mutations that are the result of gene targeting by homologous recombination in ES cells are given the symbol of the targeted gene, with a superscript consisting of three parts: the symbol tm to denote a targeted mutation, a serial number from the laboratory of origin, and the laboratory code where the mutation was engineered. For example, Cftrtm1Unc is the first targeted mutation of the cystic fibrosis transmembrane regulator (Cftr) gene produced at the University of North Carolina. Early in the use of homologous recombination, targeted mutagenesis was used consistently to generate complete knockouts of genes, usually by insertion of a neomycin cassette in early coding regions. As a result, the nomenclature designation of “tm” became associated with knockouts and complete loss of function alleles. At present, targeted mutagenesis is used to manipulate the genome in many ways; thus, the user is advised to consult databases and primary literature for complete information. In some cases, variations of the tm nomenclature have been adopted: Mutations where all or part of the coding region of one gene is replaced by another are still given a tm symbol and the details of the knockin are associated with the name in publications or databases. Exceptions occur when there has been a replacement of the complete coding region; the replacing gene symbol can be used parenthetically as part of the allele symbol of the replaced gene along with a laboratory code and serial number (e.g., En1tm1(Otx2)Wrst , where the coding region of En1 was replaced by the Otx2 gene, originating from the Wurst laboratory). There is currently no official designation to distinguish targeted mutations that contain floxed alleles or reporter alleles such as lacZ and fluorescent proteins. This can be especially problematic at loci such as Gt(rosa)26, which is frequently targeted because the promoter allows for ubiquitous expression of an inserted sequence but the disruption of the Gt(rosa)26 locus itself has no detrimental effects. However, there is nomenclature for genetically modified mice to include special designations for gene trapped alles and ENU mutagenesis, as well as RNAi and transposable elements. More details are given on the Jackson Lab Web site. Appendix 9.2: Web Resources for Laboratory Animal Genetics Mouse Genome Informatics (MGI) (http://www.informatics.jax.org/): MGI is the most comprehensive bioinformatics tool available to the mouse community, providing integrated genetic, genomic, and biological data to facilitate the study of human health and disease. MGI allows for both simple and complex queries; its organization at the highest level supports data on genes, phenotypes, gene expression, gene function, biological pathways, strain, and SNP data, as well as information orthology and mouse tumors.
84550.indb 248
11/11/10 6:25:36 PM
Laboratory Animal Genetics and Genetic Quality Control
249
International Mouse Strain Resource (IMSR) (http://www.informatics.jax.org/imsr/index.jsp): The IMSR is a searchable online database of mouse strains and stocks available worldwide, including inbred, mutant, and genetically engineered mice. The goal of the IMSR is to assist the international scientific community in locating and obtaining these resources. In addition to live mice, the IMSR also provides information on the availability of cryopreserved material and ES cell lines. The IMSR serves as the catalogue for participating repositories and consortiums generating mice and mouse-derived resources. This database is supported within MGI. The mouse phenome database (MPD) http://phenome.jax.org/pub-cgi/phenome/mpdcgi?rtn= docs/home): The goal of the MPD is to establish a collection of phenotypic baseline data on commonly used and genetically diverse inbred mouse strains. All the data are collected and disseminated from MPD. About 1,000 measurements for phenotypes, including those relevant to atherosclerosis, blood disorders, cancer susceptibility, neurological and behavioral disorders, sensory function defects, hypertension, osteoporosis, and obesity, have been acquired for many strains. The MPD also contains extensive genotypic data, which allow for genotype–phenotype association predictions and facilitate efforts to identify and determine the function of genes participating in normal and disease pathways. The rat genome database (RGD) (http://rgd.mcw.edu/): RGD serves a similar function to MGI, but for rats. It is the most comprehensive bioinformatics tool available to the rat community, providing integrated genetic, genomic, and biological data to facilitate research. RGD allows for both simple and complex queries; its organization at the highest level supports data on genes, gene function, biological pathways, strains, quantitative trait loci, phenotypes, and annotation of the rat genome. Online Mendelian Inheritance in Man (OMIM) (http://www.ncbi.nlm.nih.gov/sites/entrez?db= omim): Pioneered in the early 1960s by the late Dr. Victor McKusick, OMIM provides a comprehensive compendium of human genes and genetic phenotypes. The full-text, referenced overviews in OMIM contain information on all known Mendelian disorders and focus on the relationship between phenotype and genotype. OMIM incorporates the use of genetic animal models, its disease and gene annotations, and provides a time-lined account of detailed information. Institute for Laboratory Animal Research (ILAR) (http://dels.nas.edu/ilar_n/ilarhome/index. shtml): ILAR provides information on issues related to the scientific, technological, and ethical use of animals and related biological resources in research, testing, and education. Colony management software: JCMS (http://colonymanagement.jax.org/overview.html): This is downloadable software that allows for tracking of animal status, tracking of animal pedigree, genotype logging, creating mating records, creating litter records, animal pen management, experimental data tracking, cage card printing, generating various reports, bulk data entry, advanced database queries, comprehensive mouse sample tracking, and support for handheld devices. “MICE” (www.biomedcentral.com/1471-2156/2/4): This is another computer program that is used for automation of breeding records and tracking of individual animals and their genotypes. It is distributed free of charge to academic institutions.
84550.indb 249
11/11/10 6:25:36 PM
84550.indb 250
11/11/10 6:25:36 PM
Chapter 10
Health Status and Health Monitoring
Axel Kornerup Hansen Contents Introduction..................................................................................................................................... 252 Specific Infectious Agents in Laboratory Rodents and Rabbits..................................................... 252 Bacterial Infections.................................................................................................................... 253 Pasteurellaceae...................................................................................................................... 254 Clostridium piliforme............................................................................................................ 255 Helicobacter spp.................................................................................................................... 256 Bordetella bronchiseptica..................................................................................................... 256 Cilia-Associated Respiratory (CAR) Bacillus....................................................................... 256 Corynebacterium spp............................................................................................................ 257 Citrobacter rodentium.......................................................................................................... 257 Salmonellae........................................................................................................................... 257 Streptobacillus moniliformis................................................................................................. 257 Streptococcus spp.................................................................................................................. 258 Staphylococci........................................................................................................................ 258 Pseudomonas spp.................................................................................................................. 259 Mycoplasma spp.................................................................................................................... 259 Fungal Infections....................................................................................................................... 259 Dermatophytes...................................................................................................................... 259 Pneumocystis carinii............................................................................................................. 259 Viral Infections..........................................................................................................................260 DNA Viruses......................................................................................................................... 262 RNA Viruses.........................................................................................................................264 Parasitological Infestations................................................................................................... 268 Microbial Interference with Animal Experiments......................................................................... 272 Pathological Changes, Clinical Disease, and Mortality............................................................ 272 Contamination of Biological Products....................................................................................... 273 Immunomodulation.................................................................................................................... 273 Physiological Modulation........................................................................................................... 274 Interference with Reproduction................................................................................................. 274 Competition between Microorganisms within the Animal....................................................... 274 Modulation of Oncogenesis....................................................................................................... 274 The Impact of the Gut Microbiota of Laboratory Animals....................................................... 275 251
84550.indb 251
11/11/10 6:25:36 PM
252
Handbook of Laboratory Animal Science
Eradication Methods....................................................................................................................... 275 Caesarian Section....................................................................................................................... 275 Embryo Transfer........................................................................................................................ 278 Cessation of Breeding and Burnout........................................................................................... 279 Stamping Out............................................................................................................................. 279 Antibiotic Treatment.................................................................................................................. 279 Vaccination.................................................................................................................................280 Containment Facilities....................................................................................................................280 Barrier Housing..........................................................................................................................280 Cubicles and Filter Cabinets...................................................................................................... 282 Filter-Top Cages and Individually Ventilated Cage (IVC) Systems..........................................284 Isolators......................................................................................................................................284 Health Monitoring.......................................................................................................................... 286 Scope..........................................................................................................................................286 Sampling Strategies................................................................................................................... 287 Choice of Method....................................................................................................................... 287 Reporting.................................................................................................................................... 289 Sentinels..................................................................................................................................... 291 Characterizing the Normal Gut Microbiota............................................................................... 291 Screening of Biological Materials.............................................................................................. 292 Quarantine Housing of Animals Imported from Nonvendor Sources........................................... 292 Final Remarks................................................................................................................................. 293 References....................................................................................................................................... 294 Introduction Health status may be defined as the actual status of an individual animal related to its clinical, pathological, and physiological appearance. In more popular terms, health status tells whether the animal is ill or not. However, since no animal can be said to be either only ill or only healthy, health status can be regarded more quantitatively than qualitatively. Infections, the environment, and genetic disorders may all reduce the health of animals and counteract the aim of achieving reproducible results with low variability in groups of animals. Infections can pose a potential risk for irreversible reductions in health in a large number of animals and, consequently, have to be dealt with on a daily basis. Therefore, many laboratory animal scientists automatically think of infections when discussing the term “health,” and laboratory animal health is often considered synonymous with laboratory animal microbiology. Environment and genetics are obviously equally important for research outcome and animal welfare, but these factors may be accounted for in the design of facilities, the optimal running of animal houses, and breeding procedures—creating a set of management approaches that differ from the simple monitoring of microbiology and/or infections. The general influences of environment and genetics are discussed in considerable detail elsewhere in this handbook. In the present chapter, first some information on how infections interfere with animal research will be given to allow the reader an understanding on why this is important to consider. In the last parts of the chapter, some guidance on how to avoid such infections becoming a problem will be given. Specific Infectious Agents in Laboratory Rodents and Rabbits A specific infection may be defined as an infection caused by one specific microbial organism. Implicit in the term “infection” is that this may impose a risk of certain pathophysiological changes, which may only eventually be present in the infected animal. If there is no such risk, the term
84550.indb 252
11/11/10 6:25:36 PM
Health Status and Health Monitoring
253
“association” is used instead. Infections are equally important, whether they occur in experimental mice or experimental pigs. However, space restrictions for this chapter will not allow us to cover all animal species used in research. Therefore, a short introduction to the infections of the most commonly used laboratory animals (i.e., rodents and rabbits) is provided. Information concerning other laboratory animals can be found in textbooks dealing specifically with these species (Fox et al. 2002). Bacterial Infections Specific bacterial infections may cause disease, as well as other negative consequences for research. Laboratory animals may be kept free of single species of specified bacteria, but as far as they are not housed in isolators they always harbor a normal flora that, by itself and as later described, may also have an impact on research (Hansen et al. 2009). Specific bacteria of rodents and rabbits are listed in Table€10.1. Table€10.1â•…Important Bacterial and Fungal Infections Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) Gram-Negative Bacteria Bordetella bronchiseptica Campylobacter coli/jejuni CAR bacillus Citrobacter freundii Citrobacter rodentium Escherichia coli Francisella tularensis Fusobacterium necrophorum Haemophilus spp. Helicobacter bilis Helicobacter cholecystus Helicobacter cinnaedi Helicobacter hepaticus Helicobacter muridarum Helicobacter rappini Helicobacter rodentium Helicobacter trogontum Klebsiella pneumoniae Leptospira spp. Pasteurella pneumotropica Pasteurella multocida Pseudomonas aeruginosa Salmonella spp. Spirillum minus Streptobacillus moniliformis Treponema paraluis-cuniculi Yersinia pseudotuberculosis
M, R, GP, H, RB M, R, H, RB M, R, RB GP M M, R, GP, H, RB R, RB M, GP, RB M, R, GP, H, RB M H H M M, R M M R M, R, GP, H, RB M, R M, R, GP, H, RB M, R, GP, H, RB M, R, GP, H, RB M, R, GP, H, RB R M, R, GP RB M, R, GP, H, RB
Gram-Positive Bacteria Clostridium perfringens Clostridium difficile
M, RB GP, H, RB (continued)
84550.indb 253
11/11/10 6:25:37 PM
254
Handbook of Laboratory Animal Science
Table€10.1╅Important Bacterial and Fungal Infections Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) (continued) Gram-Positive Bacteria Clostridium piliforme Clostridium spiroforme Corynebacterium kutscheri Erysipelothrix rhusiopathiae Listeria monocytogenes Staphylococcus aureus Streptococcus group A/B/D/G Streptococcus group C Streptococcus pneumoniae
M, R, H, RB RB M, R, GP, H R M, R, GP, H, RB M, R, GP, H, RB M, R, GP, H, RB M, R, GP, H, RB M, R, GP, H, RB
Chlamydiae and Mycoplasmae Chlamydia psitacci Mycoplasma pulmonis Mycoplasma neurolyticum Mycoplasma arthritidis Mycoplasma caviae Mycoplasma cricetuli Mycoplasma collis Mycoplasma muris
GP M, R M M, R GP H M M Fungi
Aspergillus spp. Candida albicans Cryptococcus neoformans Microsporum canis Pneumocystis carinii Trichophyton mentagrophytes
R M, GP M, GP, H GP, RB M, R, GP, H, RB M, R, GP, RB
Note: Only infections that may influence research in some way are mentioned, and it should be kept in mind that animals that are not gnotobiotic harbor a great number of other bacterial species not mentioned here.
Pasteurellaceae Pasteurella pneumotropica is an important rodent bacterium. It is classified within Pasteurellaceae, but the full definition is unclear and the genus itself is heterogenous (Hayashimoto et al. 2005, 2007; Sasaki et al. 2006a, 2006b). Other Pasteurellaceae, such as Actinobacillus muris and Haemophilus influenzae murium, are likely to be of equal importance. However, DNA-based techniques have recently revealed a need for a revised classification of these agents (Christensen and Bisgaard 2010). Most conventional rodent colonies are infected, but barrier-bred colonies of rats and mice may also latently harbor this agent (Pritchett-Corning et al. 2009). Carrier prevalences in infected rodent colonies vary from just a few percent up to as much as 95% (Nakagawa et al. 1984; Hansen 1992). Infection with this agent may lead to either upper respiratory disease or pyogenic infections, such as subcutaneous abscesses or mastitis (Baker 1998). Generally, however, P. pneumotropica is a secondary pathogen in relation to a primary agent, such as Mycoplasma pulmonis or Sendai virus. Stress, including experimental stress, or immunosuppression may activate latent infections. The incidence of spontaneous deaths during inhalation anesthesia might be raised in infected animals
84550.indb 254
11/11/10 6:25:37 PM
Health Status and Health Monitoring
255
(Hansen 1999). Transmission is mainly horizontal by droplets, but newborn pups may become infected during gestational passage through the contaminated vagina. Pasteurella multocida is a facultative pathogen of rabbits (Loliger and Matthes 1989). In conventional colonies, a high number of animals may be infected (Donnio et al. 1994). Infection is mostly subclinical, and epizootic disease is connected with environmental and host-related factors. Respiratory disease occurs as “snuffles,” which may develop into conjunctivitis, abscessation, and acute septicemias, as well as acute or chronic pneumonia. Pasteurella multocida infection is mostly observed during spring and fall. Direct contact is considered the chief mode of transmission. Suckling rabbits may be infected with P. multocida from carrier does within the first week of life. The infection does not seem to spread between rabbits that are not in close contact with one another (Deeb and DiGiacomo 2000; Dillehay et al. 1991; DiGiacomo et al. 1987, 1990, 1991a, 1991b); therefore, a barrier system is an efficient way of keeping rabbits free of this infection (Scharf et al. 1981). Transmission from other species (e.g., pigs and cattle) may occur (al-Lebban et al. 1988). Clostridium piliforme Clostridium piliforme (formerly known as Bacillus piliformis) is the causative agent of Tyzzer’s disease. In mice (Tyzzer 1917), hamsters (Zook et al. 1977a, 1977b), gerbils (White and Waldron 1969), and rabbits (Allen et al. 1981), this is a fatal disease characterized by multiple focal necrosis of the liver (Figure€10.1). Long, slender bacteria are found in the cytoplasm of the hepatocytes at the periphery of the necrotic foci. These bacteria are also found in huge numbers in the alimentary tract, especially in the ileum and caecum, and especially in association with ileitis, caecitis, and colitis. It has long been known that different mutants infect different animal species (Fujiwara et al. 1983, 1985) and that infection across species is therefore a rare event. One notable exception is that Mongolian gerbils, under certain circumstances, may be sensitive to infection from rats, mice, and rabbits (Hansen 1990). In rats, Tyzzer’s disease is a mild disease of weanlings connected with megaloileitis (Figure€10.2), multiple focal necrosis of the liver, and single necroses in the myocardium (Hansen et al. 1994). Resistance to development of Tyzzer’s disease may be due to genetic traits (Hansen et al. 1990, 1992a). The organisms may persist in the intestinal epithelium of healthy animals. The prevalence of infected individuals in rat and mouse colonies varies, but it is often more than 50% (Hansen et al. 1992b). The agent may cross the placenta (Friis 1978, 1979). Infection with C. piliforme is a common finding in rats; it seems to occur less frequently in other species (Schoondermark-van de Ven et al. 2006).
Figure 10.1â•…Tyzzer’s disease in a rabbit. Note the white spots on the liver.
84550.indb 255
11/11/10 6:25:37 PM
256
Handbook of Laboratory Animal Science
Figure 10.2â•…Megaloileitis in a postweaned SPRD rat observed as a spiral on the right abdominal wall.
Helicobacter spp. Helicobacter spp. have been isolated from nearly all species of rodents (Fox and Lee 1997), but clinical significance has only been documented in a few of these. Helicobacter hepaticus and perhaps H. bilis cause chronic hepatitis in mice (Fox and Lee 1997). Helicobacter hepaticus may also be responsible for liver tumors in mice (Fox and Lee 1997). Susceptibility to disease seems to be genetically dependent; for example, A/JCr mice seem to be highly susceptible, while C57BL/6 mice seem to be resistant (Ward et al. 1994). Helicobacter cholecystus may be involved in hepatitis and pancreatitis in hamsters (Zenner 1999). Infection with Helicobacter spp. is likely to be common in rodent colonies (Fox and Lee 1997; Jacoby and Lindsey 1998). Bordetella bronchiseptica Bordetella bronchiseptica may be frequently isolated from rabbits and guinea pigs, occasionally isolated from rats, and seldom isolated from mice, hamsters, and gerbils (Hansen 1999). It may cause pneumonia, pleuritis, and pericarditis in guinea pigs, which can be fatal, especially if the animal harbors other respiratory pathogens (Hansen 1999). In rabbits, disease is mainly subclinical and characterized by focal chronic interstitial pneumonia (Uzal et al. 1989). Cilia-Associated Respiratory (CAR) Bacillus CAR bacillus has been reported in mice, rats and rabbits, but rat and mouse isolates differ from those of rabbits, and should be regarded as different bacteria. The rat and mouse version of CAR bacillus appears to be closely related to Flavobacterium (Wei et al. 1995), while the rabbit version shows a greater similarity with Helicobacter (Cundiff et al. 1995). Infected rats and mice are usually asymptomatic (Shoji-Darkye et al. 1991; Shoji et al. 1988), but CAR bacillus may be the cause of a highly contagious epizootic, a slowly progressive and uncontrollable disease referred to as chronic respiratory disease (CRD). CRD is characterized by weight loss, rough hair coat, wheezing, mucopurulent exudate, and severe peribronchial lymphoid cuffing (Itoh et al. 1987). In rabbits, no clinical signs of respiratory disease have been observed, although histopathological examination of the respiratory tree may reveal mild hyperplasia of lymphoid nodules of the respiratory mucosa with scattered bacilli in the lower respiratory system (Caniatti et al. 1998). The infection is normally not transmitted to sentinels by the dirty bedding technique (Nicklas et al. 2002).
84550.indb 256
11/11/10 6:25:38 PM
Health Status and Health Monitoring
257
Corynebacterium spp. Corynebacterium kutscheri is the cause of pseudotuberculosis in rats and mice. Although the organism has also been isolated from guinea pigs (Vallee et al. 1969) and hamsters (Amano et al. 1991), infection is normally observed only in rats and mice (Amao et al. 1995a, 1995b). In immunocompetent rats and mice, the infection may be subclinical (Amao et al. 1995a, 1995b), but it may also cause abscessation in the superficial tissues and pulmonary emboli. Embolization in the mouse affects joints, liver, and kidney (Weisbroth and Scher 1968) and is referred to as pseudotuberculosis. Genetic factors seem to be involved in the susceptibility of both rats (Suzuki et al. 1988) and mice (Hirst and Wallace 1976) and, therefore, mortality varies between infected strains. Modes of excretion and spread of the agent are not fully known, but urine and feces from infected animals are likely to be contaminated. Transplacental infection has been demonstrated experimentally (Juhr and Horn 1975). The prevalence of C. kutscheri within a colony may be less than 5%. Corynebacterium kutscheri has previously been found worldwide, but today is uncommon in laboratory animals bred and kept in modern facilities, and other Corynebacterium spp. may be more common problems; for example, Corynebacterium renale may cause urinary calculus in young rats (Osanai et al. 1996; Takahashi et al. 1995), and C. bovis may be isolated from nude mice with scaly and crusty skin, often involving more than 80% of the animals (Gobbi et al. 1999a, 1999b). Citrobacter rodentium Citrobacter rodentium (Schauer et al. 1995) (formerly known as C. freundii type 4280; Barthold et al. 1976) may be the cause of rectal prolapses, diarrhea, and dehydration in suckling and postweaned male mice. Feeding and genetic factors influence both morbidity and mortality (Barthold et al. 1977), but mortality and prevalences are normally low. Transgenic and some nontransgenic mice of all ages may be affected by chronic debilitation, loss of reproductive efficiency, rectal prolapses, and sudden death (Maggio-Price et al. 1998). Also, alterations in immunological parameters may be observed, including outgrowth of an unusual population of cells in the spleen and blood, reduction in ascites production, loss of the capacity of peritoneal exudate cells to serve as feeders for the cloning of long-term T-cell lines, and inhibition of antigen-specific cytotoxic T-cell activity (Maggio-Price et al. 1998). Antibiotic therapy may significantly reduce morbidity and mortality, increase litter size and frequency, and result in the normalization of many of the immunological assays (Maggio-Price et al. 1998). Salmonellae Salmonellae infect all species of warm-blooded animals, and until the introduction of barrier-protected breeding systems, it commonly ruined research projects involving rodents, especially mice. The prevalence of this organism has significantly diminished over the last 40 years. Salmonella typhimurium (in mice and rats) and S. enteritidis (in guinea pigs) are the most common types, causing various grades of diarrhea. Prevalence in infected colonies mostly ranges above 50%, but may vary significantly (Nakagawa et al. 1986). Uterine infections, probably without placenta barrier passage, have been described for Salmonellae (Okewole et al. 1989). Streptobacillus moniliformis Streptobacillus moniliformis may be isolated from mice (Wullenweber et al. 1990a), rats (Wullenweber et al. 1992; Koopman et al. 1991), and guinea pigs (Kirchner et al. 1992). It is transmissible to humans, which in rare cases may cause rat bite fever, a purulent wound infection developing
84550.indb 257
11/11/10 6:25:38 PM
258
Handbook of Laboratory Animal Science
into endocarditis, petechial exanthema, polyarthritis, fever, and death. Recent cases have mostly been related to pets (Cunningham et al. 1998; Peel 1993; Rygg and Bruun 1992) and wild animals (Mathiasen 1993), and it has also occurred in patients infected with human immunodeficiency virus (Rordorf et al. 2000). This is primarily a disease of mice that begins as swelling of the cervical lymph nodes and may then develop into fatal septicemia. Chronic cases are characterized by arthritis in the distal portions of the legs and the tail. Abscessation and abortions may occur (Kaspareit-Rittinghausen et al. 1990). Genetic factors seem to be important for susceptibility to infection and disease; C57BL/6 mice seem to be highly susceptible (Wullenweber et al. 1990b). In other species, symptoms are different. For example, in guinea pigs, the agent causes local abscesses that do not spread (Fleming 1976); in rats, clinical signs are uncommon, but cases of otitis media may occur (Wullenweber et al. 1992; Koopman et al. 1991). Streptococcus spp. Streptococcal infections are nonclinical and may spread between humans and animals, mostly by droplet infection through the intranasal route. According to the FELASA guidelines for health monitoring (Nicklas et al. 2002), β-hemolytic streptococci and Streptococcus pneumoniae are the only ones to be reported, although their importance is probably questionable and other types may be pathogenic as well. Prevalences within infected barrier-bred colonies are generally around 10% (Hansen 1992). Group C in guinea pigs, as well as groups G and A in rats and mice, may be the cause of various pyogenic processes. Streptococcus pneumoniae may be found in guinea pigs, is seen more seldom in rats and rabbits, and is rare in mice. The prevalence within infected colonies may vary from 15 to 55%. Disease is mostly related to stress (e.g., due to a poor environment or nutritional deficiencies). In rats, a mucopurulent discharge from the nose may initially be observed; later, the disease may progress into noisy, abdominal respiration. Pathological changes are dominated by fibrin with various grades of focal bronchopneumonia developing into lobar fibrinous pneumonia. Unexpected deaths are often the only visible signs of infection in guinea pigs, while in rabbits, dyspnoea and depression often quickly turn into septicemia (Hansen 1999). Staphylococci Staphylococci are found worldwide in all species of animals, and spread across species lines, including animal to human and vice versa, should be expected. The majority of humans and animals are carriers of staphylococci. Staphylococcus aureus is found with a high prevalence in most colonies of laboratory rodents (Hansen 1992), as well as in most humans. In wild mice, however, S.€aureus is uncommon. Some commercial breeders are now capable of producing rodents free of S.€aureus. Other types of Staphylococci common in laboratory rats and mice include S. haemolyticus, S. xylosus, S. sciuri, and S. cohnii (Vogelbacher and Bohnet 1997). The bacteria may be transmitted among hosts in various direct or indirect ways, including passive carriage by animal technicians. Staphylococcal disease in immunocompetent animals is mainly secondary (e.g., due to trauma, stress, or the equivalent). It is characterized by pyogenic processes, such as abscesses at bite or surgical wounds; pneumonia in rodents kept in poorly ventilated units; and dermatitis in gerbils kept in bedding that is too moist (Peckham et al. 1974). In immunodeficient animals, such as nude mice, S. aureus may be a primary disease-causing agent, resulting in multiple abscessation. Coagulase negative staphylococci may also cause disease in laboratory animals; for example, S. xylosus is known to cause intestinal disease in mice (Rozengurt and Sanchez 1994), dermatitis in gerbils (Solomon et al. 1990), and pneumonia in immune-suppressed rats (Detmer et€al. 1991). Negative consequences for research are mainly due to the activation of
84550.indb 258
11/11/10 6:25:38 PM
Health Status and Health Monitoring
259
latent infection by stress or immunosuppression and the presence of abscesses in immunodeficient animals (typically nude mice). Pseudomonas spp. Pseudomonas spp. may be isolated from the respiratory, digestive, and genital systems of rats and mice; the most common species are P. aeruginosa, and P. diminuta. Pseudomonas aeruginosa causes conjunctivitis and rhinitis and, under more severe or experimental conditions, pneumonia and septicemia in rats and guinea pigs (Hansen 1999). In animals with septicemia, abscessation of the liver, spleen, kidney, and middle ear may be observed. Disease due to P. aeruginosa is mainly observed in immunodeficient, immunosuppressed, or stressed animals (Urano and Maejima 1978) and, in general, is secondary to something else. Prevalence in infected colonies of immunocompetent animals kept in a high-quality environment seldom reaches more than 5–10%, but the prevalence of diseased animals in colonies of immunodeficient animals kept under poor environmental conditions may reach 100% (e.g., during ventilation breakdowns). Poor hygienic conditions, especially in relation to the water used for drinking and cleaning, may play an important role in the spread of Pseudomonas spp. Pseudomonas fluorescens and some other Pseudomonas spp. can produce mucus in drinking nipples (Hansen 1999); however, this has not been demonstrated to have any impact on the animals. This condition is normally prevented by acidification of the drinking water with hydrochloric or citric acid. Mycoplasma spp. Monoinfection with Mycoplasma pulmonis in rats causes mild symptoms. However, when complicated with other infectious agents, such as Pasteurella pneumotropica (Brennan et al. 1969) or various viruses (Schoeb et al. 1985), as well as environmental inducers such as raised ammonia levels (Broderson et al. 1976b), disease symptoms (such as snuffles, ruffled hair coat, bronchopneumonia, and arthritis, mostly in a mild form) occur. Additionally, it colonizes the genitals of both males and females and, at least in the latter, may affect reproduction (Cassell et al. 1981a, 1981b). Even in the absence of clinical symptoms, M. pulmonis may raise the incidence of respiratory tract tumors (Kimbrough and Gaines 1966), decrease cellular and humoral immune response (Lai et al. 1989), decrease the severity of adjuvant arthritis (Taurog et al. 1984), and reduce the incidence of diabetes mellitus in the inbred BB rats prone to type 1 diabetes (Voot et al. 1988). The infection is far less common in mice, but symptoms are similar. Other Mycoplasma spp. infect rabbits and guinea pigs. Fungal Infections Dermatophytes Microsporum and Trichophyton spp. may be isolated from guinea pigs and rabbits in rare cases. Clinical disease, known as ringworm or dermatomycosis, is rare, but the zoonotic potential of this infection should be kept in mind. In humans, Trichophyton forms a ring of scaly dermatitis, while Microsporum in humans is characterized by multiple erysipelas. Pneumocystis carinii Pneumocystis carinii infects a wide range of mammal species, but it is difficult to diagnose and immune-competent animals are asymptomatic. However, in SCID mice and other immune-deficient
84550.indb 259
11/11/10 6:25:38 PM
260
Handbook of Laboratory Animal Science
animals, it may cause fatal pneumonia (Shultz et al. 1989; Walzer et al. 1989). It is easily diagnosed in immune-deficient animals (e.g., by immunofluorescence staining of lung washings). It is quite widespread, even in barrier-protected colonies. Transmission seems to be airborne (Hughes 1982). Viral Infections As a rule of thumb, viruses should not be present in laboratory animals. Although surveys show that the number of virus-infected rodent colonies has been declining over the last four decades, infected rodents are still frequently housed in various parts of the world. Important viral infections in rodents and rabbits are listed in Table€10.2. Coronavirus and parvovirus infections are relatively common, while cardio-, rota- and paramyxoviruses also occur (Schoondermark-van de Ven et al. 2006; Jacoby and Lindsey 1998; Zenner and Regnault 2000).
Table€10.2╅Virus Infections Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) DNA Viruses Adenoviridae Mouse adenovirus Rat adenovirus Guinea pig adenovirus
M R GP
Herpetoviridae Mouse cytomegalovirus Rat cytomegalovirus Guinea pig cytomegalovirus Virus III of rabbits Thymic virus Guinea pig herpes-like virus Guinea pig X-virus
M R GP RB M GP GP
Papovaviridae K virus Mouse polyoma virus Rat polyoma virus Hamster papovavirus Rabbit kidney vacuolating virus Virus of oral papillomatosis Rabbit papilloma virus
M M R H RB RB RB
Parvoviridae Kilham rat virus Toolans H1 virus Minute virus of mice Hamster parvovirus Mouse parvovirus Rat parvovirus
84550.indb 260
R R M H M R
11/11/10 6:25:39 PM
Health Status and Health Monitoring
261
Table€10.2â•…Virus Infections Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) (continued) DNA Viruses Poxviridae Ectromelia virus Mouse papule virus Myxoma virus Shope’s fibroma virus Rabbit pox virus Guinea pig pox-like virus
M M RB RB RB GP RNA Viruses
Arenaviridae Lymphocytic choriomeningitis virus
M, GP, H
Bunyaviridae Hantaan virus
R
Caliciviridae Rabbit hemorrhagic disease virus
RB
Coronaviridae Mouse hepatitis virus Rat coronavirus Sialodacryaodenitis virus Guinea pig coronavirus Rabbit coronavirus
M R R GP RB
Paramyxoviridae Sendai virus Pneumonia virus of mice Guinea pig parainfluenza type 3
M, R, RB M, R, H GP
Picornaviridae Theiler’s mouse encephalomyelitis virus Strain GDVII, FA, DA Strain MHG Guinea pig cardiovirus
M R GP
Reoviridae Reovirus type 3 Mouse rotavirus Rat rotavirus Rabbit rotavirus
M, R, H, GP M R RB
Retroviridae Type A viruses Type B viruses
M (continued)
84550.indb 261
11/11/10 6:25:39 PM
262
Handbook of Laboratory Animal Science
Table€10.2╅Virus Infections Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) (continued) RNA Viruses Mouse mammary tumor virus Type C viruses Leukemia viruses Sarcoma viruses
M M, R, H, GP M, R
Togaviridae Lactate dehydrogenase elevating virus
M
Unclassified Viruses GreGrey lung virus
M, R
DNA Viruses Parvoviruses are the most problematic of the several different types of DNA viruses that may infect rodents and rabbits. Most DNA viruses, other than poxviruses, do not produce overt disease, but they still may adversely impact essential research. In general, DNA viruses cause persistent infection. Parvoviruses The classic rodent parvoviruses for rats are Kilham rat virus (KRV) and Toolan’s H1 virus (H1) (Kilham and Ferm 1961). Minute virus of mice (Crawford 1966) (MVM) is the classic mouse parvovirus. Several antigenic types of parvoviruses are known, but KRV and H1 strains share common antigens and therefore cross-react in solid-phase serological assays. Orphan parvoviruses, a group of rodent parvoviruses distinct from MVM, KRV, and H1, were first identified by the fact that antibodies to known rodent parvoviruses were detected by the immunofluorescence assay (IFA), but not by the hemagglutination inhibition assay (HAI) in commercial breeding colonies of rats and mice (McKisic et al. 1993). Today, orphan parvoviruses have been isolated from mice, rats, and hamsters, and they have further been divided into mouse parvovirus (MPV), rat parvovirus (RPV), and hamster parvovirus (HPV). RPV is assumed to be a variant of KRV (Ueno et al. 1997), while MPV resembles MVM in genome size, replication intermediates, and nonstructural proteins (Ball-Goodrich and Johnson 1994). Cross-infection between species-specific strains does not seem to occur (Ueno et al. 1997). Horizontal transmission by fecal–oral contact is most common. Vertical transmission is reported for some serotypes (Kilham and Margolis 1969). Although it is not normally seen after the infection has balanced in the colony and the female breeders have developed protective immunity, careful washing of the embryos is important during embryo transfer to avoid transmission (Janus et al. 2009). Intrauterine infections may be observed in rare cases. The prevalence among adult animals is normally high—50–80%—but falls, sometimes even to zero, after a period of infection. MVM, H1, MPV, RPV, and HPV are not known to cause any clinical disease, while some KRV serotypes have been reported to cause jaundice and ataxia in rats less than 10 days of age. Parvoviruses require a protein produced by the host cell during the S phase of cell division and therefore only replicate in rapidly dividing cells. In rats infected prior to the fourth day of life, intranuclear inclusions are present in the actively mitotic cells that comprise the external germinal layer of the cerebellum. This leads to necrosis, thereby preventing the normal development of the cerebellum and
84550.indb 262
11/11/10 6:25:39 PM
Health Status and Health Monitoring
263
resulting in granuloprival cerebellar hypoplasia. Hepatitis, with intranuclear inclusions in the hepatocytes, has also been reported. Diagnosis is mostly made by serology (Schoondermark-van de Ven et al. 2006), but polymerase chain reaction (PCR) may be applied as well (Besselsen et al. 2006). The infection rapidly spreads to sentinels through the soiled bedding technique (Besselsen et al. 2008). Seroconversion is strain dependent; for example, C57BL/6J mice respond far better to exposure than FVB/N, NMRI, ICR, and C3H/HeN mice (Janus et al. 2008). Although the situation has improved, parvovirus infections still occur in colonies of both rats and mice (Schoondermark-van de Ven et al. 2006; Besselsen et al. 2006). Results from Japan suggest that parvovirus infection in rats is most often caused by RPV (Ueno et al. 1998). Rabbits may also harbor a parvovirus. Clinical signs in neonatal rabbits consist of anorexia and listlessness, while pathological signs are mostly located in the small intestines (Metcalf et al. 1989; Matsunaga and Chino 1981; Matsunaga et al. 1977). Adenoviruses Host-specific strains of adenoviruses infect a range of species, including mice, rats, guinea pigs, and, in extremely rare cases, rabbits. The virus infects by oral or ocular transport, but close contact is required. In mice, two different substrains, MAD-FL (Larsen and Nathans 1977) and MAD-K87 (Takeuchi and Hashimoto 1976), have been identified. In rats and mice, the infection is mostly clinically silent, although myocarditis, nephritis, adenitis, encephalitis, and mortality have been observed after experimental inoculation of neonatal mice (Heck et al. 1972). In infected guinea pigs, necrotizing broncheoalveolitis is regularly observed, characterized by large basophilic intranuclear inclusion bodies in the desquamated bronchial epithelia (Hansen et al. 2000). Adenoviral disease in guinea pigs has been observed in Europe, the United States, and Canada. In most cases, no distinct clinical signs are observed, but occasionally dyspnoea symptoms are discretely scattered among animals in a room. The diagnosis can be made by serology (Schoondermark-van de Ven et€al. 2006) and/or PCR (Butz et al. 1999). Pox Viruses Ectromelia, or mousepox, is a fatal disease in mice induced by infection with a poxvirus. The disease is most severe in DBA, C3H, and BALB/c mice, while black strains seem to be relatively resistant and may even harbor latent infections (Bhatt and Jacoby 1987a). Prevalence of the overt disease may vary from few to all of the animals in a colony (Werner et al. 1981). The virus infects through skin lesions and, after approximately 10 days of incubation, the infection causes edematous skin erosions and hyperplasias, pathologically characterized by large eosinophilic cytoplasmatic inclusions in the epithelial cells. Extensive necrosis in the lymphatic organs and the liver is also observed. In mice surviving the infection, diagnosis may be made by serology, while diagnosis of acute infection is based upon PCR, immunohistochemistry, and virus isolation (Lipman et al. 2000). Today, the spontaneous disease is seldom present in laboratory mice, although incidences have occurred. Regardless, biological materials of insecure origin should be subjected to intense scrutiny (Lipman et al. 2000; Dick et al. 1996). In rabbits, several poxviruses are known (i.e., myxoma, fibroma, and rabbit poxviruses). Of major concern is the myxoma virus, the cause of myxomatosis. This is a severe disease for laboratory rabbits in which edema of the eyelids is the predominant symptom. It may occur in a peracute version, which kills the rabbits within 1 week, or in an acute version, with rabbits surviving for up to 2 weeks after edema has also developed around the anal, genital, oral, and nasal openings. Lethargy, hemorrhage, and convulsions occur just prior to death. The disease may be diagnosed by symptoms, but in countries in which the infection is under legal control, the diagnosis should be confirmed by virus isolation. Currently, screening may be performed by serology. Outbreaks in laboratory colonies are rare, since insects are the primary vector. It is important to note that the
84550.indb 263
11/11/10 6:25:39 PM
264
Handbook of Laboratory Animal Science
disease is endemic in wild populations of lagomorphs in Europe, North and South America, and Australia, even though it is listed by the World Organization for Animal Health (OIE) as a disease to be kept under control by national law. Herpes Viruses Cytomegaloviruses are also known as “salivary gland viruses.” Several species-specific strains exist; among these are strains infecting mice, rats, and guinea pigs. Megalic cells and nuclear inclusions in the glandular epithelium of the salivary glands characterize infection. Infection is rare in laboratory colonies. Mouse thymic virus infects only mice—primarily the salivary glands, but also the thymus. Infection with mouse thymic virus is occasionally reported in laboratory mouse colonies, in which it may reach high prevalences. Acute herpes virus–related disease may be induced experimentally in suckling mice (Mayo et al. 1978); however, in general, infection in rodents is asymptomatic. In guinea pigs, clinical disease may be observed in breeding females and transmission in utero cannot be excluded (Nicklas et al. 2002; Chatterjee et al. 2001). For all herpes viruses, diagnosis can easily be achieved by serology. RNA Viruses A range of different RNA viruses infect rodents and rabbits. In immune-competent animals, RNA virus infection is normally nonpersisting, except for those viruses able to incorporate themselves into the genome of the host (i.e., retroviruses) or viruses that generate an insufficient immune response (i.e., lymphocytic choriomeningitis virus). The morbidity and mortality caused by these viruses vary greatly. Most are host specific, except for lymphocytic choriomeningitis virus. RNA viruses are antigenically quite close; the same virus is often reported to be able to infect more than one species. However, infection in different species seems likely to be caused by different substrains. Coronaviruses Coronavirus infection in mice or rats is the most important and the most common viral problem encountered in laboratory rodent facilities. Coronaviruses in mice—mouse hepatitis virus (MHV)—and in rats—known as sialodacryoadenitis virus (SDAV) and rat coronavirus (RCV)— are antigenically very close, yet are different and highly species-specific viruses. Most MHV infections in mice are latent, but characteristics of both the virus and the animal will affect expression. Some strains are enterotropic (i.e., they infect through the gastrointestinal system), while others are pneumotropic (i.e., they infect through the respiratory system) (Homberger 1997). The most common symptom is diarrhea in suckling mice (Broderson et al. 1976a). Several organs are affected by the infection and the virus has a tropism for all tissues, which in connection with its immunomodulating effects makes it a significant threat to research projects and an undesired organism in rodent facilities. The liver may be pale with multiple white, yellow, or hemorrhagic foci (Homberger 1996). The spleen may be enlarged, while the thymus may be reduced in size and there is widespread necrosis of lymphoid tissue. BALB/c and C57BL mice seem to be more sensitive than other mice (Taguchi et al. 1976). In immune-deficient mice, such as SCID (Percy and Barta 1993) and nude (Sebesteny and Hill 1974) mice, infection causes high mortality. Strains such as MHV-2, MHV-3, and MHV-A59 are more virulent than some other strains, including MHV-1, MHV-S, MHV-Y, and MHV-Nu. A strain designated MHV-4 has a specific affinity for the nervous tissues (Barthold and Smith 1984). Coronavirus infection in rats caused by one of several substrains of SDAV or RCV may often be asymptomatic, but clinical symptoms seem to be more common after coronaviral infection in rats
84550.indb 264
11/11/10 6:25:39 PM
Health Status and Health Monitoring
265
than in mice. A mild necrotizing rhinotracheitis develops into interstitial pneumonia with severe necrosis and swelling of salivary and lacrimal glands. In this phase, the ventral neck region of the rat is swollen, and red-brown porphyrin rings are observed around the eyes. Diseased rats fully recover within 5 weeks (Jacoby et al. 1975). Coronaviruses easily spread in rodent facilities, and the prevalence in a rat or mouse colony can reach 100% within 4 weeks (Hansen and Jensen 1997). Although unstable, coronaviruses may be transported passively between facilities by staff or equipment. Coronaviruses are nonpersisting in immune-competent animals, but persisting in immunedeficient and certain transgenic animals. Current screening can easily be performed by serology (Hansen and Jensen 1995). Diagnosis of acute disease may be done by PCR (Casebolt et al. 1997) or by tests for anticoronaviral IgA (Hansen and Jensen 1997) in fecal samples. Coronaviruses different from murine coronaviruses may infect guinea pigs or rabbits, and, until the 1960s, pleural effusion as the result of infection with rabbit coronavirus was a common disease in laboratory rabbits (Jensen 1971). Enteric infection with rabbit coronavirus may still be common (Nicklas et al. 2002). Cardioviruses Cardioviruses are picornaviruses producing enteric infection in a wide range of mammals. Similarly to polioviruses, certain strains may, under specific conditions, invade the central nervous system and produce neurodegenerative disease. Theiler’s mouse encephalomyelitis virus (TMEV), occasionally referred to as mouse polio, can be divided into three groups of substrains infecting mice: GDVII, FA, and DA. Some strains infect rats (McConnell et al. 1964; Ohsawa et al. 1998a), but they are not well described. Antibodies to TMEV are a common finding in rats (Schoondermark-van de Ven et al. 2006; Zenner and Regnault 2000), although they occur in the absence of clinical signs. GDVII and FA are far more virulent than DA, but in most cases, spontaneous infections are asymptomatic. GDVII may cause acute encephalitis. Paralysis develops when the virus leaves the gray matter and infects the white matter, thereby damaging the upper motor neuron system (Lipton and Dal Canto 1976). CD-1, DBA/2, SJL, and SWR mice seem to be among the most susceptible strains (Lipton and Melvold 1984). DA causes a more long-term, demyelinating disease. In infected colonies, prevalences are generally high, and the virus may be transmitted vertically (Abzug and Tyson 2000), complicating rederivation. Encephalomyocarditis virus (EMCV) is a virus commonly used for experimental infection of mice, especially for diabetes research (Buschard et al. 1983). Spontaneous infections have not been found in mice. Guinea pigs may also harbor a cardiovirus; the causative agent of the disease, guinea pig lameness, is a paralytic and fatal disease. Deficiency of vitamin C may be a predisposing factor, allowing the virus to spread to the central nervous system. Therefore, colony-wide outbreaks are likely to occur if the content of vitamin C in the diet is accidentally low (Clausen et al. 2001). Serological studies have revealed the presence of antibodies cross-reacting against TMEV in guinea pigs suffering from lameness (Hansen et al. 1997), but DNA techniques show that the virus is probably closely related to EMCV (Hansen et al. 2000). This virus is rather common in laboratory guinea pigs (Hansen et al. 1997), but is seldom monitored as part of commercial breeder’s health monitoring programs. Paramyxoviruses Parainfluenza viruses are unstable viruses producing nonpersisting infection in immunecompetent rodents and persisting infection in immune-deficient rodents (Carthew and Sparrow 1980a). They may be divided into type 1, type 2, and type 3, as well as a strictly human type 4. Sendai virus, a type 1 parainfluenza virus, produces respiratory infections in rats and mice. Isolates
84550.indb 265
11/11/10 6:25:39 PM
266
Handbook of Laboratory Animal Science
are antigenically alike and show the same pathogenicity for both species. Clinical symptoms are rare, but high mortality may be seen in young mice before or around weaning. DBA/2 and 129 mice may be more sensitive than other strains (Parker et al. 1978). Pathological signs are catarrhal bronchitis, eventually extending into the alveoli of the lung. Transmission is mainly respiratory (Iida 1972). Diagnosis can easily be made by serology, but the virus does not spread efficiently among sentinels by the dirty bedding technique (Artwohl et al. 1994). Rabbits may be experimentally infected with Sendai virus (Machii et al. 1989) and spontaneously occurring antibodies have also been found in rabbits (Iwai et al. 1986). Antibodies to Sendai virus have also been observed in guinea pigs (Nakagawa et al. 1986). However, Sendai virus has not been isolated from guinea pigs, nor has experimental infection been achieved; thus, these antibodies are most likely cross-reactions from guinea pig parainfluenza virus. This type 3 parainfluenza virus was not isolated from guinea pigs until 1998 (Ohsawa et al. 1998a), but antibodies were found in guinea pigs as early as the 1970s (Welch et al. 1977). Guinea pig parainfluenza virus type 3 is a lineage of human parainfluenza virus type 3, probably introduced into guinea pig colonies via infected humans (Ohsawa et al. 1998a). In infected breeding colonies, all parainfluenza viruses normally have high prevalences. Infection with the pneumovirus, pneumonia virus of mice (PVM), normally causes a silent infection in mice, rats, hamsters, and gerbils. Serological prevalences are high in infected colonies, but pathological symptoms are absent in immune-competent mice. Mild pathological changes may be found in rats and in nude mice. In the latter, infection is also persistent (Carthew and Sparrow 1980b), in contrast to all immune-competent animals, which typically clear the infection. Currently, screening can be accomplished by serology. Reoviruses A number of wild-type and laboratory strains of reovirus type 3 have been recovered from both vertebrates and nonvertebrates. So far, the virus has only been isolated from mice, but antibodies also have been detected in rats, guinea pigs, and rabbits. Whether these antibodies are specific or are due to cross-reactions is not known. Previously, 15% of European guinea pig colonies were shown to have positive titers to reovirus type 3 (Kraft and Meyer 1990), although without clinical or pathological changes. The situation is much improved today; the virus is relatively heat stable and it is also resistant to some chemical disinfectants and can survive outside the body for long periods. Transmission is mainly by the oral route, but airborne contamination may occur. Intrauterine infection is described under experimental conditions, but is not likely to occur naturally. Antibodies can be detected by serology. Rotaviruses Rotaviruses have been isolated from numerous mammalian species. In general, they cause enteric infections leading to diarrhea, especially in newborn animals. These viruses are highly contagious, and prevalence in infected colonies is high. Infection with rabbit rotavirus is common in rabbits (Rizzi et al. 1995; Percy et al. 1993). In mice, rotaviral disease is called epizootic diarrhea of infant mice (EDIM); in rats, it is called infectious diarrhea of infant rats (IDIR) (Vonderfecht et al. 1983). EDIM and IDIR are different serotypes. Rodent rotaviral infections are becoming less common. Currently, screening can be performed in all species by serology, while diagnosis of acute disease can easily be achieved by capture ELISA or latex agglutination on feces. Different serogroups exist and monitoring in mice and rabbits must be carried out using a serogroup A antigen (Nicklas et al. 2002).
84550.indb 266
11/11/10 6:25:39 PM
Health Status and Health Monitoring
267
Togaviruses Lactate dehydrogenase elevating virus infects laboratory and wild mice worldwide. It is mostly known for its ability to elevate serum lactate dehydrogenase, as well as a range of other serum enzymes, in silently infected mice, but it may also produce paralysis after infection of the central nervous system. The c strain, also called the Murphy strain, is especially likely to produce paralysis; AKR and C58 mice seem to be more susceptible than other strains (Martinez et al. 1979). The Murphy strain is a common contaminant of transplantable tumor cell cultures (Nicklas et al. 1993). It is not clear how it actually spreads. Cannibalism and fighting, especially among male mice, may be a mechanism for transmission, but a parasitic vector may be needed, making epizootics in mouse colonies less likely. Transplacental infection may occur (Crispens 1967). It is most easily diagnosed by testing mice for elevated levels of lactate dehydrogenase; biological materials may be screened by PCR (Chen and Plagemann 1997). Arenaviruses Lymphocytic choriomeningitis virus has been found to naturally infect hamsters frequently, while less frequently infecting mice and guinea pigs. Furthermore, it is zoonotic and has the capability of infecting humans. In a few cases, mostly associated with pet hamsters, this may lead to meningitis (Biggar et al. 1975a, 1975b; Deibel et al. 1975). Infection may be persistent in mice, especially if infection occurs in utero or within the first 7 days after birth. Lymphocytic choriomeningitis virus has been widely used as an experimental agent for studying viral immunology. It spreads slowly in colonies and prevalences seldom reach more than 10%, although prevalence may be higher among animals housed on the lower racks (Smith et al. 1984). Vertical spread from mother to fetus is the principal path of infection (Parker et al. 1976). Natural infections in mice and hamsters are normally silent, while more severe symptoms may be observed after infection with specific strains in guinea pigs (Dutko and Oldstone 1983). Serology is the principal method for routine monitoring. Bunya Viruses Hantaviruses are zoonotic viruses that spread from wild animals to humans, mostly resulting in silent infections; however, in rare and unfortunate cases, they can produce hemorrhagic disease. Several types exist, but the only type of interest for laboratory animals is the Seoul strain, which produces nonclinical infection in rats (Meyer and Schmaljohn 2000). It is found mainly in the Far East and in the Balkan region (Lee et al. 1978). In other areas of the world (e.g., Scandinavia and the United States), other types of hantaviruses with specificity for other animal species may be found (LeDuc 1987). Infections in laboratory animals are rare, but have occurred after housing wild rodents in laboratory animal facilities (Desmyter et al. 1983). Screening is accomplished by serology. Caliciviruses Rabbit hemorrhagic disease virus (RHDV) has spread from Asia to Europe over several decades, primarily among slaughter rabbits. It is also found in Mexico. It can cause sudden death via hemorrhage, especially from the nasal passages. In acute cases, mortality may approach 90%, but chronic cases with lower mortality, distress, and icterus may also be seen. RHDV can be diagnosed by ELISA, but cross-reactions from nonpathogenic caliciviruses may occur (Capucci et al. 1996).
84550.indb 267
11/11/10 6:25:40 PM
268
Handbook of Laboratory Animal Science
The murine noroviruses (MNV) are persisting viruses that were not reported until 2003 (Karst et al. 2003). Today, at least four different types are known (Hsu et al. 2006) to occur in mouse colonies (Perdue et al. 2007). Infection of wild-type mice is histopathologically associated with mild inflammation in the intestine and red pulp hypertrophy and white pulp activation in the spleen (Mumphrey et al. 2007). However, clinical disease apparently only occurs in a small number of strains of immunodeficient mice lacking interferon (IFN) response (e.g., RAG mice, which lack the IFN-aβ and IFN-γ receptor signaling gene STAT1 (Karst et al. 2003; Mumphrey et al. 2007). Eradication can only be accomplished by complete rederivation: removal of infected animals, disinfection of rooms, and repopulation with uninfected animals (Kastenmayer et al. 2008). Screening is accomplished using serological techniques, such as microsphere fluorescent immunoassay (MFI) (Hsu et al. 2006). Parasitological Infestations Laboratory animals should be free of parasites. These may not have obvious clinical impact on the animals, but they do have subclinical impact and may interfere with research in various ways (e.g., by upregulating the regulatory immune response) (Maizels 2009). Parasitic infestations are symptomatic of low hygienic standards, and infested animals also may be suspected of carrying other types of infections. Important parasites in rodents and rabbits are listed in Table€10.3. The most common parasites found in laboratory rodents are the flagellates, Tritrichomonas spp., but these also seem to have low impact on the animals. Pinworms and mites are also common findings. Encephalitozoon cuniculi is now rare in colonies of rabbits and guinea pigs, but when it occurs, it can have a measurable impact on research (Kunzel and Joachim 2010). Table€10.3â•…Parasitic Infestations Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) Mastigophora (flagellates) Chilomastix spp. Giardia spp. Spironucleus muris Tritrichomonas spp. Tetratrichomonas minuta Pentatrichomas homonis Trichomitis spp. Hexamastix spp. Enteromonas spp. Retortamonas spp. Monocercomonoides spp. Chilomitus spp. Octimitus spp.
R, H, RB M, R, GP, H, RB M, R, H M, R, GP, H M, R, H M, R, H R R, GP, H R, GP R, GP, RB R, RB GP R
Sarcodina (amebas) Entamoeba muris Entamoeba cuniculi
M, R, H RB Sporozoa
Encephalitozoon cuniculi Eimeria falciformis â•… Eimeria spp.a
84550.indb 268
M, R, GP, RB M
11/11/10 6:25:40 PM
Health Status and Health Monitoring
269
Table€10.3╅Parasitic Infestations Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) (continued) Sporozoa Eimeria caviae Eimeria spp.b Eimeria stiedae Cryptosporidium spp. Toxoplasma gondii Sarcocystis muris Sarcocystis cuniculi Klossiella muris Klossiella cobayae
GP RB RB R M, R, GP, RB M, R RB M GP Ciliata
Balantidium spp.
R, GP, H Nematodes
Stomach worms Graphidium strigosum
GP, RB
Intestinal and cecal worms Trichostrongylus spp. Paraspidodera uncinata
RB GP
Pinworms Aspiculuris tetraptera Dermatoxys veligeria Passaluris ambiguus Syphacia muris Syphacia obvelata
M, R RB RB R, H M, R, H
Bladder worms Trichosomoides crassicauda
R
Threadworms Strongyloides ratti Capillaria hepatica
M, R, H M, R, RB
Lungworms Protostrongylus spp.
RB Cestodes
Adult tapeworms Cittotaenia variabilis Hymenolepis nana Hymenolepis diminuta
RB M, R, H M, R, H
Cysticerci of tapeworms Cysticercus pisiformis Coenurus serialis Strobilicercus fasciolaris
RB R M, R (continued)
84550.indb 269
11/11/10 6:25:40 PM
270
Handbook of Laboratory Animal Science
Table€10.3╅Parasitic Infestations Observed in Mice (M), Rats (R), Guinea Pigs (GP), Syrian or Chinese Hamsters (H), and Rabbits (RB) (continued) Trematodes Liver flukes Fasciola hepatica Dicrocoelium dendriticum
GP, RB GP, RB
Hair Follicle Mites Demodex aurata Demodex caviae Demodex criceti Demodex musculi Demodex nanus
GP M Ear Mange Mites
Notoedres muris Psoroptes cuniculi
R, GP, H RB Body Mange Mites
Psorergates simplex Notoedres cati Sarcoptes scabiei
M RB M, R, GP, RB Fur Mites
Cheyletiella parasitivorax Chirodiscoides caviae Myobia musculi Myocoptes musculinus Listrophorus gibbus Radfordia affinis Radfordia ensifera Trichoecius romboutsi
RB GP M M, GP RB M R M Lice
Haemodipsus ventricosus Polyplax serrata Polyplax spinulosa Gliricola porcelli Gyropus ovalis
RB M R GP GP Ticks
Haemaphysalis leporis-palustris a b
RB
E. nieschulzi, E. miyarii, E. contorta, E. separate. E. irresidua, E. magna, E. media, E. perforans, E. exigua, E. intestinalis, E. matsubayishii, E. nagpurensis, E. neoleporis, E. Piriformsis.
Pinworms Laboratory rodents often harbor pinworms (i.e., ascarids of the family Oxyuridae). In rodents, the most common species are Syphacia and Aspiculuris spp. Syphacia obvelata is the most common in mice, S. muris is the most common in rats, and hamsters may occasionally harbor a third species, S. mesocriceti. Aspiculuris tetraptera is nearly as common as Syphacia spp. in mice. Up to 70% of U.S. laboratory animal facilities have recently been shown to carry pinworm infestations (Jacoby
84550.indb 270
11/11/10 6:25:40 PM
Health Status and Health Monitoring
271
and Lindsey 1998). In rabbits, the most common species is Passalurus ambiguus. Pinworms live in the free lumen of the caecum, colon, and rectum. They are extremely infectious, easily spreading among staff and on equipment, with a small infectious dose. Clinical symptoms are rare. Diagnosis is made simply by observing the worms during necropsy or by detecting eggs in a tape or flotation test. Safe identification may be made by characterization of rDNA sequences (Parel et al. 2008). Protozoans Tritrichomonas spp. The most common endoparasite in rodents is the flagellate Tritrichomonas muris. It is routinely found in the caecum and colon of conventional rodents, but may be found in barrier-protected rodents as well (Hansen and Jensen 1995). Transmission occurs by ingestion of pseudocysts. It seems to be apathogenic. Encephalitozoon cuniculi. Encephalitozoon cuniculi, a protozoan parasite belonging to the subphylum Microspora, is the etiological agent of a spontaneous disease in rabbits, encephalitozoonosis, which is often referred to as nosematosis due to the older genus name Nosema. The most common route of transmission is via infectious feces and urine. The sporoplasm is extruded from a spore, which then enters host cells to multiply and mature into additional spores, restarting the cycle after cell rupture. Infection has been described in many species, including rabbits (Waller et al. 1978), rats (Gannon 1980), mice (Gannon 1980), and guinea pigs (Boot et al. 1988). However, infection is only common in rabbits and guinea pigs; infection in other species is rare and may be the result of contact with contaminated rabbit colonies (Gannon 1980). Infection is usually latent, but rabbits occasionally exhibit various neurological signs, including convulsions, tremors, torticollis, paresis, and coma. Lesions in the kidneys of infected rabbits are frequent and are grossly manifested as multiple pinpoint areas, randomly scattered over the surface, or, more usually, as 2–4 mm indented gray areas on the cortical surface. In histopathology, granulomatous nephritis (Flatt and Jackson 1970) and granulomatous encephalitis (Koller 1969) are observed. Guinea pigs do not seem to develop the disease. Infection is diagnosed by serology (Boot et al. 2000). The disease is usually diagnosed by recognition of typical lesions and the agent is sometimes diagnosed by histopathology of the kidneys. Eimeria spp. Various Eimeria species are known to cause intestinal coccidiosis in rabbits. Infection is fairly common, but if observed at all, clinical symptoms are mostly observed around weaning in breeding facilities and are rare in experimental facilities. Heavily infected rabbits may develop clinical symptoms, including varying degrees of diarrhea, thirst, and dehydration. Weight loss may be observed in subclinical cases. Peracute cases with deaths prior to the presence of oocysts in the feces may be seen. The small and large intestine of heavily infected rabbits may show multiple white spots on the mucosa with a mixed mononuclear and polymorphonuclear exudate. Infection occurs orally with coccidial oocysts, which burst in the intestines, releasing sporozoites to invade the intestinal mucosa cells. Here they multiply into schizonts, which break, extruding a huge number of merozoites into the lumen from where they invade new cells and repeat the process. After an unknown number of such asexual generations, designated schizogony, the merozoites differentiate into female macrogametocytes or male microgametocytes, which leave the cells and unite to form the oocysts. The most common enteric Eimeria species in the rabbit are E. irresidua, E. magna, E. media, and E. perforans. Eimeria stiedae produces hepatic coccidiosis with multiplication in the bile ducts, but this infection is uncommon in laboratory rabbits. Clinical symptoms of this infection are mostly related to chronic debilitation of the liver, weight loss, and chronic intestinal symptoms that in rare cases lead to death (Baker 1998). The diagnosis of Eimeria infection is made by the observation of oocysts in the feces in the flotation test. The diagnosis coccidiosis is made by observation of the clinical and pathological symptoms combined with the observation of large numbers of oocysts in the feces.
84550.indb 271
11/11/10 6:25:40 PM
272
Handbook of Laboratory Animal Science
Ectoparasites Ectoparasites are a common finding (Jacoby and Lindsey 1998) in conventional rats and mice. Mites such as Myobia musculim, Radfordia affinis, and Myocoptes musculinus in mice and R. ensifera in rats are the most frequently seen. Mites may cause no symptoms or a variety of skin lesions, ranging from mild pruritus to serious pyodermia. Other ectoparasites occur less frequently, although lice infestations with Polyplax spp. have been observed in conventional rodent colonies. Ectoparasites are generally absent in barrier-protected rodent colonies. Microbial Interference with Animal Experiments Clinical disease is only one of several ways that infectious agents may be detrimental to research. Therefore, the absence of disease symptoms should not be interpreted as the absence of problematic infections. Pathological Changes, Clinical Disease, and Mortality Many experiments have been ruined by disease or pathological changes caused by specific infections, as described next. Subclinical disease may also disturb essential parameters; for example, subclinical viral infections may affect body weight in rats. Additionally, behavior will often be changed during subclinical disease, leading to disturbances in the open field test, etc. The presence of some microorganisms may result in changes in the organs—for instance, presenting difficulties in the interpretation of the pathological diagnosis included in toxicological studies (a phenomenon often referred to as “background noise”). Respiratory disease of any etiology may be responsible for deaths during anesthesia. Certain strains, inbred or transgenic, may be more prone to the development of specific pathological changes. For instance, certain inbred strains suffer from particular pathological changes, and transgenic mice without various immunologically active genes suffer from a syndrome of gastric ulcers, colitis, and rectal prolapses (Bhan et al. 1999) (Figure€10.3).
Figure 10.3â•…Certain transgenic knockout mouse strains suffer from gastric ulcers, colitis, and rectal prolapses, as shown by this gastric ulcer in a plasminogen knockout mouse. (Photo courtesy Kirsten Dahl.)
84550.indb 272
11/11/10 6:25:41 PM
Health Status and Health Monitoring
273
Contamination of Biological Products Microorganisms present in the animal may contaminate samples and tissue specimens, such as cells, sera, etc. (Nicklas et al. 1993). This may interfere with experiments performed on cell cultures or isolated organs. Furthermore, the introduction of such products into animal laboratories will present a risk to the animals kept in that laboratory. In theory, any infection may contaminate products from the animal. However, viruses are known to represent the major risk, due to the ability of some viruses to produce viremia and to persist in specific organs, thereby increasing both the time and the number of tissues at risk. The most common viral contaminant is lactic dehydrogenase virus followed by reovirus type 3, lymphocytic choriomeningitis virus, minute virus of mice, mouse hepatitis virus, rat coronaviruses, Kilham rat virus, and Mycoplasma pulmonis (Nicklas et al. 1993), in that order. Other Mycoplasma spp. may contaminate biological materials (Hopert et al. 1993), but these are often nonrelevant species derived from humans or unrelated species used as serum donors for cell culture media. Ectromelia virus has previously been brought into two U.S. laboratory animal facilities through contaminated cell lines, with enormous economic impact on these institutions (Lipman et al. 2000; Dick et al. 1996), but today laboratories seem to have become more careful. Some protozoans (e.g., Encephalitozoon cuniculi), as well as bacteria, also have the potential to contaminate transplantable tumors (Nakai et al. 2000; Goto et al. 2001). Immunomodulation Many experiments are predicated on the animal model having a functional immune system (e.g., studies of autoimmune type 1 diabetes mellitus in the NOD mouse). Microorganisms may cause immunomodulation, even in the absence of clinical disease, with the effects including suppression and/or activation of different components of the immune system. Viruses are normally described as the most frequent immune modulators, due in part to the viremic phase of the pathogenesis of many viral infections. During viremic phases, such as those associated with adeno-, parvo-, and coronaviruses, cells of the immune system become infected (Mims 1986). The thymus itself may become infected by thymic virus (Cross et al. 1979). Leukemia viruses even integrate themselves into the leucocytes by reverse transcription. Such infections of the immune cells may suppress the immune system through a variety of mechanisms. One of the most well-described viral infections suppressing the immune system is mouse cytomegalovirus infection (Hamilton et al. 1979), an effect likely to be specifically associated with the emission of cytokines from macrophages and the resultant dominance of T-suppressor cells (Loh and Hudson 1981, 1982; Shanley and Pesanti 1980, 1982). The immunosuppressive effect of Mycoplasma infections seems to differ from viral immunosuppression in the secretory stimulation of the macrophages because interferon production during Mycoplasma infection seems to be impaired (Kaklamanis and Pavlatos 1972). Mycoplasma spp. have generalized effects on the specific response to some antigens (Specter et al. 1978; Westerberg et al. 1972), as well as influence on the nonspecific host defense mechanisms, especially in the respiratory system (Pollack et al. 1979; Laubach et al. 1978; Wells 1970; Ventura and Domaradzki 1967). Immunomodulatory bacteria include group A Streptococci, Pseudomonas aeruginosa, Escherichia coli, and Salmonella spp. These effects may be suppressive as well as stimulatory and are often mediated through endotoxin production, most likely lipid A (Thomsen and Heron 1979). Parasites that influence the immune system include protozoans, such as Toxoplasma gondii (Krahenbuhl et al. 1972; Ruskin and Remington 1971, 1976; Swartzberg et al. 1975) or helminths, such as Syphacia spp. (Sato et€al. 1995). There is increasing evidence that parasite infestation may have a direct impact on the formation of regulatory T-cells (Hansen et al. 2009); for instance, infection with Schistosoma reduces the incidence of T1D in NOD mice (Zaccone et al. 2003).
84550.indb 273
11/11/10 6:25:41 PM
274
Handbook of Laboratory Animal Science
Physiological Modulation Some microorganisms have a specific effect on enzymatic, hematological, and other parameters monitored during an experiment. The classic example is lactate dehydrogenase (LDH)-elevating virus in mice, which inhibits the clearance of LDH and a number of other enzymes (Notkins 1971). Also, mouse hepatitis virus alters hepatic enzyme activity (Ruebner and Hirano 1965). Such disturbances of organ function may change the outcome of an experiment; for example, the altered function of the liver and macrophages in mice infected with mouse hepatitis virus may lead to altered responses in toxicology, nutrition, and other areas in which the liver is involved (Tiensiwakul and Husain 1979). These disturbances may be irreversible for some drugs and reversible for other drugs, as has been described in mice infected with Clostridium piliforme (Friis and Ladefoged 1979). Interference with Reproduction Decreases in fertility, as a function of infections that give rise to clinical disease, can be a major problem in breeding colonies. Infections such as rotaviruses (Vonderfecht et al. 1984) or mouse hepatitis virus (Gustafsson et al. 1996), which cause high mortality in newborn animals, can also be problematic, reducing breeding output and/or disturbing experiments that involve newborn animals. Direct effects of an infection on reproduction (such as parvoviruses; Kilham and Ferm 1961), including changes in sex hormones, pathological anatomical changes in the reproductive tract, or embryonic infections that cause abortion and stillbirths, have also been observed. Many microorganisms, including retroviruses, lymphocytic choriomeningitis virus (Parker et al. 1976), parvoviruses (Kilham and Margolis 1969), ectromelia virus (Schwanzer et al. 1975), cardioviruses (Abzug and Tyson 2000), and Clostridium piliforme (Friis 1978, 1979), possess the ability to cross the placental barrier. Uterine infections—probably without crossing the placental barrier—have been observed for many bacteria, including Salmonella spp. (Okewole et al. 1989) and Pasteurella pneumotropica (Blackmore and Cassillo 1972). However, with the exception of the retroviruses and lymphocytic choriomeningitis virus, spontaneous infections of fetuses are rare. Mycoplasma pulmonis is known to be very harmful to reproduction, altering a number of reproductive parameters (Brown and Steiner 1996), and infection with Mycoplasma spp. is a risk in embryo transfer (Hill and Stalley 1991). Competition between Microorganisms within the Animal In some studies, experimental infection is the main focus of the research. In others, inactivated bacteria are used for the induction of an immunological response. In both cases, it is imperative that the animal model not carry that infection prior to the experiment. Animals used for the propagation of viruses may be experimentally immunosuppressed to increase susceptibility to the inoculated virus; however, other organisms already present in the animal may propagate instead. Some infections, such as Corynebacterium kutscheri, Kilham rat virus, Sendai virus, and sialodacryoadenitis virus (Barthold and Brownstein 1988), reduce the severity of disease caused by other agents, adversely affecting infectious disease models. Modulation of Oncogenesis Infectious agents may induce cancer, enhance the oncogenic effect of certain oncogens, or reduce the incidence of cancer in laboratory animals. Spontaneous tumors represent the most important complication in studies with aging animals or other long-term studies. Retroviruses (Schramlova et al. 1994) and Helicobacter spp. (Fox and Lee 1997) may be oncogenic by themselves. Some microorganisms (Mycoplasma pulmonis, Lynch et al. 1984; Citrobacter rodentium, Barthold and
84550.indb 274
11/11/10 6:25:41 PM
Health Status and Health Monitoring
275
Jonas 1977) that are not primary oncogens may increase the incidence of specific tumors, while other organisms (Salmonella, Ashley et al. 1976; Tindle et al. 1976; H1 parvovirus, Toolan and Ledinko 1968; Toolan 1967; Toolan et al. 1982) may decrease it. The Impact of the Gut Microbiota of Laboratory Animals Even if laboratory animals are free of all of the specific infections mentioned earlier, they still harbor 10 different bacterial species (Hayashimoto et al. 2007), unless they were raised under strict gnotobiotic conditions. Several studies indicate that this unspecific gut microbiota is of importance for the development of disease animal models (Hansen et al. 2009). A range of transgenic mice, as well as some rats, suffer from a syndrome consisting of gastric ulcers, colitis, proctitis, and rectal prolapses (Figure€10.3). This syndrome, which in its features resembles several aspects of human inflammatory bowel disease, is primarily observed in strains in which transgenesis has disrupted the normal mucosal homeostasis through cytokine imbalance, abrogation of oral tolerance, alteration of epithelial barrier and function, or loss of immunoregulatory cells. Members of the enteric flora are important factors in the development of the syndrome, as antibiotic treatment (Rath et al. 2001) and germ-free conditions (Kawaguchi-Miyashita et al. 2001) may prevent it. Helicobacter spp. has been proposed as a cause also in rodents (Chin et al. 2000), but a more parsimonious explanation is that disruptions of the gut immunological balance lead to difficulties coping with gut bacteria (Rath et al. 1996, 1999, 2001). Strong associations exist between intestinal inflammation and rheumatic arthritis (RA); germ-free status prevents experimental induction of RA (Taurog et al. 1994; Breban et al. 1993). A clean status in type 1 diabetes-prone rodent colonies results in a high incidence (Tlaskalova-Hogenova et al. 2004; Pozzilli et al. 1993; Wilberz et al. 1991), and obesity also seems to be influenced by the gut microbiota (DiBaise et al. 2008; Backhed et al. 2004; Turnbaugh et al. 2006). Transgenic mice without the galactose-presenting alpha-gal epitope shared by most mammals with a high number of gut bacteria have increased insulin resistance (Dahl et al. 2006). Phytoestrogen inflavones (genistein, daidzein, and glycetin) decrease both total and LDL cholesterol (Merritt 2004) and are rapidly reduced to the metabolite equol (Breinholt et al. 2000) in mice. This production shows higher individual variation in humans (Lu and Anderson 1998) and pigs (Kuhn et al. 2004), and equol turnover in mice may be reduced by feeding Lactobacillus gasser (Tamura et al. 2004). Gut microbiota is difficult to control completely and is therefore likely to be a cause of unwanted variation across a variety of animal models. Eradication Methods It is essential to have uninfected breeding animals when attempting to build a colony that is free of problematic infections. These are produced by so-called rederivation (e.g., caesarian section or embryo transfer)—procedures during which the uninfected future breeding animals are removed from the infected mother as embryos or fetuses. Alternatives to rederivation are cessation of breeding, stamping out, antibiotic treatment, or vaccination, although these methods are less recommendable. Caesarian Section Offspring removed from their mother’s uterus by sterile techniques are free of all infections that cannot cross the placental barrier. Caesarian sections may even eliminate infections that can pass through the placenta because not every fetus of an infected mother will harbor the infection (Hansen et al. 1992b). Therefore, building rodent colonies utilizing breeding animals born
84550.indb 275
11/11/10 6:25:41 PM
276
Handbook of Laboratory Animal Science
by caesarian section has long been an accepted rederivation strategy. The steps for the caesarian section procedure for rederivation are described next. The exact date of expected birth of the mother of fetuses that will become the future breeding animals needs to be known for planning the rest of the procedure. Therefore, timed-mating procedures need to be used (the female is given 24 hours with the male, after which the male is removed). Mating outcome needs to be checked by observation of a vaginal plug or by the microscopic observation of sperm in a vaginal smear. The latter is recommended for female rats, which often lose their plugs in their bedding. If the female reproductive cycle can be synchronized, which is possible for some larger farm animals, this can result in higher and more efficient mating rates. In rabbits, ovulation is induced by mating rather than by a cycle, so timed mating, using artificial insemination, is a simple matter (Theau-Clement 2008). Rodents, on the other hand, are difficult to synchronize. Mice can be synchronized by placing females in the same cage as the male, but separated by a grid for 2 days. When the grid is removed on the third day, a mating rate of 50% may be achieved. Five times as many female rats must be synchronized as are actually needed since only 20% will reach estrus within 24 hours. Performing a vaginal smear prior to mating, as well as only mating proestral females, can reduce the number of females to be mated by up to 50%. It may be advantageous to treat the mother with a gestagen, such as medroxyprogesteron (15 mg/kg body weight), 3 days prior to the expected birth date to prevent spontaneous births. Fetuses from treated females can only be harvested by caesarian section because the mother is unable to give birth by herself. It is the experience of this author that the exact duration of pregnancy differs across strains and the closer the caesarian procedure is performed to the real time of birth (based on days since conception), the higher the rate of success will be. To avoid contamination, strictly aseptic techniques should be employed, which are achievable in a number of different ways (Figure€10.4). To reduce the risk of infection with Pasteurella pneumotropica, the pieces of the reproductive tract should never be introduced into the isolator (Mikazuki et al. 1994). It is preferable to euthanize the donor mother mechanically rather than to anesthetize her. The uterine blood supply in rats and mice is sufficient enough to allow 15 minutes for the pups to be released; however, the entire procedure should not last more than 4 minutes for rabbits or guinea pigs. Newborn, derived pups are typically placed with a foster mother whose own offspring have recently been removed. It is the experience of this author that the highest success rates are achieved if the rodent foster mother has given birth within 2–10 days prior to receiving the derived pups. Pups should be rubbed and heated to achieve normal blood circulation prior to being placed with the foster mother because cold, pale pups are likely to be cannibalized. If germ-free animals are needed, the foster mother should be germ free and kept in an isolator. Germ-free conditions are the most attractive for the development of a colony of microbiologically defined animals, since germ-free conditions ease health monitoring procedures. Germ-free status is not possible for hamsters and guinea pigs. For these species, a foster mother of the desired microbiological status should be used. Guinea pig pups are precocial and may be hand-fed a liquid guinea pig diet. However, foster mothers are strongly recommended, since they supply the pups with colostrum and increase pup survival rates. Offspring of all laboratory mammal species may be hand-fed with irradiated milk from a mother of their own species instead of using a foster mother. Pigs may be fed cow colostrum, which is commercially available (Sangild et al. 2006). Rabbit puppies only need to be fed one time each 24 hours. If young animals are to be hand-fed, it is of the utmost importance to weigh the animals both before and after feeding, and to calculate the correct volume of milk to prevent the development of aspiration pneumonias. Pups raised with a foster mother should be weaned according to the species’ normal weaning procedures. A minimum of 8 weeks should pass before health monitoring is performed to evaluate the success of the rederivation. Several offspring and the foster mother will need to be sacrificed. It is advisable
84550.indb 276
11/11/10 6:25:41 PM
Health Status and Health Monitoring
277
III I
II IV (a)
III
IV I II
(b)
III
I II
IV
(c) Figure 10.4â•…Three ways of performing caesarian section for rederivation of laboratory animals. In (a) the uterus removed from the donor mother is transferred to a surgical isolator (I) and the offspring are released through gloves underneath a disinfective solution (II) and transferred via the isolator port (III) to a positive pressure isolator with a foster mother (IV). In (b) the uterus is placed in the disinfective solution (I) and the offspring are placed in a bottle cut open (II), which can be lifted into the high-pressure isolator through a tube passing through the isolator port (III). In (c) the section is performed in a laminar air flow bench (I) and the offspring are placed in an Ehrlen–Meyer bottle (II), which enters through the isolator port (III) by standard procedures.
84550.indb 277
11/11/10 6:25:43 PM
278
Handbook of Laboratory Animal Science
to supply the animals with a basic microflora prior to breeding them outside the isolator. This can be accomplished by breeding gnotobiotic animals of the same species with a defined flora in a different isolator and transferring one of these animals to the isolator with the germ-free animals. This flora should contain at least some anaerobes and some Gram positives (e.g., by applying the so-called Schaedler flora) (Dewhirst et al. 1999). Transfer from the isolator to the barrier unit may be accomplished in a number of ways, but is most easily done if the barrier unit has a port that fits the isolator port. Embryo Transfer Embryo transfer technologies have been used as an alternative to caesarian sections for the last 30 years (Moler et al. 1979). The advantages to caesarian section are that the embryos may be kept for long periods in liquid nitrogen and can be used for further rederivation whenever needed. Embryo transfer is required for the production of transgenic animals. It is a disadvantage that more highly specialized equipment is needed for this method than for caesarian section. Embryo transfer rederivation is also a multistep process (Figure€10.5). Embryos are normally harvested from superovulated females (i.e., multiple ovulations have been induced by hormonal treatment). In mice and rats, this is done by injection of pregnant mare serum gonadotropin (PMSG) combined with human chorionic gonadotropin (HCG) or folliclestimulating hormone (FSH) (Munoz et al. 1995; Popova et al. 2005). The best results are achieved in premature animals. Females are then mated with an appropriate male the night after hormonal treatment and checked for vaginal plugs the next morning. Superovulated matings lead to significantly higher levels of embryo recovery than do natural matings (Munoz et al. 1995). The donor female is euthanized to harvest the embryos. For rederivation procedures, two-cell stages are mostly used because these are the most appropriate for freezing. These are recovered from the salpinx 36 hours after mating. After the donor female is euthanized, the salpinx is excised and transferred into an appropriate medium, and a needle is used to flush out the embryos. Depending on the risk of collecting hazardous infections (such as parvoviruses) along with the embryos, the Donor Females
Recipient Females
Superovulation
Mating with vasectomized male
Mating Sampling of embryos from euthanized female
Placing embryos in anaesthetized female Cultivation
Storage in liquid nitrogen
Birth of puppies N2
Figure 10.5â•…Principles of embryo transfer.
84550.indb 278
11/11/10 6:25:45 PM
Health Status and Health Monitoring
279
embryos are submitted to a series of rinsing steps (Janus et al. 2009). The embryos may be cultivated in vitro in order to synchronize with mating of the recipient female. When placed in straws in an appropriate medium supplemented with sucrose and freezing protectant, embryos may be frozen in liquid nitrogen. The straws are frozen in liquid nitrogen. For thawing, the embryos are placed in a 37°C water bath and the straws are emptied into a petri dish containing a dilution medium. The embryos are then subjected to a number of elution steps. Embryos may be stored indefinitely in liquid nitrogen and viability is normally above 90%. Embryo survival rates may be increased by certain freezing and thawing procedures (Jin et al. 2010; El-Gayar et al. 2008). If the animals need to be germ free, transfer of embryos and all related procedures should be performed in isolators. The recipient mother needs to be mated with a vasectomized male to induce pseudopregnancy. This is accomplished as described for timed mating of donors for caesarian section. Mating is synchronized with the production and cultivation of embryos of the specific stage to be transferred. The pseudopregnant female is placed in a surgical isolator and anaesthetized. The abdomen is entered through the back skin and both sides of the flank muscles as described for ovariectomy. The ovary with the salpinx is carefully grasped from the abdominal cavity, and while the bursa is opened with great care, the embryos are collected with a transfer pipette. The embryos are placed in the salpinx through the infundibulum. An equivalent number of embryos are placed in each salpinx. The wound is closed with sutures and the recipient female is returned to complete her pregnancy in the isolator. From this point on, the steps are similar to those described previously for caesarian section. Cessation of Breeding and Burnout Viruses that induce an immune response in the host animal that is strong enough to eliminate the virus, protect against reinfection, and reach 100% prevalence in a colony may be eliminated by a time break in all breeding procedures. In this scenario, all animals get infected and develop protective immunity; no naïve animals are introduced until this has occurred. This technique has been successfully utilized to deal with coronaviruses in both rats (Brammer et al. 1993) and mice (Weir et al. 1987). At least 6 weeks should transpire between the last births before the break and the first matings after the break. This method is not recommended for immune-deficient rodents, such as nude or SCID mice. Great caution should be used if this is attempted with transgenic animals because viruses may behave abnormally in these animals and develop persistent states, or immunity may never develop (Rehg et al. 2001). Similar goals may be achieved in experimental facilities by burnout: not taking in new animals for a period of 6 weeks. These methods will not work for bacterial infections, which are generally persistent. Stamping Out Removal of infected animals from the colony, a method known as stamping out, has been found to be effective for some infections. This has been most successfully applied as a tool against infection with Encephalitozoon cuniculi in rabbits (Cox et al. 1977); breeding animals are serologically tested and positive responders are removed. Encephalitozoon cuniculi has a complicated life cycle and simple infections that spread easily among offspring are more difficult to eliminate using this method. Antibiotic Treatment Antibiotics have been used for the total eradication of specific pathogenic bacteria in immunecompetent animals. This method, known as selective decontamination, has been more or less successfully applied to mice (van der Waaij and Berghuis-de Vries 1974), guinea pigs (van der Waaij et al. 1974), rabbits (Heidt and Timmermanns 1975), and dogs (Walker et al. 1978). Mice may even be made germ free by decontamination (Srivastava et al. 1976). However, care should be taken
84550.indb 279
11/11/10 6:25:45 PM
280
Handbook of Laboratory Animal Science
because the effect of the antibiotic treatment may simply be that the microorganism can no longer be detected, rather than that it has been eliminated (Hansen 1995). Vaccination Vaccination may be considered if all other measures are unsuccessful or difficult to implement. If vaccination is unavoidable, it must be determined whether vaccination of only breeding females will be sufficient. In this way, pneumonia due to Bordetella bronchiseptica may be almost totally eliminated from breeding colonies of guinea pigs, even though the agent still persists in the colony (Stephenson et al. 1989). Direct vaccination of experimental animals has been used against Sendai virus pneumonia (Tagaya et al. 1995; Iwata et al. 1990; Kimura et al. 1979), ectromelia (Bhatt and Jacoby 1987b), mycoplasmosis (Cassell et al. 1981b; Cassell and Davis 1978), and the various effects of infection with cytomegalovirus (Howard and Balfour 1977). This is a common technique for dealing with infectious problems (Hansen 1998) in larger laboratory animals, such as pigs. However, the use of vaccinated animals may still result in undesirable microbial effects on research, such as the immunosuppressive effect of Sendai virus (van Hoosier 1986). Again, if the animal colony can be kept free of an infection using other techniques, without resorting to vaccination, then these techniques should be implemented.
Containment Facilities Animals from unreliable sources, wild animals (Easterbrook et al. 2008), staff contact with either of these, and biological materials (Nakai et al. 2000) are major infection threats to laboratory animals. To reduce the risk of infections, laboratory animals should be housed in facilities in which certain protective measures are implemented. This is important for both breeding colonies and experimental facilities because infections may spread to a number of studies. Reestablishment of breeding colonies and restarting experiments are time consuming, expensive, and difficult, thus justifying appropriate protective measures. Facilities in which no protective measures are applied are referred to as conventional. Barrier Housing In animal units, in which the staff is allowed to move freely around, protective measures are used for decontamination of the staff, materials, and fresh air entering the unit; that is, a barrier is physically as well as mentally in front of the unit (Figure€10.6). Such a barrier may be run at different levels. Basically, materials and diets are autoclaved or chemically decontaminated at entry and the staff should not be allowed contact with animals of the same species within a certain period (e.g., 48 hours). In the same way, animals should only be introduced if health monitoring has documented the absence of unwanted infections. In breeding units, this normally means that new breeding animals are only introduced by rederivation. In breeding units, staff members are allowed to enter only through a three-room shower, while in experimental units, staff members in some facilities are allowed to enter only after changing their clothes. In-going and preferably also outgoing air is filtered, and the air pressure in the facility is maintained at approximately 15 mm Hg above the surrounding pressure to prevent airborne infection. In some facilities, protection is further enhanced by the use of face masks and gloves by the staff (Figure€10.7). However, even if the staff is equipped with protective clothing, animals kept in barrierprotected facilities may only be kept free of certain species-specific infections. They will not be free of infections shared between their own species and humans (Hansen 1992; Wullenweber et al. 1990b).
84550.indb 280
11/11/10 6:25:46 PM
Health Status and Health Monitoring
281
(a)
(b) Figure 10.6â•…A color version of this figure follows page 336. The outside of two separated barrier units. (a) The shower entrance of barrier 1 is closest, followed by a diptank for chemical disinfection into barrier 1, autoclave for barrier 1, autoclave for barrier 2, diptank for barrier 1, shower entrance and chemical disinfection lock for barrier 2 (Taconic Europe, Denmark). (b) The entrance to a larger autoclave for decontamination through the barrier (Scanbur, Sweden).
84550.indb 281
11/11/10 6:25:49 PM
282
Handbook of Laboratory Animal Science
Figure 10.7â•…A color version of this figure follows page 336. An animal technician dressed for working in a barrier-protected animal unit. (Ellegaard Göttingen Minipigs, Denmark.)
Animals bred in barrier-protected facilities are sold under different terms than animals bred in conventional facilities. The term “microbiologically defined” is preferred, indicating that the animals have been both protected and health monitored. Occasionally, the term “specific pathogen free” (SPF) is used, although the specific pathogens are not always clearly defined. If the aim is to protect the outside of the barrier because animals of unknown, or known but unacceptable, microbiological quality are housed inside the barrier, then containment facilities may be established to work in the opposite direction. In these cases, air pressure would be negative and the decontamination procedures would be applied to staff and materials leaving the facility. Cubicles and Filter Cabinets To enhance the protection of the animals inside a facility, all animals may be placed behind a glass wall in a number of built-in cubicles (Figure€10.8) containing a limited number of cages. Ventilation inlets and outlets are inside each cubicle. Cubicles reduce the spread of infection across individual animals and the spread of allergens in the room. As an alternative to a stationary cubicle, a transportable filter cabinet (TFC) may be used (Figure€10.9). This is a closed cabinet, either supplied with its own ventilation motor or individually plugged into the central ventilation system. Transportable filter cabinets confer the same advantages as cubicles in relation to reduced spread of infections and allergens. When equipped with their own ventilation motors, TFCs allow for the establishment of animal facilities without expensive investments in central ventilation systems. TFCs may also be used for protecting animals and staff in facilities that are distant from
84550.indb 282
11/11/10 6:25:49 PM
Health Status and Health Monitoring
283
Figure 10.8â•…A color version of this figure follows page 336. Animals placed behind glass walls in so-called cubicles. (University of Washington; photo by Gavin Sisk.)
Figure 10.9â•…A ventilated cabinet for maintenance of laboratory animals (Scanbur, Denmark).
84550.indb 283
11/11/10 6:25:50 PM
284
Handbook of Laboratory Animal Science
Figure 10.10â•…A cage protected by a lid with a filter, a so-called filter top (Tecniplast, Italy).
central facilities—for example, when animals are housed in the research laboratory for a 1-day acute study, thereby minimizing the animals’ exposure to stressful noises or the staff’s exposure to allergens. Filter-Top Cages and Individually Ventilated Cage (IVC) Systems In some cases, it may be practical or economical to place the barrier around the individual cage, rather than the entire unit. This approach may be used inside a barrier-protected unit to enhance the protection of the individual animal or as an alternative to running such a facility. The simplest application is to place a filter-top lid on the cage that only allows the passage of air (Figure€10.10): the so-called microisolator. However, filter tops are not completely sealed to the cage, so animals cannot be considered completely isolated. Microisolators reduce the risks of spreading infections and allergens, but do not eliminate them. Occasionally, concentrations of trace gases in a filtertopped cage may be elevated (Krohn and Hansen 2002), a potential hazard to the animals (Corning and Lipman 1991). In IVC systems (Figure€10.11), each filter-topped cage is supplied with its own ventilation and all cages are microbiologically isolated from one another and from the surroundings. Principles and degrees of sealing depend on the brand of the system. To allow continued separation, cages are changed in a specially equipped laminar airflow system. Some IVC systems may be run at both positive and negative pressure, so they can be used for protecting the animals from their surroundings or protecting the surroundings from the animals. Isolators In some studies, it is necessary to characterize the microflora of the animals used completely. For such studies, animals are kept in isolators (Figure€10.12)—fully closed systems in which the animals are only handled through gloves integrated into and tightly sealed to the isolator wall. In-going air is filtered through an absolute filter and outgoing air is blown through a silicone lock. All materials and diets are introduced only after autoclaving or γ-irradiation. Chemicals, such as potassium monopersulphate, are used to sterilize all surfaces in the isolator lock. Animals are transported from one isolator to another in a closed cylinder that fits tightly with the port of the isolator. The pressure in the isolator is positive to the surroundings.
84550.indb 284
11/11/10 6:25:50 PM
Health Status and Health Monitoring
285
Figure 10.11â•…An individually ventilated cage (IVC) rack (Tecniplast, Italy).
Animals kept in isolators may be germ free or axenic. If supplied with a fully defined flora, in which every microorganism is known, then these animals are referred to as gnotobiotic. Gnotobiotic may be used to refer to germ-free animals as well. Isolators are necessary for certain types of studies. It may be essential in microbiological and nutritional investigations to know the role of specific microorganisms. This can be achieved by using germ-free animals monoinfected with the organism of interest. This will allow the separation of physiological characteristics into germ-free-associated characteristics (GAC) or microfloraassociated characteristics (MAC), factors related to the animal or the microorganism, respectively (Gustafsson and Norin 1977). Isolators are commonly used in cancer research, where strong
Figure 10.12â•…A flexible film isolator for laboratory animals (Isotec, UK).
84550.indb 285
11/11/10 6:25:51 PM
286
Handbook of Laboratory Animal Science
pharmacological immunosuppressors are administered, potentially resulting in problems related to opportunistic pathogens. Isolators are also essential for rederivation of laboratory animals. Health Monitoring Scope Animals maintained as microbiologically defined according to the rederivation and containment principles described previously should be regularly monitored to confirm this status. Therefore, a number of animals are sampled from the colony at frequent intervals and subjected to a range of tests (Figure€10.13). This practice is called health monitoring, although microbiological monitoring would be a more appropriate term. Because quite a number of examinations are necessary for health monitoring, the procedures will normally be performed on animals sampled only for this purpose, and the observed status will be used as a picture of the population rather than of the individual. For instance, a mouse colony may be infected with Helicobacter hepaticus, but mice from this colony may or may not have hepatic changes. An individual animal used in a research project may not harbor the target agent; thus, health monitoring is based upon the principle that a few animals can be sampled for examination, and the results can be used to describe the risks facing the entire colony. Therefore, if one animal is infected with a certain organism, then the entire colony is considered to be infected with that particular organism. Conversely, if the infection is not found in any of the animals sampled, then the entire colony is considered free of that organism. It is difficult, but essential, to define the microbiological entity (e.g., one isolator, one individually vented cage, or one room within the barrier) for which a surveillance sample is predictive. Some bacteria easily spread from one room to another, while others do not; it would be safest, but also impossible, to define each cage as a microbiological entity. Therefore, individual judgments are necessary. Lnn cervicales supf./axill./inguin. Gl. submaxillaris (only rats) Larynx Lungs Serum Heart Serology (ELISA, IFA, HAI) Pelt & skin Macroscopic Thymus pathology Spleen Inspection under microscope for ectoparasites Gastrointestinal syst. Kidneys Adrenals Uterus/Testes Nose Nasal cavity Cav. oris Bacteriology Genitalia Bacteriology Faeces Flotation PCR
Ileum Microscopy
Trachea Bacteriology
Caecum Bacteriology Microscopy
Figure 10.13â•…A color version of this figure follows page 336. Examples of tests performed on each sampled animal in health (microbiological) monitoring of rodents.
84550.indb 286
11/11/10 6:25:52 PM
Health Status and Health Monitoring
287
When an appropriate number of sentinel animals have been identified, sampled, examined, and assayed, the presence or absence of a list of important agents is noted, and each animal’s microbiological status is specified. Most health monitoring programs have been developed by commercial breeders, without whom this process would have been much slower. However, because reports from different breeders were difficult to compare, in the late 1980s, the need for some standardization of health monitoring programs became apparent. To fulfill this need, the Federation of European Laboratory Animal Science Associations (FELASA) issued guidelines for health monitoring of various species: rodents and rabbits under breeding as well as experimental conditions (Nicklas et al. 2002) and pigs, dogs and cats (Rehbinder et al. 1998), primates (Weber et al. 1999), and ruminants (Rehbinder et al. 2000). These papers set standards that define the agents for which to test, the methods to use, the number of animals to test, the frequency of testing, and the format for reporting results. Agents and methods recommended for rodents and rabbits are shown in Table€10.4. Some of these recommendations are very pragmatic. If more thorough health monitoring is required, some of the principles described next may be used in addition to the FELASA recommendations. Sampling Strategies The number of animals to sample, the sampling frequency, the agents to look for, and the analysis methods to use may be based upon scientific judgment, rather than strict obedience to international guidelines. The prevalence reached by a certain infection in an animal colony depends on many factors, including contact between animals, the resistance of the animals, and the characteristics of the agent itself. The sample size needed to detect a specific agent effectively in a colony therefore depends upon the acceptable risk of a false negative result, the sensitivity of the applied method, and the estimated minimum prevalence of the infection, if it were present in the colony (Table€10.5) (Hansen 1993). Once a sample has been taken, it becomes historical and only curiosity will dictate when to take the next sample. The result at one sampling time point elucidates whether changes have occurred since the previous sampling. Choice of Method Some microorganisms are readily cultivated from easily accessible sites of the animal, while others can only be cultivated if a range of specific conditions are fulfilled. A few rodent organisms (e.g., Spirillum minus, the cause of the Japanese variant of rat bite fever, sodoku) cannot even be grown in vitro. Therefore, a complete health monitoring program will necessarily consist of a variety of different types of assays. Enteric helminths and protozoans may be diagnosed by direct microscopy after a flotation test, on smears from the caecum and ileum, or on tape used for sampling around the anus. Inspection under a stereomicroscope may reveal ectoparasites. A range of organs should be inspected macroscopically to decide whether further investigation by histopathology is needed. Histopathological examination is seldom the method of choice for routine health monitoring and is more efficient as a tool for a final diagnosis and validation of the impacts of a certain microorganism. Cultivation is the diagnostic method of choice for microorganisms (bacteria and fungi) readily grown on artificial media. Samples for bacteriological cultivation are typically taken from organs such as the nose, trachea, genitals, liver, and caecum, with inoculation on both selective and nonselective media. Further procedures are then designed to grow pure cultures and to identify the exact target organisms (Hansen 1999).
84550.indb 287
11/11/10 6:25:52 PM
288
Handbook of Laboratory Animal Science
Table€10.4â•…Agents and Methods Recommended for Inclusion in Health Monitoring Programs According to FELASA Guidelines for Health Monitoring Species to Be Tested Viruses Coronaviruses Ectromeliaavirus Guinea pig adenovirus Guinea pig cytomegalovirus Hantaviruses Lymphocytic choriomeningitis virus Parvoviruses Pneumonia virus of mice Rabbit hemorrhagic disease virus Rabbit pox viruses Rabbit rotavirus Reovirus type 3 Sendai virus Theiler’s mouse encehalomyelitis virus Bacteria and fungi Bordetella bronchiseptica Chlamydia psitacci Citrobacter rodentium Clostridium piliforme Corynebacterium kutscheri Dermatophytes Helicobacter spp. Mycoplasmae spp. Pasteurellaceae Salmonellae Streptobacillus moniliformis Streptococci (β-hemolytic) Streptococcus pneumoniae Parasites Ectoparasites Endoparasites Encephalitozoon cuniculi
Test Method Serology Serology Serology Serology Serology Serology Serology Serology Serology Serology Serology Serology Serology Serology Cultivation No recommendation Cultivation Pathology, serology Cultivation Cultivation Cultivation, PCR Serology Cultivation Cultivation Cultivation Cultivation Cultivation Microscopy Microscopy Serology
Mice
Rats
+ +
+
Guinea Pigs
Hamsters
Rabbits
+ + + + + +
+ +
+ +
+ + + + + +
+ + + +
+ + + + + + +
+
+ +
+ +
+ +
+ + + + + + +
+ +
+ +
+
+
+
+
+ + + + +
+ +
+ +
+ + +
+ +
+ +
+
+ + +
For microorganisms that are not easily cultivated, serological assays, such as immunofluorescence assay (IFA), enzyme-linked immunosorbent assay (ELISA), and western immunoblotting are widely used. However, western immunoblotting is laborious and is therefore not recommended for routine use in the FELASA guidelines (Nicklas et al. 2002). Serology is applied for all viruses except for lactate dehydrogenate (LDH) elevating virus, which is simply monitored by testing for LDH activity. A few bacteria, such as Clostridium piliforme (Hansen et al. 1994), and protozoans, such as Encephalitozoon cuniculi (Boot et al. 2000), are screened by serological methods. Some organisms, like Mycoplasma spp., may be diagnosed by both serological and culture methods (Cassell et al. 1981a). Parvoviruses are typically diagnosed by a combination of solid-phase assays, such as ELISA or IFA, to detect common parvoviral antigens
84550.indb 288
11/11/10 6:25:52 PM
Health Status and Health Monitoring
289
Table€10.5╅Calculation of Sample Sizes for Health Monitoring of Laboratory Animals Nosografic Sensitivity (N1) (n1) =
infected animals reacting in the assay infected an nimals whether reacting or not
The estimated prevalence (p) in the colony p=
infected animals total number of animals
The risk of a false negative result in an infected colony (C) C=
number of infected colonies tested with a negative ressult total number of infected colonies tested
Sample size (S) for colonies with more than 1,000 animals s≥
log C log(1− (p * n1))
Sample size (S) for colonies with less than 1,000 animals S ≥ (1 – C1/D) * (T-((D-1)/2)) D = number of infected animals; T = total number of animals
and hemagglutination inhibition assays to differentiate between the different types (Riley et al. 1996a). Serological results are historical and do not provide any evidence concerning whether the animal actually harbors the particular microorganism. The main pitfalls of serological assays are lack of sensitivity due to a pure antibody response to the infection or low specificity due to cross-reactions. The latter is especially the case for bacterial serology as bacteria contains a high number of epitopes and many of those, which induce a high level of antibodies, are shared between a range of different bacteria (Jensen et al. 1985; Moreno et al. 2010). Viruses are simpler, with only few epitopes, and it is easier to base an assay on an epitope and at least limit cross-reactivity to group level (Tischler et al. 2008; Hsu et al. 2006). Molecular biological techniques represent another attractive choice for microorganisms that cannot be cultivated. These methods are generally divided into two types: those based upon DNA (and occasionally RNA) already being present in detectable amounts (in situ hybridization and southern blotting) and those in which the DNA or RNA must be amplified before detection (PCR). Helicobacter hepaticus is an example for which PCR may be the method of choice (Riley et al. 1996a). Microchip arrays may be applied for PCR screening for multiple organisms (C. piliforme, H. bilis, H. hepaticus, and mouse hepatitis virus) from the same fecal sample (Goto et al. 2007). Figure€10.13 shows a very typical range of methods applied for health monitoring of rodents. Reporting When all examinations have been performed and the results are known, a report is issued (Table€10.6). These reports are made available by vendors, via the Internet, to all purchasers of the animals (Table€10.7). No animals should be allowed into an animal facility without careful study of health reports prior to their arrival. Veterinarians at experimental facilities should make the results of the most current screening procedures of the animals in experiments available to animal users.
84550.indb 289
11/11/10 6:25:53 PM
290
Handbook of Laboratory Animal Science
Table€10.6╅An Example of a Health Monitoring Report from a Commercial Vendor FELASA-Approved Health Monitoring Report Name and address of the breeder: The Breeding Company Ltd. Date of issue: May 22 2001 Unit No: 100-12 Latest test date: 08-04-01 Rederivation: 1999 Protocol No: 000397-000398 Species: Mice Strains: C57Bl/6J//XXX-pmn,B10M/XXX, XXX:NMRI, DBA1/J/XXX
Historical Results
Latest Test Results
Laboratory
Method
Viral Infections Minute virus of mice Mouse hepatitis virus Pneumonia virus of mice Reovirus type 3 Sendai virus Theiler’s encephalomyelitis virus Ectromelia virus Hantaviruses Lymphocytic choriomeningitis virus Lactic dehydrogenase virus
Neg. Neg. Neg. Neg. Neg. Neg. Neg. Neg. Neg. N.T.
0/10 0/10 0/10 0/10 0/10 0/10 N.T. N.T. 0/10 N.T.
Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd.
ELISA ELISA ELISA ELISA ELISA ELISA IFA IFA IFA
Bacterial and Fungal Infections Bordetella bronchiseptica Campylobacter spp. Citrobacter rodentium Clostridium piliforme Corynebacterium kutscheri Helicobacter spp. Leptospira spp. Mycoplasma spp. Pasteurella spp. Pasteurellaceae Pasteurella pneumotropica Other Pasteurella spp. Salmonellae Streptobacillus moniliformis β-Hemolytic streptococci Streptococcus pneumoniae Other species associated with lesions: none
Neg. Neg. Neg. Neg. Neg. Neg. N.T. Neg.
0/10 0/10 0/10 0/10 0/10 0/10 N.T. 0/10
Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd.
Culture Culture Culture ELISA Culture Culture
Lab. Ltd.
ELISA
Pos. Pos. Neg. Neg. Neg. Neg. Neg.
8/10 4/10 0/10 0/10 0/10 0/10 0/10
Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd.
ELISA Culture Culture Culture Culture Culture Culture
Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd. Lab. Ltd.
Inspection Flotation Flotation Microscopy Microscopy Microscopy
Parasitological Infections Arthropods Helminths Eimeria spp. Giardia spp. Spironucleus spp. Other flagellates Klossiella spp. Encephalitozoon cuniculi Toxoplasma gondii Pathological lesions observed: Stock: XXX:NMRI Lesions: None
Neg. Neg. Neg. Neg. Neg. Neg. N.T. N.T. N.T.
0/10 0/10 0/10 0/10 0/10 0/10 N.T. N.T. N.T.
Abbreviations for laboratories Lab. Ltd.:╇ Laboratory Company Ltd, X Boulevard, X-Town, X-Country Pos.:╇ Positive results previously observed Neg.:╇ Positive results never observed 0/10:╇No positives out of 10 samples N.T.:╇Not tested
84550.indb 290
11/11/10 6:25:54 PM
Health Status and Health Monitoring
291
Table€10.7â•…Internet Addresses at Which Health Monitoring Reports May Be Found Breeder B&K Universal Charles River Laboratories Clea Japan Ellegaard Göttingen Minipigs Harlan Jackson Laboratories RCC Taconic
Origin
URL
Worldwide Worldwide Japan Denmark and U.S.A. Worldwide U.S.A. Switzerland U.S.A./Denmark
www.bku.com www.criver.com www.clea-japan.co.jp www.minipigs.dk www.harlan.com www.jax.org www.rcc.ch www.taconic.com
Sentinels It is often not possible to sample experimental animals directly for health monitoring purposes. Sentinel animals—those placed in the same unit as the experimental animals solely for health monitoring purposes—can be used instead (Hansen and Jensen 1995). Sentinels are sampled and replaced at regular intervals and are subjected to a range of tests in a manner similar to those used in breeding facilities. Sentinel animals should be as similar (strain, age, sex, etc.) to experimental animals as possible. Certain precautions in relation to age and genetics of the sentinels may have to be considered specifically for the infections to be monitored as the sentinels should have the immunological capacity to raise antibodies against the infection if they are to be used for serology (Hansen et al. 1994). Sentinels should be placed on the lower shelves of racks in cages that contain some dirty bedding from the other animals in the room. This may spread some existing infections to the sentinels. It appears to be effective for mouse hepatitis virus (Homberger and Thomann 1994), but unsafe for Sendai virus (Artwohl et al. 1994). Because some infectious agents are easily found by PCR on fecal samples, it might be considered more reliable to do such testing on feces from animals in experiments rather than testing sentinels (Manuel et al. 2008). Characterizing the Normal Gut Microbiota As described before, the normal gut microbiota can be the cause of essential variation in animal models. In spite of this, commercial breeders so far do not make major efforts to control and monitor this flora, but it is reasonable to assume that this will be a claim from their customers in the future. Only 10–20 of these bacterial species can be cultivated (Hansen et al. 2006), so monitoring might be, for example, by PCR to propagate species-specific bacterial rRNA sequences, which are then separated by specific electrophoresis techniques based on a temperature gradient (TGGE) or a denaturization gradient (DGGE) (Fushuku and Fukuda 2008a, 2008b). Profiles may then be cluster analyzed to provide a semiquantitative and semiqualitative picture of the flora and to show percentage similarities between the individuals (Figure€10.14). Gas chromatography is an easily accessible tool that differentiates bacterial species based on profiles of bacterial cellular fatty acids (CFAs). If the profile of a bacterial species is known, the bacteria will be identified, and any undefined species will also be included in the total profile (Vaahtovuo et al. 2001). Gas chromatography shows a rather uniform gut microbiotic profile for inbred mice—a profile that is mostly dependent on age and strain and is fairly independent of gender (Vaahtovuo et al. 2001). Profiles of mice of an outbred stock are typically 65–70% similar, while those of an inbred strain are 75–80% similar (Hufeldt et al. 2010). The most detailed approach to characterizing the normal gut microbiota is to use pyrosequencing (Zhang et al. 2010).
84550.indb 291
11/11/10 6:25:54 PM
292
Handbook of Laboratory Animal Science
Figure 10.14â•…A DGGE profile from mice. Each band represents specific genomic codes of the bacteria. Each profile is arranged in the cluster analysis with those individuals with which it has highest similarity. Some individuals have a similarity of around 70%, while others have a similarity of up to 90%. (Courtesy of Majbritt Ravn Hufeldt.)
Screening of Biological Materials As described previously, biological materials, such as cells and sera, are as likely as or more likely than the animals to introduce hazardous infections, such as lactic dehydrogenase virus, Pasteurella pneumotropica, and Mycoplasma spp. (Nakai et al. 2000) into an animal facility. There are three basic methods for protecting the facility against this risk. The first method is to allow studies including such materials to take place only in negative pressure isolators. The second is to allow the use of materials derived directly from facilities in which the microbiological status is clearly defined, as described in the preceding sections. This reduces the risk to a level comparable to introducing animals from that facility. The third method is to test all biological materials before allowing them into the animal facility. Viruses impose the highest risk and, although it is not possible to test biological materials for viruses by simple culture, PCR is applicable for most laboratory animal viruses (Homberger et al. 1991). A combination of freezing and serial in vivo passages may be used to eliminate the contamination (Dagnaes-Hansen and Horsman 2005). Quarantine Housing of Animals Imported from Nonvendor Sources If the supplier of animals for an experimental facility is unable to supply a health report or the supplier’s containment facilities are regarded as unsatisfactory, then animals should not be introduced directly into an experimental facility. This issue has increased in prominence with the increased use of transgenic animals. Many of these animals, primarily mice, are not produced by commercial breeders, but rather by a range of different research groups, who regularly engage in the global exchange of these animals.
84550.indb 292
11/11/10 6:25:55 PM
Health Status and Health Monitoring
293
University animal facilities typically receive requests from their animal users to accept transgenic mice from different sources, often with health statuses that are difficult to assess empirically. Since these transgenic animals are often very costly and difficult to get for restocking if accidentally lost, animal facilities must take precautions to reduce the risk of infections in their resident animals, while still permitting the intake of mice from different sources. This is most efficiently done by breeding animals on site in a barrier-protected unit into which animals are only introduced by rederivation. To include smaller numbers of animals into experiments rapidly, it may be necessary to establish a facility in which it is possible to quarantine animals before entry into the main experimental facility. Animals should be maintained for at least 4 weeks in quarantine before health monitoring is performed to define the status of the animals. Four weeks of quarantine may not be sufficient if sentinels are used. If the health monitoring process determines that unwanted microorganisms are absent, then the animals may be transferred to the experimental facility. In larger institutional units, animals may be delivered on an as-needed basis and therefore each delivery must be kept separate inside the quarantine facility. This makes the use of negative pressure isolators or IVC systems the most appropriate choice for quarantine housing. A single sample for health monitoring can never be considered 100% safe; therefore, it may be necessary for units dealing with many minor deliveries from different sources to rederive all delivered animals routinely in the institution’s own barrier breeding facility. Animals delivered for individual projects in the experimental facilities would then only come from the breeding stock inside the unit. This may prove impossible for all animals, so, in practice, it may become necessary to differentiate between safe and unsafe deliveries. Animals from known commercial breeders, with high standard barrier facilities and efficient health monitoring procedures, may be considered safe deliveries and therefore allowed more direct access to the experimental facility through the quarantine unit. All animal deliveries from other sources should be regarded as unsafe and be subject to rederivation. This method does not take into consideration the possibility that animals might get infected during transport. Another security measure involves the division of the experimental unit into small separate units, protected from one another in ways that ensure that each can be regarded as its own microbiological entity. For instance, each separate unit would have its own barrier and staff. Inside each separation, all animals would be derived from the same source. Each institution must independently judge which system to use based on a comprehensive analysis of the costs and benefits of the protective measures and the research requirements.
Final Remarks In the first part of this chapter, a range of infections and the problems to which they may lead were described. This may give rise to the thought that it is almost impossible to avoid the fact that infections disturb one’s experiments. However, this is actually an area in which the laboratory animal community has had quite a success. By following the recommendations given in the last part of this chapter, a range of infections have been eliminated from research facilities and, for instance, some viruses, which used to be common problems for scientists, are no longer found frequently in Europe, Japan, and North America. Improvement of facility standards and health monitoring programs has enabled research to be performed without constantly being under the risk of invalidating infections in the animals. However, it is still a must that researchers, laboratory animal scientists, and animal caretakers feel a daily responsibility to keep infections out of their laboratory animal units.
84550.indb 293
11/11/10 6:25:55 PM
294
Handbook of Laboratory Animal Science
References Abzug, M. J., and R. W. Tyson. 2000. Picornavirus infection in early murine gestation: Significance of maternal illness. Placenta 21:840–846. al-Lebban, Z. S. et al. 1988. Rabbit pasteurellosis: Induced disease and vaccination. American Journal of Veterinary Research 49:312–316. Allen, A. M. et al. 1981. Tyzzer’s disease syndrome in laboratory rabbits. American Journal of Pathology 46:859–882. Amano, H. et al. 1991. Isolation of Corynebacterium kutscheri from aged Syrian hamsters (Mesocricetus auratus). Laboratory Animal Science 41:265–268. Amao, H. et al. 1995a. Natural and subclinical Corynebacterium kutscheri infection in rats. Laboratory Animal Science 45:11–14. ———. 1995b. Natural habitats of Corynebacterium kutscheri in subclinically infected ICGN and DBA/2 strains of mice. Laboratory Animal Science 45:6–10. Artwohl, J. E. et al. 1994. The efficacy of a dirty bedding sentinel system for detecting Sendai virus infection in mice: A comparison of clinical signs and seroconversion. Laboratory Animal Science 44:73–75. Ashley, M. P. et al. 1976. Local and systemic effects in the nonspecific tumor resistance induced by attenuated Salmonella enteritidis 11RX in mice. Australian Journal of Experimental Biology and Medical Science 54:157–168. Backhed, F. et al. 2004. The gut microbiota as an environmental factor that regulates fat storage. Proceedings of the National Academy of Sciences USA 101:15718–15723. Baker, D. G. 1998. Natural pathogens of laboratory mice, rats, and rabbits and their effects on research. Clinical Microbiology Reviews 11:231–266. Ball-Goodrich, L. J., and E. Johnson. 1994. Molecular characterization of a newly recognized mouse parvovirus. Journal of Virology 68:6476–6486. Barthold, S. W., and D. G. Brownstein. 1988. The effect of selected viruses on Corynebacterium kutscheri infection in rats. Laboratory Animal Science 38:580–583. Barthold, S. W., and A. M. Jonas. 1977. Morphogenesis of early 1, 2-dimethylhydrazine-induced lesions and latent period reduction of colon carcinogenesis in mice by a variant of Citrobacter freundii. Cancer Research 37:4352–4360. Barthold, S. W., and A. L. Smith. 1984. Mouse hepatitis virus strain-related patterns of tissue tropism in suckling mice. Archives of Virology 81:103–112. Barthold, S. W. et al. 1976. The etiology of transmissible murine colonic hyperplasia. Laboratory Animal Science 26:889–894. ———. 1977. Dietary, bacterial, and host genetic interactions in the pathogenesis of transmissible murine colonic hyperplasia. Laboratory Animal Science 27:938–945. Besselsen, D. G. et al. 2006. Identification of novel murine parvovirus strains by epidemiological analysis of naturally infected mice. Journal of General Virology 87:1543–1556. ———. 2008. Transmission probabilities of mouse parvovirus 1 to sentinel mice chronically exposed to serial dilutions of contaminated bedding. Comparative Medicine 58:140–144. Bhan, A. K. et al. 1999. Colitis in transgenic and knockout animals as models of human inflammatory bowel disease. Immunological Reviews 169:195–207. Bhatt, P. N., and R. O. Jacoby. 1987a. Effect of vaccination on the clinical response, pathogenesis and transmission of mousepox. Laboratory Animal Science 37:610–614. ———. 1987b. Mousepox in inbred mice innately resistant or susceptible to lethal infection with ectromelia virus. III. Experimental transmission of infection and derivation of virus-free progeny from previously infected dams. Laboratory Animal Science 37:23–27. Biggar, R. J. et al. 1975a. Letter: Lymphocytic choriomeningitis associated with hamsters. Lancet 1:856–857. ———. 1975b. Lymphocytic choriomeningitis outbreak associated with pet hamsters. Fifty-seven cases from New York State. JAMA 232:494–500. Blackmore, D. K., and S. Cassillo. 1972. Experimental investigation of uterine infections of mice due to Pasteurella pneumotropica. Journal of Comparative Pathology 82:471. Boot, R. et al. 1988. Serological evidence for Encephalitozoon cuniculi infection (nosemiasis) in gnotobiotic guineapigs. Laboratory Animals 22:337–342.
84550.indb 294
11/11/10 6:25:55 PM
Health Status and Health Monitoring
295
———. 2000. Comparison of assays for antibodies to Encephalitozoon cuniculi in rabbits. Laboratory Animals 34:281–289. Brammer, D. W. et al. 1993. Elimination of sialodacryoadenitis virus from a rat production colony by using seropositive breeding animals. Laboratory Animal Science 43:633–634. Breban, M. A. et al. 1993. Influence of the bacterial-flora on collagen-induced arthritis in susceptible and resistant strains of rats. Clinical and Experimental Rheumatology 11:61–64. Breinholt, V. et al. 2000. Estrogenic activity of flavonoids in mice. The importance of estrogen receptor distribution, metabolism and bioavailability. Food and Chemical Toxicology 38:555–564. Brennan, P. C. et al. 1969. Role of Pasteurella pneumotropica and Mycoplasma pulmonis in murine pneumonia. Journal of Bacteriology 97:337–349. Broderson, J. R. et al. 1976a. Lethal enteritis in infant mice caused by mouse hepatitis virus. Laboratory Animal Science 26:824. ———. 1976b. The role of environmental ammonia in respiratory mycoplasmosis of rats. American Journal of Pathology 85:115–130. Brown, M. B., and D. A. Steiner. 1996. Experimental genital mycoplasmosis: Time of infection influences pregnancy outcome. Infection and Immunity 64:2315–2321. Buschard, K. et al. 1983. Virus-induced diabetes mellitus in mice and the thymus-dependent immune system. Diabetologia 24:42–46. Butz, N. et al. 1999. Pathogenesis of guinea pig adenovirus infection. Laboratory Animal Science 49:600–604. Caniatti, M. et al. 1998. Cilia-associated respiratory (CAR) bacillus infection in conventionally reared rabbits. Zentralblatt für Veterinarmedizin [B] 45:363–371. Capucci, L. et al. 1996. Detection and preliminary characterization of a new rabbit calicivirus related to rabbit hemorrhagic disease virus but nonpathogenic. Journal of Virology 70:8614–8623. Carthew, P., and S. Sparrow. 1980a. Persistence of pneumonia virus of mice and Sendai virus in germfree (nu/nu) mice. British Journal of Experimental Pathology 61:172–175. ———. 1980b. Sendai virus in nude and germ-free rats. Research in Veterinary Science 29:289–292. Casebolt, D. B. et al. 1997. Detection of enterotropic mouse hepatitis virus fecal excretion by polymerase chain reaction. Laboratory Animal Science 47:6–10. Cassell, G. H., and J. K. Davis. 1978. Protective effect of vaccination against Mycoplasma pulmonis respiratory disease in rats. Infection and Immunity 21:69–75. Cassell, G. H. et al. 1981a. Adherence and colonization of Mycoplasma pulmonis to genital epithelium and spermatozoa in rats. Israel Journal of Medical Science 17:593–598. ———. 1981b. Control of Mycoplasma pulmonis infection in rats and mice: Detection and elimination vs. vaccination. Israel Journal of Medical Science 17:674–677. ———. 1981c. Respiratory and genital mycoplasmosis of laboratory rodents: Implications for biomedical research. Israel Journal of Medical Science 17:548–554. Chatterjee, A. et al. 2001. Modification of maternal and congenital cytomegalovirus infection by anti-glycoprotein B antibody transfer in guinea pigs. Journal of Infectious Diseases 183:1547–1553. Chen, Z. Y., and P. G. W. Plagemann. 1997. Detection of lactate dehydrogenase-elevating virus in transplantable mouse tumors by biological assay and RT-PCR assays and its removal from the tumor cell. Journal of Virological Methods 65:227–236. Chin, E. Y. et al. 2000. Helicobacter hepaticus infection triggers inflammatory bowel disease in T-cell receptor alpha beta mutant mice. Comparative Medicine 50:586–594. Christensen, H., and M. Bisgaard. 2010. Molecular classification and its impact on diagnostics and understanding the phylogeny and epidemiology of selected members of Pasteurellaceae of veterinary importance. Berliner und Münchener Tierarztliche Wochenschrift 123:20–30. Clausen, F. B., J. Lykkesfeldt, and A. K. Hansen. 2001. On the possible connection between vitamin C, cardiovirus and guinea pig lameness. In Animal research and welfare. A partnership. Proceedings of the International Joint Meeting, ed. J. Tur-Mari et al. Twelfth ICLAS General Assembly and Conference. Seventh FELASA Symposium. London: Laboratory Animals Ltd., 101–124. Corning, B. F., and N. S. Lipman. 1991. A comparison of rodent caging systems based on microenvironmental parameters. Laboratory Animal Science 41:498–503. Cox, J. C. et al. 1977. Application of immunofluorescence to the establishment of an Encephalitozoon cuniculifree rabbit colony. Laboratory Animal Science 27:204–209.
84550.indb 295
11/11/10 6:25:55 PM
296
Handbook of Laboratory Animal Science
Crawford, L. V. 1966. A minute virus of mice. Virology 29:605–612. Crispens, C. G., Jr. 1967. Lactate dehydrogenase virus and mouse embryos. Nature 214:819. Cross, S. S. et al. 1979. Biology of mouse thymic virus, a herpes virus of mice, and the antigenic relationship to mouse cytomegalovirus. Infection and Immunity 26:1186–1195. Cundiff, D. D. et al. 1995. Characterization of cilia-associated respiratory bacillus in rabbits and analysis of the 16S rRNA gene sequence. Laboratory Animal Science 45:22–26. Cunningham, B. B. et al. 1998. Rat bite fever in a pet lover. Journal of the American Academy of Dermatologists 38:330–332. Dagnaes-Hansen, F., and M. R. Horsman. 2005. Experience with mouse hepatitis virus sanitation in three transplantable murine tumor lines. Laboratory Animals 39:394–399. Dahl, K. et al. 2006. Glucose intolerance in a xenotransplantation model: Studies in alpha-gal knockout mice. Acta Pathologica Microbiologica et Immunologica Scandinavica 114:805–811. Deeb, B. J., and R. F. DiGiacomo. 2000. Respiratory diseases of rabbits. Veterinary Clinics of North America Exotic Animal Practice 3:465. Deibel, R. et al. 1975. Lymphocytic choriomeningitis virus in man. Serologic evidence of association with pet hamsters. JAMA 232:501–504. Desmyter, J. et al. 1983. Laboratory rat associated outbreak of hemorrhagic fever with renal syndrome due to Hantaan-like virus in Belgium. Lancet 2:1445–1448. Detmer, A. et al. 1991. Xylose-positive staphylococci as a cause of respiratory disease in immunosuppressed rats. Scandinavian Journal of Laboratory Animal Science 18:13–18. Dewhirst, F. E. et al. 1999. Phylogeny of the defined murine microbiota: Altered Schaedler flora. Applied and Environmental Microbiology 65:3287–3292. DiBaise, J. K. et al. 2008. Gut microbiota and its possible relationship with obesity. Mayo Clinic Proceedings 83:460–469. Dick, E. J. et al. 1996. Mousepox outbreak in a laboratory mouse colony. Laboratory Animal Science 46:602–611. DiGiacomo, R. F. et al. 1987. Transmission of Pasteurella multocida in rabbits. Laboratory Animal Science 37:621–623. ———. 1990. Naturally acquired Pasteurella multocida infection in rabbits: Immunological aspects. Laboratory Animal Science 40:289–292. ———. 1991a. Naturally acquired Pasteurella multocida infection in rabbits: Clinicopathological aspects. Canadian Journal of Veterinary Research 55:234–238. ———. 1991b. Naturally acquired Pasteurella multocida subsp. multocida infection in a closed colony of rabbits: Characteristics of isolates. Laboratory Animals 25:236–241. Dillehay, D. L. et al. 1991. Pathogenicity of Pasteurella multocida A:3 in Flemish giant and New Zealand white rabbits. Laboratory Animals 25:337–341. Donnio, P. Y. et al. 1994. Pasteurella-multocida-oropharyngeal carriage and antibody-response in breeders. Veterinary Research 25:8–15. Duga, S. et al. 1998. Analysis of the 16S rRNA gene sequence of the coryneform bacterium associated with hyperkeratotic dermatitis of athymic nude mice and development of a PCR-based detection assay. Molecular and Cellular Probes 12:191–199. Dutko, F. J., and M. B. Oldstone. 1983. Genomic and biological variation among commonly used lymphocytic choriomeningitis virus strains. Journal of General Virology 64 (Pt 8): 1689–1698. Easterbrook, J. D. et al. 2008. A survey of rodent-borne pathogens carried by wild-caught Norway rats: A potential threat to laboratory rodent colonies. Laboratory Animals 42:92–98. El-Gayar, M. et al. 2008. One-step dilution of open-pulled-straw (OPS)-vitrified mouse blastocysts in sucrosefree medium. Cryobiology 57:191–194. Flatt, R. E., and S. J. Jackson. 1970. Renal nosematosis in young rabbits. Veterinary Pathology 7:492–497. Fleming M. P. 1976. Streptobacillus moniliformis isolations from cervical abscesses of guinea pigs. Veterinary Record 99:256. Fox, J. G., L. Anderson, F. Loew, and F. W. Quimby. 2002. Laboratory animal medicine. San Diego, CA: Academic Press. Fox, J. G., and A. Lee. 1997. The role of Helicobacter species in newly recognized gastrointestinal tract diseases of animals. Laboratory Animal Science 47:222–255. Friis, A. S. 1978. Demonstration of antibodies to Bacillus piliformis in SPF colonies and experimental transplacental infection by Bacillus piliformis in mice. Laboratory Animals 12:23–26.
84550.indb 296
11/11/10 6:25:56 PM
Health Status and Health Monitoring
297
———. 1979. Studies on Tyzzer’s disease: Transplacental transmission by Bacillus piliformis in rats. Laboratory Animals 13:43–46. Friis, A. S., and O. Ladefoged. 1979. The influence of Bacillus piliformis (Tyzzer) infections on the reliability of pharmacokinetic experiments in mice. Laboratory Animals 13:257–261. Fujiwara, K. et al. 1983. Antigenic relatedness of Bacillus piliformis (Tyzzer) from different sources. Laboratory Animal Science 33:490. ———. 1985. Antigenic relatedness of Bacillus piliformis from Tyzzer’s disease occurring in Japan and other regions. Japanese Journal of Veterinary Science 47:9–16. Fushuku, S. and K. Fukuda. 2008a. Gender difference in the composition of fecal flora in laboratory mice, as detected by denaturing gradient gel electrophoresis (DGGE). Experimental Animals 57:489–493. ———. 2008b. Inhomogeneity of fecal flora in separately reared laboratory mice, as detected by denaturing gradient gel electrophoresis (DGGE). Experimental Animals 57:95–99. Gannon, J. 1980. A survey of Encephalitozoon cuniculi in laboratory animal colonies in the United Kingdom. Lab Anim 14:91–94. Gobbi, A. et al. 1999a. Corynebacterium bovis infection in immunocompetent hirsute mice. Laboratory Animal Science 49:209–211. ———. 1999b. Corynebacterium bovis infection in waltzing mice. Laboratory Animal Science 49:132–133. Goto, K. et al. 2001. Contamination of transplantable human tumor-bearing lines by Helicobacter hepaticus and its elimination. Journal of Clinical Microbiology 39:3703–3704. ———. 2007. Specific and quantitative detection of PCR products from Clostridium piliforme, Helicobacter bilis, H. hepaticus, and mouse hepatitis virus infected mouse samples using a newly developed electrochemical DNA chip. Journal of Microbiological Methods 69:93–99. Gustafsson, B. E. and K. E. Norin. 1977. Development of germfree animal characteristics in conventional rats in antibiotics. Acta Pathologica et Microbiologica Scandinavica B 85B:1–8. Gustafsson, E. et al. 1996. Maternal antibodies protect immunoglobulin deficient neonatal mice from mouse hepatitis virus (MHV)-associated wasting syndrome. American Journal of Reproductive Immunology 36:33–39. Hamilton, J. D. et al. 1979. Effects of murine cytomegalovirus infection on the immune response to a tumor allograft. Reviews of Infect Diseases 1:976–987. Hansen, A. K. 1990. The use of the Mongolian gerbil as sentinel for infection with Bacillus piliformis in laboratory rats. Proceedings of the 4th FELASA symposium. Lyon: Federation of European Laboratory Animal Science Associations, 449–452. ———. 1992. The aerobic bacterial flora of laboratory rats from a Danish breeding centre. Scandinavian Journal of Laboratory Animal Science 19:59–68. ———. 1993. Statistical aspects of health monitoring of laboratory animal colonies. Scandinavian Journal of Laboratory Animal Science 20:11–14. ———. 1995. Antibiotic treatment of nude rats and its impact on the aerobic bacterial flora. Laboratory Animals 29:37–44. ———. 1998. Microbiological quality of laboratory pigs. Scandinavian Journal of Laboratory Animal Science 25:145–152. ———. 1999. Handbook of laboratory animal bacteriology. Boca Raton, FL: CRC Press. Hansen, A. K., and H. J. S. Jensen. 1995. Experience from sentinel health monitoring in units containing rats and mice in experiments. Scandinavian Journal of Laboratory Animal Science 22:1–9. ———. 1997. Mice infected with mouse hepatitis virus shed diagnostically detectable amounts of IgA in feces. Scandinavian Journal of Laboratory Animal Science 24:66–71. Hansen, A. K. et al. 1990. Epidemiological studies of Bacillus piliformis infection and Tyzzer’s disease in laboratory rats. Zeitschrift für Versuchstierkunde 33:163–169. ———. 1992a. Correlation between megaloileitis and antibodies to Bacillus piliformis in laboratory rat colonies. Laboratory Animal Science 42:449–453. ———. 1992b. Rederivation of rat colonies seropositive for Bacillus piliformis and the subsequent screening for antibodies [published erratum appears in Laboratory Animal Science 1993 43 (1): 114]. Laboratory Animal Science 42:444–448. ———. 1994. Studies on the diagnosis of Tyzzer’s disease in laboratory rat colonies with antibodies against Bacillus piliformis (Clostridium piliforme). Laboratory Animal Science 44:424-9.
84550.indb 297
11/11/10 6:25:56 PM
298
Handbook of Laboratory Animal Science
———. 1997. A serological indication of the existence of a guineapig poliovirus. Laboratory Animals 31:212–218. ———. 2000. New infections to be considered in health monitoring of laboratory rodents. Scandinavian Journal of Laboratory Animal Science 27:65–83. ———. 2006. Diabetes preventive gluten-free diet decreases the number of caecal bacteria in nonobese diabetic mice. Diabetes—Metabolism Research & Reviews 22:220–225. ———. 2009. The impact of the postnatal gut microbiota on animal models. Proceedings of the Tenth FELASA Symposium and the XIV ICLAS General Assembly & Conference. Cernobbio, Italy, June 11–14, 2007, 95–99. ———. In press. The impact of the postnatal gut microbiota on animal models. Laboratory Animals. Hayashimoto, N. et al. 2005. Genetic diversity on 16S rDNA sequence and phylogenic tree analysis in Pasteurella pneumotropica strains isolated from laboratory animals. Current Microbiology 51:239–243. ———. 2007. A specific polymerase chain reaction based on the gyrB gene sequence and subsequent restriction fragment length polymorphism analysis of Pasteurella pueumotropica isolates from laboratory mice. Journal of the American Association for Laboratory Animal Science 46:54–58. Heck, F. C., Jr. et al. 1972. Pathogenesis of experimentally produced mouse adenovirus infection in mice. American Journal of Veterinary Research 33:841–846. Heidt, P. J., and C. P. J. Timmermanns. 1975. Selective decontamination of the digestive tracts of pregnant rabbits: A method for producing Enterobacteriaceae-free rabbits. Laboratory Animal Science 25:594–596. Hill, A. C., and G. P. Stalley. 1991. Mycoplasma pulmonis infection with regard to embryo freezing and hysterectomy derivation. Laboratory Animal Science 41:563–566. Hirst, R. G., and M. E. Wallace. 1976. Inherited resistance to Corynebacterium kutscheri in mice. Infection and Immunity 14:475–482. Homberger, F. R. 1996. Mouse hepatitis virus. Schweizer Archiv für Tierheilkunde 138:183–188. ———. 1997. Enterotropic mouse hepatitis virus. Lab Anim 31:97–115. Homberger, F. R., and P. E. Thomann. 1994. Transmission of murine viruses and mycoplasma in laboratory mouse colonies with respect to housing conditions. Lab Anim 28:113–120. Homberger, F. R. et al. 1991. Detection of rodent coronaviruses in tissues and cell cultures by using polymerase chain reaction. Journal of Clinical Microbiology 29:2789–2793. Hopert, A. et al. 1993. Mycoplasma detection by PCR analysis. In Vitro Cellular & Developmental Biology— Animal 29A:819–821. Howard, R. J., and H. H. Balfour, Jr. 1977. Prevention of morbidity and mortality of wild murine cytomegalovirus by vaccination with attenuated cytomegalovirus. Proceedings of the Society for Experimental Biology and Medicine 156:365–368. Hsu, C. C. et al. 2006. Persistent infection with and serologic cross-reactivity of three novel murine noroviruses. Comparative Medicine 56:247–251. Hufeldt, M. R. et al. 2010. Characterizing the variation in the composition of the intestinal microbiota of laboratory mice. Comparative Medicine. Hughes, W. T. 1982. Natural-mode of acquisition for de novo infection with Pneumocystis carinii. Journal of Infectious Diseases 145:842–848. Iida, T. 1972. Experimental study on the transmission of Sendai virus in specific pathogen-free mice. Journal of General Virology 14:69–75. Itoh, T. et al. 1987. Naturally occurring CAR bacillus infection in a laboratory rat colony and epizootiological observations. Jikken Dobutsu 36:387–393. Iwai, H. et al. 1986. Prevalence of antibodies to Sendai virus and rotavirus in laboratory rabbits. Jikken Dobutsu 35:491–494. Iwata, H. et al. 1990. Aerosol vaccination with a Sendai virus temperature-sensitive mutant (HVJ-pB) derived from persistently infected cells. Journal of Infectious Diseases 162:402–407. Jacoby, R. O., and P. N. Bhatt. 1987. Mousepox in inbred mice innately resistant or susceptible to lethal infection with ectromelia virus. II. Pathogenesis. Laboratory Animal Science 37:16–22. Jacoby, R. O., and J. R. Lindsey. 1998. Risks of infection among laboratory rats and mice at major biomedical research institutions. ILAR Journal 266–271. Jacoby, R. O. et al. 1975. Pathogenesis of sialodacryoadenitis in gnotobiotic rats. Veterinary Pathology 12:196–209. Janus, L. M. et al. 2008. Minute virus of mice: Antibody response, viral shedding, and persistence of viral DNA in multiple strains of mice. Comparative Medicine 58:360–368.
84550.indb 298
11/11/10 6:25:56 PM
Health Status and Health Monitoring
299
———. 2009. Risk assessment of minute virus of mice transmission during rederivation: Detection in reproductive organs, gametes, and embryos of mice after in vivo infection. Biology of Reproduction 81:1010–1015. Jensen, H. J. 1971. Elimination of intercurrent death among rabbits inoculated with Treponema pallidum. Acta Pathologica et Microbiologica Scandinavica B: Microbiology and Immunology 79:124–15. Jensen, K. et al. 1985. A serological study with monoclonal antibodies to an antigen common to a wide range of bacteria. FEMS Microbiology Letters 29:129–133. Jin, B. et al. 2010. Equilibrium vitrification of mouse embryos. Biology of Reproduction 82:444–450. Juhr, N. C., and J. Horn. 1975. [Model infection with Corynebacterium kutscheri in the mouse]. Zeitschrift für Versuchstierkunde 17:129–141. Kaklamanis, E., and M. Pavlatos. 1972. The immunosuppressive effect of mycoplasma infection. I. Effect on the humoral and cellular response. Immunology 22:695–702. Karst, S. M. et al. 2003. STAT1-dependent innate immunity to a Norwalk-like virus. Science 299:1575–1578. Kaspareit-Rittinghausen, J. et al. 1990. [Pathological changes in Streptobacillus moniliformis infection of C57bl/6J mice]. Berliner und Münchener Tierarztliche Wochenschrift 103:84–87. Kastenmayer, R. J. et al. 2008. Eradication of murine norovirus from a mouse barrier facility. Journal of the American Association for Laboratory Animal Science 47:26–30. Kawaguchi-Miyashita, M. et al. 2001. An accessory role of TCR gamma delta(+) cells in the exacerbation of inflammatory bowel disease in TCR alpha mutant mice. European Journal of Immunology 31:980–988. Kilham, L., and H. V. Ferm. 1961. Rat virus (RV) infections of pregnant, fetal and newborn rats. Proceedings of the Society of Experimental Biology and Medicine 106:825–829. Kilham, L., and G. Margolis. 1969. Transplacental infection of rats and hamsters induced by oral and parenteral inoculations of H-l and rat viruses (RV). Teratology 2:111–123. Kimbrough, R., and T. B. Gaines. 1966. Toxicity of hexamethylphosphoramide in rats. Nature 211:146–147. Kimura, Y. et al. 1979. Protection of mice against virulent virus infection by a temperature-sensitive mutant derived from an HVJ (Sendai virus) carrier culture. Archives of Virology 61:297–304. Kirchner, B. K. et al. 1992. Isolation of Streptobacillus moniliformis from a guinea pig with granulomatous pneumonia. Laboratory Animal Science 42:519–521. Koller, L. D. 1969. Spontaneous Nosema cuniculi infection in laboratory rabbits. Journal of American Veterinary Medical Association 155:1108–1114. Koopman, J. P. et al. 1991. Isolation of Streptobacillus moniliformis from the middle ear of rats. Laboratory Animals 25:35–39. Kraft, V., and B. Meyer. 1990. Seromonitoring in small laboratory animal colonies. A five year survey: 1984– 1988. Zeitschrift für Versuchstierkunde 33:29–35. Krahenbuhl, J. L. et al. 1972. The use of killed vaccines in immunization against an intracellular parasite: Toxoplasma gondii. Journal of Immunology 108:425–431. Krohn, T. C., and A. K. Hansen. 2002. Carbon dioxide concentrations in unventilated IVC cages. Laboratory Animals 36:209–212. Kuhn, G. et al. 2004. Growth performance, carcass characteristics and bioavailability of isoflavones in pigs fed soy bean based diets. Archives of Animal Nutrition 58:265–276. Kunzel, F., and A. Joachim. 2010. Encephalitozoonosis in rabbits. Parasitology Research 106:299–309. Lai, W. C. et al. 1989. Mycoplasma pulmonis depresses humoral and cell-mediated responses in mice. Laboratory Animal Science 39:11–15. Larsen, S. H., and D. Nathans. 1977. Mouse adenovirus: Growth of plaque-purified FL virus in cell lines and characterization of viral DNA. Virology 82:182–195. Laubach, H. E. et al. 1978. Lung lysophospholipase activity in specific-pathogen-free rats infected with Pasteurella pneumotropica or Mycoplasma pulmonis. Infection and Immunity 22:295–297. LeDuc, J. W. 1987. Epidemiology of Hantaan and related viruses. Laboratory Animal Science 37:413–418. Lee, H. W. et al. 1978. Isolation of the etiologic agent of Korean hemorrhagic fever. Journal of Infectious Diseases 137:298–308. Lipman, N. S. et al. 2000. Mousepox resulting from use of ectromelia virus-contaminated, imported mouse serum. Comparative Medicine 50:426–435. Lipton, H. L., and M. C. Dal Canto. 1976. Chronic neurologic disease in Theiler’s virus infection of SJL/J mice. Journal of Neurological Sciences 30:201–207.
84550.indb 299
11/11/10 6:25:56 PM
300
Handbook of Laboratory Animal Science
Lipton, H. L., and R. Melvold. 1984. Genetic analysis of susceptibility to Theiler’s virus-induced demyelinating disease in mice. Journal of Immunology 132:1821–1825. Loh, L., and J. B. Hudson. 1981. Murine cytomegalovirus infection in the spleen and its relationship to immunosuppression. Infection and Immunity 32:1067–1072. ———. 1982. Murine cytomegalovirus-induced immunosuppression. Infection and Immunity 36:89–95. Loliger, H. C., and S. Matthes. 1989. [Infectious factor diseases in domestic small animals (carnivorous and herbivorous fur animals, wool and meat rabbits)]. Berliner und Münchener Tierarztliche Wochenschrift 102:364–371. Lu, L. J. W., and K. E. Anderson. 1998. Sex and long-term soy diets affect the metabolism and excretion of soy isoflavones in humans. American Journal of Clinical Nutrition 68:1500S–1504S. Lynch, D. W. et al. 1984. Carcinogenic and toxicologic effects of inhaled ethylene oxide and propylene oxide in F344 rats. Toxicology and Applied Pharmacology 76:69–84. Machii, K. et al. 1989. Infection of rabbits with Sendai virus. Laboratory Animal Science 39:334–337. Maggio-Price, L. et al. 1998. Diminished reproduction, failure to thrive, and altered immunologic function in a colony of T-cell receptor transgenic mice: Possible role of Citrobacter rodentium. Laboratory Animal Science 48:145–155. Maizels, R. M. 2009. Exploring the immunology of parasitism—From surface antigens to the hygiene hypothesis. Parasitology 136:1549–1564. Manuel, C. A. et al. 2008. Soiled-bedding sentinel detection of murine norovirus 4. Journal of the American Association for Laboratory Animal Science 47:31–36. Martinez, D. et al. 1979. Viral etiology of age-dependent polioencephalomyelitis in C58 mice. Infection and Immunity 23:133–139. Mathiasen, T., and M. Rix. 1993. [Rat-bite—An infant bitten by a rat]. Ugeskrift for Laeger 155:1475–1476. Matsunaga, Y., and F. Chino. 1981. Experimental infection of young rabbits with rabbit parvovirus. Archives of Virology 68:257–264. Matsunaga, Y. et al. 1977. Isolation and characterization of a parvovirus of rabbits. Infection and Immunity 18:495–500. Mayo, D. et al. 1978. Activation of latent murine cytomegalovirus infection: Cocultivation, cell transfer, and the effect of immunosuppression. Journal of Infectious Disease 138:890–896. McConnell, S. J. et al. 1964. Isolation and characterization of a neurotropic agent from adult rats. Proceedings of the Society for Experimental Biology and Medicine 115:362–367. McKisic, M. D. et al. 1993. Identification and propagation of a putative immunosuppressive orphan parvovirus in cloned T-cells. Journal of Immunology 150:419–428. Merritt, J. C. 2004. Metabolic syndrome: Soybean foods and, serum lipids. Journal of the National Medical Association 96:1032–1041. Metcalf, J. B. et al. 1989. Natural parvovirus infection in laboratory rabbits. American Journal of Veterinary Research 50:1048–1051. Meyer, B. J., and C. S. Schmaljohn. 2000. Persistent hantavirus infections: Characteristics and mechanisms. Trends in Microbiology 8:61–67. Mikazuki, K. et al. 1994. Colonization pattern of Pasteurella pneumotropica in mice with latent pasteurellosis. Experimental Animals 43:375–379. Mims, C. 1986. Virus-related immunomodulation. In Viral and mycoplasmal infections of laboratory rodents, effects on biomedical research, ed. P. N. Bhatt, R. O. Jacoby, H. C. Morse III, and A. E. New. New York: Academic Press Inc. Moler, T. L. et al. 1979. Simple technique for non-surgical embryo transfer in mice. Laboratory Animal Science 29:353–356. Moreno, A. T. et al. 2010. Immunization of mice with single PspA fragments induces antibodies capable of mediating complement deposition on different pneumococcal strains and cross-protection. Clinical and Vaccine Immunology 17:439–446. Mumphrey, S. M. et al. 2007. Murine norovirus 1 infection is associated with histopathological changes in immunocompetent hosts, but clinical disease is prevented by STAT1-dependent interferon responses. Journal of Virology 81:3251–3263. Munoz, I. et al. 1995. Use of follicle-stimulating hormone (Fsh) to increase the in vitro fertilization (Ivf) efficiency of mice. Journal of Assisted Reproduction and Genetics 12:738–743.
84550.indb 300
11/11/10 6:25:56 PM
Health Status and Health Monitoring
301
Nakagawa, M. et al. 1984. Ten years-long survey on pathogen status of mouse and rat breeding colonies. Jikken Dobutsu 33:115–120. ———. 1986. A survey of Streptococcus pneumoniae, Streptococcus zooepidemicus, Salmonella spp., Bordetella bronchiseptica and Sendai virus in guinea pig colonies in Japan. Jikken Dobutsu 35:517–520. Nakai, N. et al. 2000. Detection and elimination of contaminating microorganisms in transplantable tumors and cell lines. Experimental Animals 49:309–313. Nicklas, W. et al. 1993. Contamination of transplantable tumors, cell lines, and monoclonal antibodies with rodent viruses. Laboratory Animal Science 43:296–300. ———. 2002. Recommendations for the health monitoring of rodent and rabbit colonies in breeding and experimental units. Laboratory Animals 36:20–42. Notkins, A. L. 1971. Enzymatic and immunologic alterations in mice infected with lactic dehydrogenase virus. American Journal of Pathology 64:733–746. Ohsawa, K. et al. 1998a. Genetic characterization of parainfluenza virus 3 derived from guinea pigs. Journal of Veterinary Medical Science 60:919–922. Ohsawa, K., Y. Watanabe, H. Miyata, and H. Sato. 1998b. Genetic analysis of TMEV-like virus isolated from rats: Nucleic acid characterization of 3-D protein region. Laboratory Animal Science 48 (4): 418. Okewole, P. A. et al. 1989. Uterine involvement in guinea pig salmonellosis. Laboratory Animals 23:275–277. Osanai, T. et al. 1996. Distribution of Corynebacterium renale among apparently healthy rats. Veterinary Microbiology 52:313–315. Parel, J. D. C. et al. 2008. Characterization of rDNA sequences from Syphacia obvelata, Syphacia muris, and Aspiculuris tetraptera and development of a PCR-based method for identification. Veterinary Parasitology 153:379–383. Parker, J. C. et al. 1976. Lymphocytic choriomeningitis virus infection in fetal, newborn, and young adult Syrian hamsters (Mesocricetus auratus). Infection and Immunity 13:967–981. ———. 1978. Susceptibility of inbred and outbred mouse strains to Sendai virus and prevalence of infection in laboratory rodents. Infection and Immunology 19:123–130. Peckham, J. C. et al. 1974. Staphylococcal dermatitis in Mongolian gerbils (Meriones unguiculatus). Laboratory Animal Science 24:43–47. Peel, M. M. 1993. Dog-associated bacterial infections in humans: Isolates submitted to an Australian reference laboratory, 1981–1992. Pathology 25:379–384. Percy, D. H., and J. R. Barta. 1993. Spontaneous and experimental infections in SCID and SCID/beige mice. Laboratory Animal Science 43:127–132. Percy, D. H. et al. 1993. The enteritis complex in domestic rabbits—A field study. Canadian Veterinary Journal—Revue Veterinaire Canadienne 34:95. Perdue, K. A. et al. 2007. Naturally occurring murine norovirus infection in a large research institution. Journal of the American Association for Laboratory Animal Science 46:39–45. Pollack, J. D. et al. 1979. Lecithin changes in murine Mycoplasma pulmonis respiratory infection. Infection and Immunity 24:94–101. Popova, E. et al. 2005. Strain differences in superovulatory response, embryo development and efficiency of transgenic rat production. Transgenic Research 14:729–738. Pozzilli, P. et al. 1993. Nod mouse colonies around the world—Recent facts and figures. Immunology Today 14:193–196. Pritchett-Corning, K. R. et al. 2009. Contemporary prevalence of infectious agents in laboratory mice and rats. Laboratory Animals 43:165–173. Rath, H. C. et al. 1996. Normal luminal bacteria, especially bacteroides species, mediate chronic colitis, gastritis, and arthritis in HLA-B27/human beta(2) microglobulin transgenic rats. Journal of Clinical Investigation 98:945–953. ———. 1999. Varying cecal bacterial loads influences colitis and gastritis in HLA-B27 transgenic rats. Gastroenterology 116:310–319. ———. 2001. Different subsets of enteric bacteria induce and perpetuate experimental colitis in rats and mice. Infection and Immunity 69:2277–2285. Rehbinder, C. et al. 1998. FELASA recommendations for the health monitoring of breeding colonies and experimental units of cats, dogs and pigs—Report of the Federation of European Laboratory Animal Science Associations (FELASA) Working Group on Animal Health. Laboratory Animals 32:1–17.
84550.indb 301
11/11/10 6:25:56 PM
302
Handbook of Laboratory Animal Science
———. 2000. FELASA recommendations for the health monitoring of experimental units of calves, sheep and goats. Report of the Federation of European Laboratory Animal Science Associations (FELASA) Working Group on Animal Health. Lab Anim 34:329–350. Rehg, J. E. et al. 2001. Persistent transmission of mouse hepatitis virus by transgenic mice. Comparative Medicine 51:369–374. Riley, L. K. et al. 1996a. Expression of recombinant parvovirus NS1 protein by a baculovirus and application to serologic testing of rodents. Journal of Clinical Microbiology 34:440–444. ———. 1996b. Identification of murine helicobacters by PCR and restriction enzyme analyses. Journal of Clinical Microbiology 34:942–946. Rizzi, V. et al. 1995. Electrophoretic typing of some strains of enteric viruses isolated in rabbits suffering from diarrhea. Microbiologica 18:77–81. Rordorf, T. et al. 2000. Streptobacillus moniliformis endocarditis in an HIV-positive patient. Infection 28:393–394. Rozengurt, N., and S. Sanchez. 1994. Enteropathogenic catalase-negative cocci. In Welfare and science, ed. J. Bunyan. Proceedings of the Fifth Symposium of the European Laboratory Animal Science Associations, 11 June 8–11, Brighton, U.K. London: Royal Society of Medicine Press, 402–403. Ruebner, B. H., and T. Hirano. 1965. Viral hepatitis in mice. Changes in oxidative enzymes and phosphatases after murine hepatitis virus (MHV-3) infection. Laboratory Investigation 14:157. Ruskin, J., and J. S. Remington. 1971. Resistance to intracellular infection in mice immunized with toxoplasma vaccine and adjuvant. Journal of Reticuloendothelial Society 9:465–479. ———. 1976. Toxoplasmosis in the compromised host. Annals of Internal Medicine 84:193–199. Rygg, M., and C. F. Bruun. 1992. Rat bite fever (Streptobacillus moniliformis) with septicemia in a child. Scandinavian Journal of Infectious Disease 24:535–540. Sangild, P. T. et al. 2006. Diet- and colonization-dependent intestinal dysfunction predisposes to necrotizing enterocolitis in preterm pigs. Gastroenterology 130:1776–1792. Sasaki, H. et al. 2006a. Molecular typing of Pasteurella pneumotropica isolated from rodents by amplified 16S ribosomal DNA restriction analysis and pulsed-field gel electrophoresis. Microbiology and Immunology 50:265–272. ———. 2006b. Phylogenetic relationship of Pasteurella pneumotropica isolates from laboratory rodents based on 16S rDNA sequence. Journal of Veterinary Medical Science 68:639–641. Sato, Y. et al. 1995. Antibody production in Syphacia obvelata infected mice. Journal of Parasitology 81:559–562. Scanziani, E. et al. 1997. Outbreaks of hyperkeratotic dermatitis of athymic nude mice in northern Italy. Lab Anim 31:206–211. ———. 1998. Hyperkeratosis-associated coryneform infection in severe combined immunodeficient mice. Lab Anim 32:330–336. Scharf, R. A. et al. 1981. A modified barrier system for maintenance of Pasteurella-free rabbits. Laboratory Animal Science 31:513–515. Schauer, D. B. et al. 1995. Genetic and biochemical characterization of citrobacter rodentium sp-nov. Journal of Clinical Microbiology 33:2064–2068. Schoeb, T. R. et al. 1985. Exacerbation of murine respiratory mycoplasmosis in gnotobiotic F344/N rats by Sendai virus infection. Veterinary Patholology 22:272–282. Schoondermark-van de Ven, E. M. E. et al. 2006. Prevalence of naturally occurring viral infections, Mycoplasma pulmonis and Clostridium piliforme in laboratory rodents in Western Europe screened from 2000 to 2003. Laboratory Animals 40:137–143. Schramlova, J. et al. 1994. Electron-microscopic demonstration of virus particles in acute lymphoblastic leukaemia in Sprague–Dawley rats. Folia Biologica (Praha) 40:113–118. Schwanzer, V. et al. 1975. [Intrauterine infection of mice with ectromelia virus]. Zeitschrift für Versuchstierkunde 17:110–120. Sebesteny, A., and A. C. Hill. 1974. Hepatitis and brain lesions due to mouse hepatitis virus accompanied by wasting in nude mice. Lab Anim 8:317–326. Shanley, J. D., and E. L. Pesanti. 1980. Replication of murine cytomegalovirus in lung macrophages: Effect of phagocytosis of bacteria. Infection and Immunity 29:1152–1159. ———. 1982. Effects of antiviral agents on murine cytomegalovirus-induced macrophage dysfunction. Infection and Immunity 36:918–923.
84550.indb 302
11/11/10 6:25:56 PM
Health Status and Health Monitoring
303
Shoji, Y. et al. 1988. Pathogenicities of two CAR bacillus strains in mice and rats. Jikken Dobutsu 37:447–453. Shoji-Darkye, Y. et al. 1991. Pathogenesis of CAR bacillus in rabbits, guinea pigs, Syrian hamsters, and mice. Laboratory Animal Science 41:567–571. Shultz, L. D. et al. 1989. Pneumocystis carinii pneumonia in SCID mice. Current Topics in Microbiology and Immunology 152:243–249. Smith, A. L. et al. 1984. Two epizootics of lymphocytic choriomeningitis virus occurring in laboratory mice despite intensive monitoring programs. Canadian Journal of Comparative Medicine 48:335–337. Solomon, H. F. et al. 1990. A survey of staphylococci isolated from the laboratory gerbil. Laboratory Animal Science 40:316–318. Specter, S. C. et al. 1978. Macrophage-induced reversal of immunosuppression by leukemia viruses. Federal Proceedings 37:97–101. Srivastava, K. K. et al. 1976. Bacterial decontamination and antileukemic therapy of AKR mice. Infection and Immunity 14:1179–1183. Stephenson, E. H. et al. 1989. Efficacy of a commercial bacterin in protecting strain 13 guinea pigs against Bordetella bronchiseptica pneumonia. Lab Anim 23:261–269. Suzuki, E. et al. 1988. Naturally occurring subclinical Corynebacterium kutscheri infection in laboratory rats: Strain and age related antibody response. Laboratory Animal Science 38:42–45. Swartzberg, J. E. et al. 1975. Dichotomy between macrophage activation and degree of protection against Listeria monocytogenes and Toxoplasma gondii in mice stimulated with Corynebacterium parvum. Infection and Immunology 12:1037–1043. Tagaya, M. et al. 1995. Efficacy of a temperature-sensitive Sendai virus vaccine in hamsters. Laboratory Animal Science 45:233–238. Taguchi, F. et al. 1976. Difference in response to mouse hepatitis virus among susceptible mouse strains. Japan Journal of Microbiology 20:293–302. Takahashi, T. et al. 1995. Assignment of the bacterial agent of urinary calculus in young rats by the comparative sequence analysis of the 16S rRNA genes of corynebacteria. Journal of Veterinary Medical Science 57:515–517. Takeuchi, A., and K. Hashimoto. 1976. Electron microscope study of experimental enteric adenovirus infection in mice. Infection and Immunity 13:569–580. Tamura, M. et al. 2004. Lactobacillus gasseri: Effects on mouse intestinal flora enzyme activity and isoflavonoids in the caecum and plasma. British Journal of Nutrition 92:771–776. Taurog, J. D. et al. 1984. Infection with Mycoplasma pulmonis modulates adjuvant- and collagen-induced arthritis in Lewis rats. Arthritis & Rheumatism 27:943–946. ———. 1994. The germ-free state prevents development of gut and joint inflammatory disease in Hla-B27 transgenic rats. Journal of Experimental Medicine 180:2359–2364. Theau-Clement, M. 2008. Factors of success of rabbit doe insemination and methods for estrus induction. Productions Animales 21:221–230. Thomsen, A. C., and I. Heron. 1979. Effect of mycoplasmas on phagocytosis and immunocompetence in rats. Acta Pathologica et Microbiologica Scandinavica C 87C:67–71. Tiensiwakul, P., and S. S. Husain. 1979. Effect of mouse hepatitis virus infection on iron retention in the mouse liver. British Journal of Experimental Pathology 60:161–166. Tindle, R. W. et al. 1976. Resistance of mice to Krebs ascites tumor, sarcoma S180 and PC6 plasmacytoma after immunization with Salmonella enteritidis 11RX. Australian Journal of Experimental Biology and Medical Science 54:149–155. Tischler, N. D. et al. 2008. Characterization of cross-reactive and serotype-specific epitopes on the nucleocapsid proteins of hanta viruses. Virus Research 135:1–9. Tlaskalova-Hogenova, H. et al. 2004. Commensal bacteria (normal microflora), mucosal immunity and chronic inflammatory and autoimmune diseases. Immunology Letters 93:97–108. Toolan, H. W. 1967. Lack of oncogenic effect of the H-viruses for hamsters. Nature 214:1036. Toolan, H. W., and N. Ledinko. 1968. Inhibition by H-1 virus of the incidence of tumors produced by adenovirus 12 in hamsters. Virology 35:475–478. Toolan, H. W. et al. 1982. Inhibition of 7,12-dimethylbenz(a)anthracene-induced tumors in Syrian hamsters by prior infection with H-1 parvovirus. Cancer Research 42:2552–2555. Turnbaugh, P. J. et al. 2006. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 444:1027–1031.
84550.indb 303
11/11/10 6:25:57 PM
304
Handbook of Laboratory Animal Science
Tyzzer, E. E. 1917. A fatal disease of the Japanese waltzing mouse caused by a spore-bearing bacillus (Bacillus piliformis N.Sp.). Journal of Medical Research 37:307–338. Ueno, Y. et al. 1997. Epidemiological characterization of newly recognized rat parvovirus, “rat orphan parvovirus.” Journal of Veterinary Medical Science 59:265–269. ———. 1998. Prevalence of “orphan” parvovirus infections in mice and rats. Experimental Animals 47:207–210. Urano, T., and K. Maejima. 1978. Provocation of pseudomoniasis with cyclophosphamide in mice. Laboratory Animals 12:159–161. Uzal, F. A. et al. 1989. A study of lung lesions in asymptomatic rabbits naturally infected with B. bronchiseptica. Scandinavian Journal of Laboratory Animal Science 16:3. Vaahtovuo, J. et al. 2001. Study of murine fecal microflora by cellular fatty acid analysis; effect of age and mouse strain. Antonie Van Leeuwenhoek International Journal of General and Molecular Microbiology 80:35–42. Vallee, A. et al. 1969. [Isolation of a strain of Corynebacterium kutscheri in a guinea pig]. Bulletin de l’Academie Veterinaire de France 42:797–800. van der Waaij, D., and J. M. Berghuis-de Vries. 1974. Selective elimination of Enterobacteriaceae species from the digestive tract in conventional and antibiotic treated mice. Journal of Hygiene (Cambridge) 69:405–411. van der Waaij, D. et al. 1974. Mitigation of experimental bowel disease in guinea pigs by selective decontamination of the aerobic Gram negative flora. Gastroenterology 67:460–472. van Hoosier, G. L. 1986. Answer on a question concerning Sendai virus vaccination. In Viral and mycoplasmal infections of laboratory rodents, effects on biomedical research, ed. P. N. Bhatt, R. O. Jacoby, H. C. Morse III, and A. E. New, 61. New York: Academic Press Inc. Ventura, J., and M. Domaradzki. 1967. Role of mycoplasma infection in the development of experimental bronchiectasis in the rat. Journal of Pathology and Bacteriology 93:342–348. Vogelbacher M., and W. Bohnet. 1997. Distribution of Staphylococcus species in laboratory mice and laboratory rats compared with those found in house mice and Norway rats. In Harmonization of laboratory animal husbandry, ed. P. N. O’Donoghue. Proceeding of the Sixth Symposium of the Federation of the European Laboratory Animal Science Associations, June 19–21, Basel, Switzerland. London: Royal Society of Medicine. Vonderfecht, S. L. et al. 1983. Infectious diarrhea of infant rats—Clinicopathologic observations and characterization of a viral agent. Laboratory Animal Science 33:502. ———. 1984. Infectious diarrhea of infant rats produced by a rotavirus-like agent. Journal of Virology 52:94–98. Voot, L. et al. 1988. The development of bb-rat diabetes is delayed or prevented by infections or applications of immunogens. Zeitschrift für Versuchstierkunde 31:197. Walker, R. I. et al. 1978. Antibiotic decontamination of the dog and its consequences. Laboratory Animal Science 28:55–61. Waller, T. et al. 1978. Humoral immune response to infection with Encephalitozoon cuniculi in rabbits. Lab Anim 12:145–148. Walzer, P. D. et al. 1989. Outbreaks of pneumocystis carinii pneumonia in colonies of immunodeficient mice. Infection and Immunity 57:62–70. Ward, J. M. et al. 1994. Chronic active hepatitis in mice caused by Helicobacter hepaticus. American Journal of Pathology 145:959–968. Weber, H. et al. 1999. Health monitoring of nonhuman primate colonies. Recommendations of the Federation of European Laboratory Animal Science Associations (FELASA) Working Group on nonhuman primate health accepted by the FELASA Board of Management, November 21, 1998. Lab Anim 33 (Suppl 1): S1–S18. Wei, Q. et al. 1995. Taxonomic status of CAR bacillus based on the small subunit ribosomal RNA sequences. Chinese Medical Science Journal 10:195–198. Weir, E. C. et al. 1987. Elimination of mouse hepatitis virus from a breeding colony by temporary cessation of breeding. Laboratory Animal Science 37:455–458. Weisbroth, S. H., and S. Scher. 1968. Corynebacterium kutscheri infection in the mouse. II. Diagnostic serology. Laboratory Animal Care 18:459–468.
84550.indb 304
11/11/10 6:25:57 PM
Health Status and Health Monitoring
305
Welch, B. G. et al. 1977. Development of a guinea pig colony free of complement-fixing antibodies to parainfluenza virus. Laboratory Animal Science 27:976–979. Wells, A. B. 1970. The kinetics of cell proliferation in the tracheobronchial epithelia of rats with and without chronic respiratory disease. Cell Tissue Kinetics 3:185–206. Werner, R. M. et al. 1981. Clinical manifestations of mousepox in an experimental animal holding room. Laboratory Animal Science 31:590–594. Westerberg, S. C. et al. 1972. Mycoplasma-virus interrelationships in mouse tracheal organ cultures. Infection and Immunity 5:840–846. White, D. J., and M. M. Waldron. 1969. Naturally occurring Tyzzer’s disease in the gerbil. Veterinary Record 85:111–114. Wilberz, S. et al. 1991. Persistent Mhv (mouse hepatitis virus) infection reduces the incidence of diabetes mellitus in nonobese diabetic mice. Diabetologia 34:2–5. Wullenweber, M. et al. 1990a. Staphylococcus aureus phage types in barrier-maintained colonies of SPF mice and rats. Zeitschrift für Versuchstierkunde 33:57–61. ———. 1990b. Streptobacillus moniliformis epizootic in barrier-maintained C57BL/6J mice and susceptibility to infection of different strains of mice. Laboratory Animal Science 40:608–612. ———. 1992. Streptobacillus moniliformis isolated from otitis media of conventionally kept laboratory rats. Journal of Experimental Animal Science 35:49–57. Zaccone, P. et al. 2003. Schistosoma mansoni antigens modulate the activity of the innate immune response and prevent onset of type 1 diabetes. European Journal of Immunology 33:1439–1449. Zenner, L. 1999. Pathology, diagnosis and epidemiology of the rodent Helicobacter infection. Comparative Immunology Microbiology and Infectious Diseases 22:41–61. Zenner, L., and J. P. Regnault. 2000. Ten-year long monitoring of laboratory mouse and rat colonies in French facilities: A retrospective study. Laboratory Animals 34:76–83. Zhang, C. H. et al. 2010. Interactions between gut microbiota, host genetics and diet relevant to development of metabolic syndromes in mice. Isme Journal 4:232–241. Zook, B. C. et al. 1977a. Tyzzer’s disease in the Chinese hamster (Cricetulus griseus). Laboratory Animal Science 27:1033–1035. ———.1977b. Tyzzer’s disease in Syrian hamsters. Journal of the American Veterinary Medical Association 171:833–836.
84550.indb 305
11/11/10 6:25:57 PM
84550.indb 306
11/11/10 6:25:57 PM
Chapter 11
Nutrient Requirements, Experimental Design, and Feeding Schedules in Animal Experimentation
Jo H. A. J. Curfs, André Chwalibog, Bart S. Savenije, and Merel Ritskes-Hoitinga Contents Introduction.....................................................................................................................................308 Nutrient Requirements....................................................................................................................308 Food Intake................................................................................................................................309 Digestion.................................................................................................................................... 310 Metabolism................................................................................................................................. 311 Carbohydrate.............................................................................................................................. 312 Fat............................................................................................................................................... 313 Protein........................................................................................................................................ 314 Energy........................................................................................................................................ 316 Requirements............................................................................................................................. 318 Allowances................................................................................................................................. 319 Types of Diets................................................................................................................................. 321 Natural-Ingredient Diets............................................................................................................ 321 Between-Brand Variation...................................................................................................... 322 Within-Brand Variation......................................................................................................... 323 Purified Diets............................................................................................................................. 323 Standardization..................................................................................................................... 323 Achieving Desired Nutrient Levels....................................................................................... 325 Fulfilling Essential Needs—Avoiding Toxic Levels............................................................. 326 Purified Diets versus Species................................................................................................ 327 Points to Consider when Preparing Purified Diets................................................................ 327 Pelletability and Feeding Devices......................................................................................... 328 Contaminants............................................................................................................................. 329 Quality Considerations............................................................................................................... 330 Storage Conditions................................................................................................................ 330 Sterilization........................................................................................................................... 330 Pellet Hardness...................................................................................................................... 330
307
84550.indb 307
11/11/10 6:25:57 PM
308
Handbook of Laboratory Animal Science
Feeding Schedules.......................................................................................................................... 331 Ad Libitum Intake and Isocaloric Exchange............................................................................. 331 Restricted Feeding..................................................................................................................... 331 Meal Feeding............................................................................................................................. 333 Pair Feeding............................................................................................................................... 333 Influence of Feeding Schedules in Pharmacological Studies.................................................... 333 Influence of Feeding Schedules on Circadian Rhythms............................................................ 334 Diet and Welfare............................................................................................................................. 335 Transport and Acclimatization................................................................................................... 335 Enrichment and Variation versus Need for Standardization..................................................... 335 Conclusion....................................................................................................................................... 336 References....................................................................................................................................... 337 Introduction This chapter on nutrition of laboratory animals discusses the influence of diets, dietary composition, and feeding schedules on experimental results, and animal health and welfare. Food is consumed every day, and many processes in the body are dependent upon and affected continuously by what is eaten, and when and how it is ingested. The chapter provides scientific information and examples of designing sound experiments in the field of nutrition. This is important when doing not only nutritional studies, but also other studies in which nutritional interference is undesirable. Nutrition is also important in regulating the expression of a host of genes. It is not the genetic makeup alone that is important in determining gene activity; interactions with the environment are crucial as well (epigenetics). A good experimental design will contribute to reliable and reproducible results without unnecessarily compromising animal welfare. Because many species spend a large part of their day foraging, suggestions concerning dietary enrichment are presented. An increased effort in the field of dietary enrichment is expected to contribute to the improvement of the welfare of laboratory animals and more reliable experimental results. The possible conflict that might arise between the need for standardization and the need for enrichment is a current research challenge for laboratory animal scientists and nutritionists. Nutrient Requirements Nutrients and energy are required for a number of different functions in the body—for the vital processes necessary for an animal to survive, and for the productive processes, such as reproduction, growth, and lactation. The requirement of a single nutrient or energy may be defined as the smallest quantity of the nutrient or energy required for a particular body function, provided all other nutrients (and/or energy) are supplied in optimal quantities. As stated by Fuller and Wang (1987), in attempting to estimate requirements, one must ask three questions: • What animal or population of animals are requirements for? • What are requirements for ? • How are requirements to be specified?
The first question arises from great differences between animals in requirements. Animals of different age, body weight, and sex have different requirements. Furthermore, even when these differences are excluded, the individual animals will still differ in their requirements, due to genetic constitution—a more difficult influence to quantify. It is often observed from animal experiments that even individuals of the same age, sex, and genetic line, while kept in similar environments, will show interindividual variation of up to 15 or 20% for many nutritional parameters.
84550.indb 308
11/11/10 6:25:57 PM
Nutrient Requirements, Experimental Design, and Feeding
309
The second question relates to the criteria we choose to define requirement. For example, a requirement for animal performance can be defined in terms of body weight gain, feed conversion for gain, or metabolic responses. The third question requires a specification of terms: whether requirements are, for example, for metabolizable energy or for net energy, or for digestible protein or for amino acids, or for whatever combination of energy and protein is preferred. In general, the basis for the nutritional requirements of farm animals and laboratory animals are similar. All require energy, protein, carbohydrate, lipid, minerals, and vitamins supplied in diets that should be palatable and free from chemical and biological contaminations. Therefore, much information regarding farm animal nutrition may be applied to laboratory animals. However, we have to keep in mind that the aim of laboratory animal nutrition is not the highest production, but rather optimum performance and nutritionally unbiased responses to biomedical treatments. It is well established that a diet deficient in composition or quantity may influence not only animal growth or reproductive performance, but also immune response and resistance to diseases. Furthermore, prior to changes in animal performance, a number of physiological and anatomical changes caused by a deficiency or an excess of nutrients may interact with the action of biomedical treatments. Therefore, this chapter will outline some of the important effects of nutrition on the performance of laboratory animals, concomitantly emphasizing possible interactions between nutrition and experimental results. Most biomedical experiments currently being conducted use small rodents as subjects and, because the existing data are most abundant for rats and mice, this chapter will focus on these species. Due to the diversity of factors associated with the supply of nutrients and energy for laboratory animals, it would simply be impossible to debate all possible effects of nutrition on experimental results. Therefore, only some of the major nutritional characteristics of macronutrients and energy will be discussed in the following sections. Food Intake Laboratory rodents are usually fed ad libitum. It is generally accepted that voluntary food intake is related to the energy requirement of the animal. The classic experiments of Adolph in 1947 demonstrated that when rat diets were diluted with inert materials to produce a wide range of energy concentrations, the animals were able to adjust the amount of food eaten so that their energy intake remained constant (McDonald et al. 1981). However, the concept that “animals eat for energy” is more complicated than it sounds. In the case of extensive dilution of the diet with materials of low digestibility, the ability of the subjects to adjust their intake may be impaired by the limiting factor of gastrointestinal capacity. Food intake may therefore be insufficient to cover energy and often also nutrient requirements. On the other hand, when dietary energy density is high enough to cover energy requirements, an increase in energy density by supplementing the diet with extra fat or carbohydrate may result in nutrient deficiencies. This occurs because the animal usually stops eating when its energy requirement has been met. Furthermore, not only do energy density of a diet and the capacity of the gastrointestinal tract influence food intake, but reduced intake is also commonly observed in deficiency states. This is especially true with diets low in protein (Forbes 1995), unbalanced in amino acids, or deficient in or excessive for some trace minerals and vitamins. Diets that are highly palatable (taste good), such as “hamburger diets,” result in rodents eating more than necessary to meet their energy requirement, making them obese. The physiological state of the animal also plays an essential role in food intake (McDonald et al. 2002). Several reports show increased intake with the onset of pregnancy, but other reports suggest few if any changes (Forbes 2007). Lactation is usually associated with a marked increase in food intake; a rat at peak lactation will consume nearly three times the amount of food consumed by a nonlactating rat. Considering that voluntary food intake may be subject to marked variation
84550.indb 309
11/11/10 6:25:58 PM
310
Handbook of Laboratory Animal Science
Table€11.1╅Estimated Average Food Intake (g/day) Species
Growing
Adult
Pregnant
Lactating
Mouse Rat Hamster Guinea pig Rabbit
3–5 8–25 6–12 35–45 120–200
5–7 25–30 10–12 45–70 200–300
6–8 25–35 12–15 70–80 300
7–15 35–65 20–25 100–130 300–400
depending on nutritional and physiological factors, it is difficult to specify an expected daily consumption. Consequently, Table€ 11.1 includes the approximate voluntary daily food intake in the most common laboratory animals when they are fed commercial pelleted diets. The metabolizable energy content of these diets usually lies within the range of 8–12 MJ/kg diet. For growing animals, the wide range of daily food intake is directly related to the age and live weight (LW), increasing during the growth period. For example, the voluntary food intake of growing rats, reported by Thorbek and colleagues (1982) was 15 g/day at the age of 5 weeks and a LW of 100 g; from 7 to 8 weeks (200 g LW) until 18 weeks (400 g LW) of age, it was remarkably constant at about 25 g/day. With a dietary gross energy concentration of 16.5 MJ/kg of food, the values correspond to about 1400 kJ/LW per kilogram of metabolic live weight (LW, kg0.75) at 100 and 200 g LW, but only to 820 kJ/LW, kg0.75 at 400 g LW. (The metabolic live weight is a parameter used to compare species. It is used in order to correct for differences in the metabolic rate per kilogram; Kleiber 1961.) Comparable values can be calculated from the experiments of Pullar and Webster (1974) with “lean” rats indicating a pattern of decreasing energy consumption and thereby food consumption in relation to metabolic live weight. Kleiber (1961) has suggested that the maximum food intake, for all species, is proportionally related to basal metabolic rate (BMR) with the ratio of 4:1. Assuming a constant BMR of 320 kJ/LW, kg0.75, the maximum intake in growing laboratory animals would be about 1.3 MJ/LW, kg0.75. In accordance with Kleiber’s principle, Clarke and colleagues (1977) presented general equations for growth, pregnancy, and lactation in all laboratory animals, suggesting constant proportions between energy supply for different life processes and metabolic live weight. Despite the fact that such approaches may give some indication of the level of food intake, the accuracy of the predicted values is doubtful because energy intake in relation to metabolic live weight is not constant at the maintenance level (Eggum and Chwalibog 1983) or during growth (Thorbek et al. 1982). Nevertheless, it is practical to use when performing nutritional studies, allowing one to compare groups of animals of the same strain, age, sex, etc. at the same time to determine whether the intake of food and nutrients is similar (unless the researcher wants intake to differ). Digestion The supply of nutrients required for body functions depends on the transformation of the dietary constituents into simpler elements (amino acids, glucose, fatty acids) before they can pass through the mucous membrane of the gastrointestinal tract into the blood and lymph. The process of digestion results from muscular contraction of the alimentary canal, microbial fermentation, and the action of digestive enzymes secreted in digestive juices. In monogastric animals like rats and mice, microbial activity in the large intestine is low; these animals mainly process food compounds by means of the digestive enzymes and acids (McDonald et al. 2002). In suckling rats or mice, the action of the digestive system and the secretion of enzymes are restricted to hydrolysis of the components of maternal milk. The serous glands of the tongue produce a lingual lipase that is important for the digestion of milk triglycerides (Henning 1987). In contrast to the serous glands, the salivary glands of neonatal rats and mice are functionally immature, and amylase activity is negligible during the first two postnatal weeks. In the pancreas, amylase activity
84550.indb 310
11/11/10 6:25:58 PM
Nutrient Requirements, Experimental Design, and Feeding
311
does not begin to increase until 2 weeks after birth (Prochazka et al. 1964). The gastric secretion of HCl, pepsinogen, and pepsin is minimal, thus allowing intact protein to pass into the small intestine, where it is absorbed as intact macromolecules by the process of pinocytosis (Henning 1987). Weaning of rats and mice, which normally begins at 17 days of age and is completed by day 26, constitutes a significant change in dietary composition from milk, with a high content of lipids (9%) and a low content of carbohydrate (4%), to a diet low in fat and high in carbohydrate. The change in dietary composition necessitates changes in digestive function. The intestinal hydrolases (maltase, sucrase, isomaltase, trehalase) involved in digestion of carbohydrate from solid food cannot be detected in the intestines of rats during the first two postnatal weeks, but their activities rise rapidly later (Rubino et al. 1964). The activities of amylase, chymotrypsin, trypsinogen, and lipase change little before weaning, but increase dramatically at the time of weaning (Descholdt-Lanckman et al. 1974). The gastric secretion of acids and pepsinogen rises to adult levels during the third and fourth postnatal weeks, coincident with the transition to solid food (Henning 1987). Although the enzymatic changes that occur in the gastrointestinal tract about weaning time seem to be directly related to the change of diet from milk to solid food, there is evidence that the primary cause of the enzymatic development is not a change of diet (Henning 1981). Among other regulatory factors that have been suggested, glucocorticoids, thyroxine, glucagon, gastrin, cholecystokinin, prostaglandins, and insulin play an important role as potential regulators of postnatal development of the gastrointestinal tract (Henning 1987). In older rats, digestive capacity is stabilized and remains almost constant under normal feeding conditions. The presence of food in the stomach may have a significant influence on the bioavailability and pharmacokinetics of certain drugs (Melander 1978). Rats eat at night and are usually fasted overnight prior to dosing with different drugs, based on the assumption that overnight fasting will result in the postabsorptive state (Jefery et al. 1987). However, substantial evidence indicates that the rate of gastric emptying (e.g., the rate of passage in the alimentary tract) depends on diet composition. In balance experiments with rats at about 140 g LW and using glass beads as markers, it was demonstrated (Raczynski et al. 1982) that the highest quantity of markers was recovered in feces about 30 h after the beginning of eating. Only marginal amounts could be detected in the digestive tract at 72 h after feeding. Protein and fat levels in the diet did not affect the rate of passage. However, crude fiber strongly increased the rate of passage, with the highest recovery of the markers about 20 h after feeding. It was also demonstrated that reduced levels of microbial activity in the hind-gut decreased the rate of passage time by about 15 h. It is interesting to note that in these experiments, the level of microbial activity was regulated by the administration of the antibiotic Nebacitin, thereby indicating that compounds that alter microbial metabolism might affect the rate of food passage in the alimentary tract and, subsequently, the digestibility of nutrients. Metabolism The starting point of metabolism is the substances produced by the digestion of food. Digested carbohydrate, fat, and protein are the main groups of nutrients involved in a variety of catabolic and anabolic processes in the body. The general relations between the intake of digested nutrients and the end products of their metabolism in the body are presented in Figure 11.1. The soluble part of digested carbohydrate is mainly absorbed as glucose, which can be stored as glycogen. It can be oxidized and used as a source of energy and utilized in the process of de novo lipogenesis. The insoluble carbohydrates (fibers) are fermented and transformed into short chain fatty acids (SCFAs), which finally are sources of energy and precursors in the synthesis of glucose and body fat. Although SCFAs are a significant source of energy for herbivorous animals (in the guinea pig and the rabbit, approximately 30% of the energy is supplied from SCFA metabolism), the microbial
84550.indb 311
11/11/10 6:25:58 PM
Handbook of Laboratory Animal Science
DCHO Glucose
SCFA Acetic Butyric
DFAT
DPROT TG
FFA
Intestine
312
AA
Propionic Keto acids
OX
Glycogen
TG
Glycerol
FFA
OX
AA
NH3 Body Tissue
FFA
Glucose
PROT
Acetate BHBA OX
Glucose FAT
OX
AA FAT
OX
Urea
PROT
Figure 11.1â•…Nutrient partition in the body: digested carbohydrate (DCHO), fat (DFAT), and protein (DPROT); short-chain fatty acids (SCFAs), free fatty acids (FFAs), triacylglycerols (TGs), amino acids (AAs) oxidized substrates (OX), and β-hydroxybutyric acid (BHBA).
capacity for SCFA production in monogastric animals such as the rat and the mouse is limited. The free fatty acids (FFAs) and triglycerides from digested fat are transformed into body fat, and they can be oxidized to become an efficient energy source. In the case of inadequate energy supply from a diet, body fat can be mobilized as an additional energy source. Amino acids from digested protein are synthesized and retained in the body or milk and partly deaminated and oxidized with concomitant transfer of their carbon skeletons into gluconeogenesis and ketogenesis conversion of ammonia into urea. Carbohydrate Carbohydrate (CHO) constitutes the largest proportion of food consumed by laboratory animals, except among carnivores. CHO is the most important component of plants, constituting up to 75% of the dry matter present in feeds of plant origin. Dietary CHO, consisting of α-monosaccharide units (soluble CHO), is readily digested by endogenous enzymes and constitutes the major energy source for laboratory animals. A number of CHOs can be used by the rat, and, as reviewed by the National Research Council (1995), glucose, sucrose, maltose, fructose, and starch support similar levels of performance. However, excessive lactose or galactose in the diet may cause diarrhea and poor performance. Diets for rats and mice are mainly based on starch, which is a relatively inexpensive energy source that yields 17.6 kJ/g. There is a direct relation between CHO intake and the level of fatness of the animals, because CHO that exceeds the amount needed to meet the energy requirement will
84550.indb 312
11/11/10 6:25:59 PM
Nutrient Requirements, Experimental Design, and Feeding
313
be stored as body fat (Chwalibog et al. 1992; Chwalibog et al. 1998b). Although rats and mice are able to regulate food intake, depending on the energy density of the diet, an extensive supply of soluble CHO may result in obesity—especially in animals with genetic predispositions to obesity, like Zucker “fatty” rats (Rafecas et al. 1991). Obesity is also more likely to occur when the diet is high in sugar because sugar is utilized more efficiently than starch (Glick et al. 1984). All rodents exhibiting obesity or obesity-related diabetes syndrome are characterized by diminished glucose tolerance that may result in hyperinsulinemia and inappropriate hyperglycemia (Herberg 1991) and thereby in reduced longevity. The CHOs composed of β-monosaccharide units are called insoluble CHOs and are collectively referred to as “dietary fiber.” They resist the action of digestive enzymes, but can be utilized by microorganisms in the large intestine. Plant cell wall material, the major source of dietary fiber, is composed chiefly of cellulose, hemicellulose, and pectins. Microbial metabolism is limited in rats and mice compared with herbivorous animals, and fiber digestion is almost entirely confined to microbial activity in the hind-gut (Eggum et al. 1982). Only a small amount of dietary fiber (2–5% of the diet) should be included in the diets of rats and mice. This proportion can be processed by the microorganisms. Fiber is important because of its water-holding capacity and its influence on the peristaltic motility of the intestines—the process by which food components are driven through the alimentary tract. Furthermore, dietary fiber components influence the composition and function of the intestinal microflora and activate the microbial synthesis of several vitamins. It has been demonstrated that pectin stimulates the microbial synthesis of thiamin, riboflavin, and niacin (Rotenberg et al. 1982). Pectin can absorb various antimetabolites and thus reduce their degree of absorption (Rotenberg and Andersen 1982). There are also indications that fiber increases peripheral sensitivity to insulin, perhaps by increasing the number of insulin receptors (Anderson 1985). On the other hand, an excess of dietary fiber has a negative influence on nutrient and energy digestibility because it increases the rate of passage of food components in the digestive tract and subsequently reduces the time of action of digestive enzymes. Total glucose, lipid, and protein absorption is decreased when high levels of dietary fiber are consumed (Cummings 1978; Zhao, Jørgensen, and Eggum 1995). Additionally, an association between dietary fiber and periodontitis and oronasal fistulation in rats has been reported in several investigations (Robinson et al. 1991) and is presumed to relate to the presence of the long, sharp fibers of oats or barley (Madsen 1989). Fat Dietary fat is required as a source of essential fatty acids (EFAs) for the absorption of fat-soluble vitamins and to enhance the palatability of food. Lipids are an excellent source of energy, providing 2.5 times more energy (39.8 kJ/g) than carbohydrates (17.6 kJ/g) and protein (18.4 kJ/g). However, if the diet contains adequate CHO and protein, fat is not used as a source of energy, but rather is stored as body lipids (Chwalibog and Thorbek 2000). Both the amount and composition of dietary fat are important in laboratory animal nutrition. A high fat level in the diet increases cholesterol synthesis. Especially high inputs of saturated fat furnish acetyl-CoA in excess of that required for energy production and body fat synthesis, and the excess acetyl-CoA is used for cholesterol formation. The fatty acid composition of the dietary fat affects antioxidant mechanisms in the colon mucosa, presumably because the composition of cell membranes reflects the fatty acid composition of the diet (Kuratko and Pence, 1992). There is evidence that high-fat diets elevate the toxic effects of nuclear-damaging agents and carcinogens (Bird and Bruce 1986). It has been demonstrated that an increase in fat level may alter the acute genotoxic effects of carcinogens, a phenomenon associated with the initiation of colon cancer (Bull et al. 1989).
84550.indb 313
11/11/10 6:25:59 PM
314
Handbook of Laboratory Animal Science
Furthermore, mammary tumor incidence is related to the fatty acid composition of the diet, with a greater incidence of tumors in rats fed diets containing polyunsaturated fatty acids (PUFAs) than in rats fed diets with saturated fat. In the rat, increasing levels of linoleic acid were correlated with increasing chemically induced mammary tumor incidence up to a maximum at 4.5% of dietary linoleic acid (Ip, Carter, and Ip 1985; Ip 1987). The same effect was observed in the mouse, but at a higher level of linoleic acid (8.4%) (Fisher et al. 1992). Feeding high levels of fat, particularly PUFA, may depress immune responsiveness (Crevel et al. 1992), which may lead to an increased susceptibility to infections and tumor development (J. Ritskes-Hoitinga et€al.€1996). The early work of Burr and Burr (1929, 1930) established that the rat does not thrive on diets rigidly devoid of fat and will develop a number of deficiency syndromes. The linoleic acid family of PUFA was shown to reverse the effects of fat-free diets. The linoleic, linolenic, and arachidonic acids are usually referred to as essential fatty acids (EFAs). Mammals lack the enzymes that introduce double bonds at carbon atoms beyond C-6 in the fatty acid chain. This makes the double bond at the 12th carbon atom of linoleic acid “essential.” After absorption, linoleic acid can be oxidized, accumulated in the adipose tissue, converted to PUFA, and incorporated into structural lipids (Innis 1991). The list of symptoms ascribed to EFA deficiency includes classical signs, such as reduced growth rate, dermal lesions, increased water permeability of the skin, increased susceptibility to bacteria, decreased prostaglandin synthesis and reproductive failure, reduced myocardial contractility, abnormal thrombocyte aggregation, swelling of liver mitochondria, and increased heat production. Dietary requirements of EFA are usually stated in terms of linolate. An amount equivalent to 1–1.5% of the metabolizable energy (ME) of the diet has been found to be adequate for most monogastric animals (McDonald et al. 1981). However, studies on growing pigs indicate lower requirements (0.26% of the ME) (Christensen 1985). This level is likely to be present in all natural food compounds used for laboratory animals, except for highly refined or purified diets, which must be fortified with EFA-rich sources like soybean oil. Also, diets containing a high level of saturated fatty acids (>5%) may require a larger supply of EFAs; hence, EFAs enhance the utilization of saturated fatty acids (Holman 1968). Protein The nutritive effects of a protein depend on the amino acids released from the protein by digestive processes. For nutritional purposes, amino acids are classified into two groups: nonessential (NE) and essential (E) amino acids (AAs). NEAAs are not necessary as dietary components because they can be synthesized in the body via intermediates of carbohydrate metabolism or by transformation of some EAAs into certain NEAAs. EAAs, however, cannot be synthesized in the body, at least not at a rate adequate to meet physiological requirements, and they must therefore be supplied within the diet (Table€11.2). Protein synthesis can only take place when all the amino acids required to form a certain protein are present together; thus, a relative inadequacy of a single amino acid can impair utilization of the rest. The amino acid in lowest concentration, relative to the requirement, will therefore determine the rate at which protein can be synthesized in the body. Consequently, amino acids present in excess of the requirement for protein synthesis will not be used for synthesis, but their nitrogen-free components will be oxidized or used in gluconeogenesis and ketogenesis. The nitrogenous component (ammonia) is converted by the liver to urea and excreted by the kidneys with concomitant energy loss. It is evident that EAA must be present in the diet in correct quantities and proportions in order to be synthesized into animal protein. However, the animal must also receive a sufficient amount of NEAA as a nitrogen source for protein synthesis. If an inadequate amount of NEAAs is absorbed (and produced from body protein turnover), they will be resynthesized from dietary EAA.
84550.indb 314
11/11/10 6:25:59 PM
Nutrient Requirements, Experimental Design, and Feeding
315
Table€11.2╅Essential and Main Nonessential Amino Acids for Growing Rats and Mice Essential
Nonessential
Arginine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Tryptophan Valine
Alanine Aspartic acid Cystine Glutamic acid Glycine Hydroxyproline Proline Serine Tyrosine
Inadequate amino acid supply is the most common of all nutrient deficiencies. Signs of protein deficiency include reduced protein concentration in the blood, reduced protein synthesis rate in the tissues and synthesis of certain enzymes and hormones, decreased food intake, reduced growth rate, and infertility. On the other hand, an excess supply of one or more amino acids may cause amino acid imbalance and, consequently, decreased protein utilization. The classic experiments with growing rats fed a rice diet, with lysine and threonine as limiting amino acids, demonstrated that if the lysine content of the diet was held constant and the threonine content was increased stepwise, a point was reached at which the growth of rats fed on a threoninesupplemented diet was retarded unless the lysine content of the diet was also increased (Eggum 1989). The same phenomenon was seen in the reverse situation, when threonine was held constant and lysine increased. Furthermore, the sites in the brain that regulate food intake are sensitive to an alteration in the proportion of amino acids in the blood plasma, and an imbalance of amino acids may cause reduced food intake (Rogers and Leung 1973). There is also evidence that a surplus of arginine and histidine may depress protein utilization, while the ingestion of a large amount of methionine or tyrosine (20–50 g/kg food) is followed by serious metabolic and histopathological changes, in addition to depressed food intake and retarded growth in the rat (Harper et al. 1970). Excess methionine inhibits ATP synthesis, causing irregularities in energy metabolism, and excess tyrosine causes a specific toxic syndrome with histopathological changes in the skin, pancreas, and liver, and severe eye lesions. For some amino acids, negative effects of an excess supply may only be prevented by the addition of other structurally similar amino acids. Growth depression in rats caused by surplus isoleucine and valine can be prevented by addition of their “antagonistic” amino acid, leucine. It is also interesting to note that a high-quality protein source is much less affected by an excess of a single amino acid than is a protein source of poorer quality, as demonstrated for egg protein compared to barley and potato protein (Eggum et al. 1981). High dietary protein levels should be avoided in case of kidney disease because the relatively high urea excretion will negatively influence kidney function. Dietary protein can also interfere with tumor studies. For example, supplementation of methionine to soybean protein isolates increased mammary tumor progression in rats (Hawrylewicz et al. 1991). It was also suggested that decreased levels of dietary methionine may decrease tumor cell proliferation. When discussing the role of dietary protein in laboratory animal nutrition, it has to be emphasized that the utilization of protein for different life processes is dependent on energy supply. The relationship between protein balance and protein and energy intake has been recognized for years (Dean and Edwards 1985; Edwards et al. 1985; Edwards and Dean 1985) (Figure€11.2).
84550.indb 315
11/11/10 6:26:00 PM
316
Handbook of Laboratory Animal Science
Retained Protein
High protein level
Low protein level
Level of Energy Intake A
B
Figure 11.2â•…Relation between protein retention and energy intake.
For a low-protein diet, an increase of energy intake to point A (Figure€11.2) will increase protein retention, while the extended energy supply from A to B will not have any effect on protein retention. For a high-protein diet, the pattern is the same; however, because of higher protein supply, protein retention is elevated. For such a diet, the increase from A to B will stimulate protein retention until the limit is reached at point B. The presented relationship between energy and protein is, of course, a simplification, and other factors should also be considered, such as the extent to which energy is supplied from carbohydrate or fat (Yoshida et al. 1957; Eggum 1973), the supply of other nutrients (Szelényi-Galantai et al. 1981), amino acid profile in the diet (Eggum 1973), and endocrinological regulation of anabolic and catabolic processes. Energy All functions of the body require energy, which is supplied from carbohydrate, fat, and protein. The results of inadequate energy supply are obvious, but excessive energy intake can be harmful as well. Excess energy intake causes obesity, several obesity-associated diseases, and reproductive failure, and it reduces longevity (Solleveld et al. 1986; Keenan et al. 1994, 1995a,b, 1996, 1997). There is evidence that rapid growth rates and obesity are associated with an increased occurrence of spontaneous and induced tumors in laboratory animals (Suzuki et al. 1991; Ross and Brass 1965). The total energy—gross energy (GE)—available in food can be determined by complete combustion in the calorimetric bomb. As shown in Figure€11.3, the animals cannot use all GE; some energy is lost in feces, urine, methane, and hydrogen. The remaining energy, called metabolizable energy (ME), is the energy available for maintenance, growth, reproduction, lactation, and work (Figure€11.3). The ME value of a food varies according to the species of animal to which it is given. For rats and mice, ME values are almost similar because these animals digest foods to much the same extent, and losses in the form of methane are negligible (Chwalibog et al. 1998b). However, for herbivorous animals like guinea pigs and rabbits, the same foods are digested to a lesser extent, and due to the fermentation process, more methane is lost, reducing the ME value. Part of the ME is lost as heat increment caused by the energetic expenses of digestion and absorption of nutrients and due to energetic inefficiency of the reactions by which absorbed nutrients are metabolized. The deduction of the heat increment of a food (dietary-induced thermogenesis) from its ME gives the net energy (NE) value of the food. The NE is the remaining part of food energy used for different life processes and is therefore the unique measure of the energetic value of the food.
84550.indb 316
11/11/10 6:26:00 PM
Nutrient Requirements, Experimental Design, and Feeding
317
GE Gross Energy FE Feces Energy DE Digestible UE Urine + CH4 ME Metabolizable
HE Heat energy
HI Heat Increment
HEm Heat of Maintenance
RE Retained Energy
NE Net Energy
Figure 11.3â•…The partition and losses of food energy in the body.
An energy system based on NE is the most accurate system to evaluate energetic values of foods, but it is difficult to access. From a practical point of view, it is advisable to use ME for the evaluation of foods for laboratory animals. Relatively few types of foodstuffs are used for laboratory animals—primarily grain and oilseed cakes, in which the utilization of ME varies little. Extensive studies on different animals have shown that ME can be calculated with reasonable certainty on the basis of the digestible quantities of nutrients as follows (Schiemann et al. 1971):
ME, kJ = + + +
Rats 18.4 39.4 15.2 17.5
Rabbits 18.2 39.5 18.8 17.1
× digest. crude protein, g × digest. fat, g × digest. crude fiber, g × digest. nitrogen-free extract, g
For example, ME content in a diet for rats would be calculated from the dietary content of digestible nutrient as ME, kJ =18.4 × digestible crude protein, g + 39.4 × digestible fat, g + 15.2 × digestible crude fiber, g + 17.5 × digestible nitrogen-free extract, g The calculation of ME will always be subject to some uncertainty concerning the energy constants for the respective nutrients. In practice, the same accuracy can be obtained by using only the
84550.indb 317
11/11/10 6:26:01 PM
318
Handbook of Laboratory Animal Science
content of organic matter (OM) as a basis. With growing rabbits, Thorbek and Chwalibog (1981) found the following relation between ME and digestible OM:
ME, kJ = 18.7 × digest. OM, g
Because the digestibility of organic matter generally is about 65%, ME in food for rabbits and guinea pigs could be calculated as ME, kJ = OM, g × 0.65 × 18.7. For rats, mice, and hamsters, a factor of 19 kJ/g digestible OM can be applied. Assuming that the digestibility of organic matter is 85%, the following calculation could be used: ME, kJ = OM, g × 0.85 × 19 (Chwalibog 1989). Usually, the ME value of a certain diet is given by the manufacturer, but it is often unclear whether this is an analyzed value or calculated by using fixed, theoretical constants. If the energy value of foods is expressed in ME units, the requirements must be expressed in the same units; thus, we need to know the animal’s maintenance requirement for metabolizable energy (MEm) as well as the efficiency of ME utilization for growth, pregnancy, and lactation. Theoretically, the maintenance requirement is defined as the amount of energy necessary to balance anabolism and catabolism, yielding an energy retention around zero (Chwalibog 1991). For laboratory animals, there are few empirical results concerning MEm (National Research Council 1995; Eggum and Beames 1986). MEm values for rats and mice are usually suggested to be around 100 kcal (420 kJ) per LW, kg0.75 (Cañas et al. 1982; Chwalibog et al. 1998a). Although this constant value has been used as the measure of MEm in laboratory animals, it is debatable whether there is any constant value of MEm independent of nutritional, genetic, and environmental conditions (Chwalibog 1991; Blaxter 1989). There is substantial information about ME utilization for growth, pregnancy, and lactation in farm animals (Cañas et al. 1982); surprisingly, these values are not used for laboratory animals. The standards for energy requirements of laboratory animals are based on the measurements of voluntary food intake at different physiological states (National Research Council 1995) or on equations produced by Clarke et al. (1977). According to these equations, requirements of ME for maintenance of rats are 450 kJ/LW, kg0.75 and requirements for the other physiological states (inclusive of maintenance) are calculated as growth = 1200 kJ/LW, kg0.75, pregnancy = 600 kJ/LW, kg0.75, and lactation€= 1300 kJ/LW, kg0.75. Requirements The primary element in the evaluation of nutritional requirements is knowledge of the ability of animals to transform nutrients and energy obtained from a diet into body components and products. In spite of much information available on different aspects of laboratory animal nutrition, only a few methodical investigations concerning nutrient and energy balances during growth (Klein and Hoffmann 1989), pregnancy (Imai et al. 1986), and lactation (Cañas et al. 1982) have been conducted. Thorbek and colleagues (1982) conducted a series of methodical studies concerning protein and energy metabolism in growing rats for 5 months after weaning. The results of these studies furnish valuable data concerning nutrient and energy utilization and accretion during growth, providing the necessary basis for calculation of the requirements. The experiments were performed with male albino rats fed ad libitum on nonpurified commercial diets. During the growth period from 5 to 18 weeks of age, retained protein (RP) followed the pattern for growing farm animals (Tauson et al. 1998), increasing RP to about 1.7 g/day per animal between 7 and 8 weeks of age and then decreasing to a constant plateau of 0.5 g/day at 16–18 weeks of age. Fat retention (RF) gradually increased above 2 g/day until 10–14 weeks of age, after which a slow decrease was observed. The amount of retained energy (RE), a sum of energy retained in protein and fat, reached the highest level of 125 kJ/d at the age of 8–10 weeks, then gradually decreased to about 80 kJ/d due to decreasing protein and fat retention.
84550.indb 318
11/11/10 6:26:02 PM
Nutrient Requirements, Experimental Design, and Feeding
319
Table€11.3╅Requirement for Digestible Protein for Growing Rats in Relation to Live Weight (LW) and Retained Protein (RP) Age (week)
Requirement (g/day) LW (g)
5 8 10 18
RP (g/d)
100 220 290 400
Maintenance
Growth
Total
0.4 0.8 1.0 1.3
1.7 2.3 1.7 0.7
2.1 3.1 2.7 2.0
1.3 1.7 1.3 0.5
Based on the presented values, the requirements for protein and energy in the growing rat are calculated by the factorial approach demonstrated in Tables€11.3 and 11.4. The calculations are performed for weeks 5, 8 (around maximum RP), 10 (around maximum RF), and 18 (mature animals). Protein requirement is estimated assuming an average protein digestibility of 80% (Edwards and Dean 1985) and that endogenous nitrogen excretion in urine and feces is related to the live weight of animals (Chwalibog 2000). Metabolizable energy required for maintenance (MEm) is calculated from the following function: MEm, kJ/d = 32.5 + 251 × LW, kg 0.75. The amount of ME required for energy retention in protein (RPE) and in fat (RFE)—that is, for growth—is calculated with the efficiencies of ME utilization as 0.50 and 0.77 for RPE and RFE, respectively (Thorbek et al. 1983) (Tables€11.3 and 11.4). The pattern of protein retention showed that the highest requirement for digestible protein was at 8 weeks of age, with relatively low fat retention at 5 weeks of age. The total requirement for ME was markedly lower than in the latter part of the growth period. The presented values demonstrate the pattern of requirement during growth, but for practical diet formulation, it is interesting to note the changes in the required concentration of DP per 100 kJ ME. It decreased linearly from 1.2 g/100€kJ at 5 weeks to 0.7 g/100 kJ at 18 weeks of age, thus indicating the necessity of providing diets with different concentrations of protein during the growth period. Nevertheless, only one dietary composition is typically used for the entire growth period for laboratory animals. Allowances Knowing the amount of nutrients and energy required by animals is necessary to establish nutritional allowances for laboratory animals. There is a clear distinction between the terms “requirement” and “allowance.” The requirement is a statement of what animals, on average, require for a particular function. The allowance, however, is greater than this amount by a safety margin designed principally to allow for variations in requirements between individual animals and to account for possible variations of nutrient content in the same foods or diets. Table€11.4â•…Requirement for Metabolizable Energy for Growing Rats in Relation to Live Weight (LW), Retained Energy in Protein (RPE), and Retained Energy in Fat (RFE)
84550.indb 319
Age (week) 5 8 10 18
LW (g)
RPE (kJ/day)
RFE (kJ/day)
100 220 290 400
30 40 30 12
20 80 95 70
Requirement (kJ/day) Maintenance 80 110 130 160
Growth 90 180 180 120
Total 170 290 310 280
11/11/10 6:26:02 PM
320
Handbook of Laboratory Animal Science
As a result of the large number of factors that can influence dietary requirements, manufacturers add nutrients that are in excess of the estimated requirements (Lang and Harrell 2000). However, there are inconsistencies in the magnitude of these safety margins and, unfortunately, the distinctions between requirements and allowances are not strictly defined for laboratory animals. In many publications, it is not clear whether the “requirements,” or so-called nutritional standards, refer to requirements or to allowances. This is one of the main reasons for many of the discrepancies between different recommendations. Keeping in mind the possible confusion caused by the use of inconsistent terminology, the tables in this section present only a general outline of nutritional allowances recommended for common laboratory animals. For a detailed description of nutritional standards, the publications by Coates (1976), Clarke et al. (1977), the National Research Council (1995), and Eggum and Beames (1986) are suggested. Nutrient and energy allowances for laboratory animals are rarely expressed in terms of quantity per day; more typically, the means of the concentrations of nutrients and energy in the diet are used. In practice, most laboratories use only two different diets for each animal species: one during growth and maintenance (adult, nonproducing animals) and the other for pregnant and lactating animals. Knowledge concerning the requirements for individual amino acids for different laboratory animals and different life processes is limited. The greatest quantity of information exists for laboratory rats, although, depending on the source of the information, there is a considerably broad range of recommended values, as demonstrated in Table€11.5. In general, adult mice, rats, and hamsters require 70–120 g of crude protein per kilogram of diet, with 90% of dry matter for their protein supply. Depending on protein digestibility and biological value, this is equivalent, in the natural diet, to a supply of about 50–70 g digestible protein per kilogram of food. For the other body functions, like growth, pregnancy, and lactation, a supply of 200–240 g crude protein per kilogram of food may be recommended, corresponding to 120–140 g digestible protein for maintenance and productive functions. The recommended level of crude fat in the diet for adult animals is about 20 g/kg of diet and about 50 g/kg for productive animals. For the mouse, rat, and hamster, crude fiber in the diet should preferably not exceed 80 g/kg. On the other hand, in the guinea pig and rabbit, considerable amounts of Table€11.5â•…Range of Recommended Levels for Essential Amino Acids for Laboratory Rats Amino acid Arginine Histidine Isoleucine Leucine Lysine Methionine + cystine Phenylalanine + tyrosine Threonine Tryptophan Valine
g/100 g Protein
g/kg Diet
mg/kg0.75
g/MJ
5.0–6.0 2.5–3.0 5.0–6.0 7.5–8.0 6.0–9.0 5.0–10.0 6.0–10.0 4.0–4.5 1.2–1.5 5.0–6.0
4.3 2.8 6.2 10.7 9.2 9.8 10.2 6.2 2.0 7.4
0–10 0–17 30–49 16–64 10–33 20–43 16–52 20–54 5–10 18–47
0.39 0.19 0.32 0.49 0.45 0.30 0.52 0.32 0.10 0.39
Sources: g/100 g protein: Chwalibog, A. et al. 1998a. Comparative Biochemistry and Physiology 121:423–429; g/kg diet: National Research Council. 1995. Nutrient Requirements of Laboratory Animals, 4th ed. Washington, D.C.: National Academy of Sciences; mg/kg0.75 per day: Owens, F. N., and J. E. Pettigrew. 1989. In Absorption and Utilization of Amino Acids, ed. M. Friedman, chap. 2. Boca Raton, FL: CRC Press, Inc.; g/MJ, ME per day: Eggum, B. O., and R. M. Beames. 1986. In Laboratory Animals, ed. E. J. Ruitenberg and P. W. J. Peters, chap. 9. Amsterdam: Elsevier Science Publishers B.V.
84550.indb 320
11/11/10 6:26:02 PM
Nutrient Requirements, Experimental Design, and Feeding
321
cellulose, hemicellulose, and pectin are broken down in the large intestine (50–75% of the digestion capacity of ruminants). For these animals, a content of 100–200 g crude fiber/kg is recommended. As mentioned previously, the nutritional standards for laboratory animals are not comprehensive. A substantial amount of work is required to establish nutrient and energy requirements and, consequently, nutritional allowances for different species, life processes, housing conditions, and research purposes. The presented values for nutritional allowances are therefore to be used as general recommendations that should be revised when new scientific data become available. In order to improve standardization, the use of the National Research Council guidelines (1995) for defining the minimum requirements per species is recommended. Types of Diets Natural-Ingredient Diets The pelleted diets that comprise the typical standard diet used in most laboratory animal facilities are, in most instances, made from natural ingredients. Manufacturers can produce diets according to variable or fixed formulas. In a variable formula, the final product levels are kept as constant as possible (i.e., the goal is to maintain the level of nutrients in the end product as constant as possible). This means that the quantity of each individual ingredient is adjusted for variations in nutrient levels of the raw materials (British Association of Research Quality Assurance 1992). In a fixed formula, the recipe does not change for a particular type of diet (i.e., the same proportions of raw material ingredients are used each time a batch is produced) (British Association of Research Quality Assurance 1992). Because natural ingredients can differ, depending on weather conditions, soil, etc., the commercially available pelleted natural-ingredient diets are also subject to variation. This has to be taken into consideration when one uses so-called standard diets. Natural-ingredient diets can also be divided into open- and closed-formula diets (Thigpen et al. 1999). In open-formula diets, all dietary ingredients and their concentrations are reported and should not vary from batch to batch. In closed-formula diets, the dietary ingredients used are reported, but the concentration of each dietary ingredient is not stated by the manufacturing company. The concentration of dietary ingredients may vary from batch to batch or with availability of ingredients (Thigpen et al. 1999). Natural-ingredient diets can be offered in several physical forms, including in meal form, as extruded pellets, or as expanded pellets. The extruded form is the conventional pelleted form. The expanded form is produced by forcing a wet mixture of the basal ingredients of the diet through a mold at high pressure, accompanied by injection of superheated steam. This process yields pellets that are more voluminous. It is claimed that expanded diets reduce food intake (higher availability of nutrients), increase food conversion, reduce microbial counts, and reduce wastage as compared to the conventional extruded pellets (Ward 2008). Manufacturers provide information on the dietary products in their catalogs and on their Web sites. The amount of information provided differs, but generally, information is provided on the ingredients that have been used and the nutrient and contaminant concentrations to be expected in these diets. Also, microbiological quality is often presented. Catalog values for nutrient concentrations have been established, but have not been standardized. The most common way to determine catalog values is to analyze a few production batches of the same diet and then present the average value in the catalog; however, no information on batch-to-batch variation is typically included. It is a well-known fact that dietary batches produced by different manufacturers can differ (between-brand variation), yet batches produced at the same production unit can also differ, based on variation in natural ingredients (between-batch variation) (Beynen and Coates 2001). In the
84550.indb 321
11/11/10 6:26:02 PM
322
Handbook of Laboratory Animal Science
Table€11.6â•…Between-Brand Variation: Variation in Catalog Values of Selected Nutrients of Nine Brands of Minipig Maintenance Diets Nutrient Metabolizable energy (MJ/kg) Crude protein (%) Lysine (%) Isoleucine (%) Cu (mg/kg) Mn (mg/kg) Fe (mg/kg) Se (µg/kg) Vitamin B (mg/kg) Vitamin C (mg/kg)
Lowest Value
Highest Value
9.5 13.2 0.57 0.40 10.0 50.0 130.0 150.0 3.0 0.0
11.2 15.4 0.94 0.74 38.0 160.0 770.0 520.0 18.0 100.0
Source: Modified from Ritskes-Hoitinga, J., and P. Bollen. 1997. Pharmacology and Toxicology 80 (SII): 5–9.
case of a production firm that produces a similar type of diet at different production sites, betweensite variation must be expected as well, because natural ingredients used at the different sites are unlikely to be identical. Even if the natural ingredients used are exactly the same, variations in storage and production methods may still result in variation of the final product. Between-Brand Variation By definition, natural ingredients are not constant in their composition, and manufacturing companies use different natural ingredients in varying amounts; therefore, the so-called “standard” diets for, for example, growth and maintenance, show variation. In Table€11.6, catalog values for a number of nutrients from minipig maintenance diets from nine different manufacturing companies are given, illustrating the variation in composition between brands (J. Ritskes-Hoitinga and Bollen 1999). Variations in dietary composition can result in different experimental data. By feeding female Wistar rats 10 brands of standard maintenance rodent diets, varying experimental results were obtained (Table€ 11.7) (J. Ritskes-Hoitinga et al. 1991). This demonstrates that historical control groups cannot always be used reliably and that, most of the time, there is a need for a concurrent control group. Dietary differences are likely to affect the outcomes of experiments, but are unlikely Table€11.7â•…Between-Brand Variation: Variation in Seven Parameters after Feeding 10 Different Brands of Rodent Maintenance Diets Ad Libitum to Female Outbred Wistar Rats for a 4-Week Period Parameter Body weight (g) Food intake (g/day) Water intake (mL/day) Kidney calcification score (0–3) Urine production (mL/day) Urine pH Caecal weight (g/100 g BW)
Lowest Mean Group Value 137.0 11.0 18.2 0.0 8.6 6.3 1.1
Highest Mean Group Value 163.0 15.6 24.8 1.3 17.0 8.4 3.9
Source: Modified from Ritskes-Hoitinga, J. et al. 1991. Lab Anim 25:126–132. Note: Six animals per group.
84550.indb 322
11/11/10 6:26:02 PM
Nutrient Requirements, Experimental Design, and Feeding
323
Table€11.8╅Within-Brand Variation: Comparison of Catalog Values of a Maintenance Minipig Diet with Analyzed Values from a Number of Nutrients from Five Dietary Batches from the Same Manufacturing Company Parameter Crude oil (%) Crude protein (%) Crude fiber (%) Calcium (%) Phosphorus (%) Iron (mg/kg) Zinc (mg/kg) Vitamin E (mg/kg)
Catalog Value 2.4 13.9 11.6 1.02 0.68 130.0 110.0 53.0
Lowest Value 2.2 14.1 7.1 0.89 0.53 130.0 109.0 32.0
Highest Value 3.4 16.5 13.2 1.13 0.73 299.0 200.0 216.0
Source: Modified from Ritskes-Hoitinga, J., and P. Bollen. 1997. Pharmacology and Toxicology 80 (SII): 5–9.
to have a significantly interacting influence on the outcomes of all experiments. This will depend on the topic of the study and how dietary components interfere. Within-Brand Variation If one sticks to the same type of diet from the same manufacturing company, one will be inclined to think that the dietary composition will remain the same over time. However, as mentioned previously, natural ingredients vary in their composition (Pullar and Webster 1974). As a consequence, dietary batches produced from “the same” natural ingredients will also show variation in nutrient analyses over time. Table€11.8 illustrates the variation in five batches of minipig maintenance diets from the same manufacturing company (J. Ritskes-Hoitinga and Bollen 1997). Generally speaking, nutrient levels in commercial diets fulfill requirements of all essential nutrients more than sufficiently, at least when diets have been transported and stored under proper conditions. If one looks at the vitamin E levels in Table€11.8, a batch level of 32 mg/kg will fulfill the needed minimum requirement. However, vitamin E is unstable under higher environmental temperatures, so inappropriate storage may quickly reduce the vitamin E level, potentially resulting in a deficiency by the time of feeding. In another batch, the vitamin E level was as high as 216 mg/kg. Because vitamin E is an antioxidant, such a high level can theoretically provide protection—for instance, if one works with oxidative stress models. Due to this batch-to-batch variation, it is advisable always to buy diets with a batch analysis certificate. Although this will increase costs, at least basic knowledge on nutrient content, microbiology, and contaminants of a particular batch is supplied. This will also provide an opportunity to reject a batch before the start of a particular experiment or to exclude dietary composition as a possible confounding factor if the batch analysis yields results outside those expected. In the case of studies performed under good laboratory practice (GLP) guidelines, it is mandatory to buy diets with a batch analysis certificate in order to document all necessary details of the experiment performed. Table€11.9 gives an example of a typical batch analysis certificate. Purified Diets Standardization Purified or semipurified diets (also named synthetic or semisynthetic diets) are defined as being formulated with a combination of natural ingredients, pure chemicals, and ingredients of varying
84550.indb 323
11/11/10 6:26:03 PM
324
Handbook of Laboratory Animal Science
Table€11.9╅An Example of a Batch Analysis Certificate
Source: Modified from Ritskes-Hoitinga, J., and P. Bollen. 1997. Pharmacology and Toxicology 80 (SII): 5–9.
degrees of refinement (Beynen and Coates 2001). This results in diets having a much more standardized composition than natural-ingredient diets; consequently, this leads to (more) reproducible results and thereby more responsible use of laboratory animals. In a study evaluating the influence of dietary P level on the induction of nephrocalcinosis (kidney calcification) in female Wistar rats, where purified diets were used, a 0.2% P level prevented the occurrence of kidney calcification in each study, while P levels between 0.5 and 0.6% induced a severe degree of nephrocalcinosis (J. Ritskes-Hoitinga 1992). Specially prepared, purified diets make it is possible to approximate the desired nutrient levels closely and reproducibly. Table€11.10 shows the between-batch variation for eight batches of purified diets used in rat experiments in which the dietary etiology of nephrocalcinosis was studied. The levels obtained are Table€11.10╅Between-Batch Variation in Eight Batches of Purified Diets Nutrient Calcium (%) Phosphorus (%) Magnesium (%)
Targeted Concentration
Lowest Value
0.50 0.40 0.05
0.52 0.40 0.06
Highest Value 0.57 0.43 0.06
Source: Modified from Ritskes-Hoitinga, J. 1992. Thesis, Utrecht University, the Netherlands.
84550.indb 324
11/11/10 6:26:05 PM
Nutrient Requirements, Experimental Design, and Feeding
Phytoestrogen Levels in Lab Animal Chow
Isoflavone Levels in µg/g Chow
500
400
300
200
100
0
325
AIN 76A soy Altromin 1324 CRF-1 HSD 7012 MF pellet diet NIH-07 NIH-31 NTP-2000 PMI 5001 PMI 5002 PMI 5015 PMI 5058 RM1 RM3 Ssniff R/M-H AIN 76A AIN 93M HSD 2014S HSD 2919 PMI 5k96 Altromin C 1000 Zeigler 5412-01 HSD 2016S NTP 88 Zeigler 5412-00
Figure 11.4â•…A color version of this figure follows page 336. Phytoestrogen levels in laboratory animal chow. (Modified from Nygaard Jensen, M., and M. Ritskes-Hoitinga. 2007. Laboratory Animals 41: 1–18.)
close to the desired concentrations, and between-batch variation is relatively small. These highly standardized and reproducible levels usually can be attained only when purified diets are used. Figure€11.4 illustrates the variation in isoflavone (phytoestrogen) levels that can be found in natural ingredient and purified diets (Nygaard Jensen and Ritskes-Hoitinga 2007). Only the (purified) diets that are formulated without soy reach negligible isoflavone levels. Isoflavone levels around 300–400 µg/g and higher are known to interfere with the results of different types of experiments. Achieving Desired Nutrient Levels In case it is desirable to achieve low or very low levels of certain nutrients, it is often necessary to use purified diets; due to contamination, it is virtually impossible to achieve these low concentrations with natural-ingredient diets. For rodent toxicity studies, it is particularly advisable to use purified diets composed according to the American Institute of Nutrition guidelines (Reeves et al. 1993): the AIN93 G (G = growth) or AIN 93 M (M = maintenance) diets. This will lead to a constant dietary composition for each study. These guidelines are a “cookbook recipe” that ensures that all the nutrient requirements for rodents are fulfilled (National Research Council 1995). However, the vitamin B12 level is lower in the AIN-93 diet than that advised by the National Research Council (1995), so it is advised to double the vitamin B12 level in the AIN diet in order to achieve the minimum requirements. Use of these purified diets will yield more reproducible results that should reduce interlaboratory variation. Haseman, Huff, and Boorman (1984) and Roe (1994) clearly demonstrated the great variation that may be found in control groups in 2-year toxicological studies (Table€ 11.11). The mammary tumor incidence in control groups could vary from 2 to 44%. Part of this large variation is the result of the variation in dietary composition when natural-ingredient diets are used.
84550.indb 325
11/11/10 6:26:07 PM
326
Handbook of Laboratory Animal Science
Table€11.11╅Variation in Results among Control Groups in 2-Year Toxicological Studies in Five Different Laboratories Fischer 344 Rats Number of studies Percent survival to 2 years Liver neoplastic nodules (%) Mammary fibroadenomas (%)
Results in Males
Results in Females
41 44–78 0–12 0–8
42 50–86 0–12a 2–44a
Sources: Modified from Haseman, J. K., J. Huff, and G. A. Boorman. 1984. Toxicologic Pathology 12:126–135; Roe, F. C. J. 1994. Lab Anim 28:148–154. a Significant interlaboratory variation (p < 0.05).
Fulfilling Essential Needs—Avoiding Toxic Levels In order to fulfill the essential needs for each species, the National Research Council documents provide the best documented scientific evidence as all available literature on nutrient requirements has been assembled (National Research Council Collection). For minipigs, the scientific documentation for nutrient requirements is still uncertain and insufficient and is therefore being investigated (J. Ritskes-Hoitinga and Bollen 1997, 1999; Bollen et al. 2006). When reducing specific nutrient levels, one must be aware that certain nutrients are essential and therefore should be present at the minimum necessary levels in order to avoid deficiency problems, unless deficiency is a component of the study. It should be noted that a level just below the minimum requirement does not necessarily lead to deficiency symptoms but may still affect experimental results. The literature should be consulted to facilitate responsible choices of the degree to which levels of certain nutrients can be reduced without causing serious health or welfare problems or the premature deaths of the animals. Obviously, this will also compromise experimental results. Methionine is an example of a nutrient with a narrow safety margin, making it necessary to modify dietary concentrations within very strict boundaries in order to prevent the premature death of subjects, especially mice. In a study by DeWille and colleagues (1993), the influence of dietary linoleic acid on mammary tumor development in transgenic mice was studied. Three levels of dietary linoleic acid were given: 0, 1.2, and 6.7%. Mammary tumor development was reduced significantly in the group receiving 0% linoleic acid compared to the other two dietary groups (Figure€11.5). Because linoleic acid is an 60
MT (%)
50 40 30 20 10 0
0.0 LA
1.2 LA
6.7 LA
Figure 11.5â•…Relationship between dietary linoleic acid levels and mammary tumor frequency in transgenic mice. MT: mammary tumor frequency (%: number of animals with mammary tumors); LA: dietary linoleic acid level (%). Mammary tumor frequencies on the two highest levels of linoleic acid were significantly higher than on the 0 level of linoleic acid. The frequencies on the two highest dietary linoleic acid levels were not significantly different from each other.
84550.indb 326
11/11/10 6:26:09 PM
Nutrient Requirements, Experimental Design, and Feeding
327
essential fatty acid necessary for the development of cell membranes (National Research Council 1995), this cannot be considered a reliable, valid control group. A diet that does not contain linoleic acid compromises the general health of the subjects; there were 25 animals in the 0% group at the start of the study, yet data from only 15 animals were presented in the results section (DeWille et al. 1993). If cell membranes cannot develop, it is questionable whether tumors can arise. Much of what applies when nutrient levels are lowered is also applicable to increasing nutrient levels. The literature must be examined to determine toxic levels for a certain nutrient for a certain species because toxic levels will obviously negatively influence animal health and welfare and, thereby, experimental results. Some nutrients, like selenium and methionine, have a narrow “safety margin,” and relatively small increases in dietary levels will lead to toxic effects. Information on toxicity levels of nutrients is available in documents published by the National Research Council (1987; NRC Collection). Purified Diets versus Species One of the disadvantages of purified diets is that palatability is often lower than that of natural-ingredient diets. This may induce the need for an acclimatization phase, in which a naturalingredient diet is gradually replaced by the purified diet. It is the authors’ experience that this is species dependent; rats will usually accept purified diets readily, and a 1-week acclimatization period to this diet is sufficient. In mice, a drop in body weight in the first week and irregular food intake over 2–4 weeks may occur when an acute change from a natural-ingredient diet to a purified diet has been made. Feeding purified diets containing “unnatural” components, like fish oil, to (herbivorous) rabbits required a 6-month, gradual adaptation phase. During this period, the purified diet gradually replaced the natural-ingredient diet (J. Ritskes-Hoitinga et al. 1998). If rabbits stop eating, some grass meal should be included in addition to the diet to increase palatability. It is important to keep rabbits eating in order to prevent them from dying within 2 or 3 days due to fasting-induced hyperlipemia. When dietary ingredients and animal models are chosen, one must be aware of possible speciesspecific characteristics. Increasing concentrations of fish oil fed to the rabbits mentioned before caused liver pathology, coinciding with a higher degree of aortic atherosclerosis (Figure€ 11.6). Therefore, other animal species, such as (omnivorous) pigs, are considered better animal models in which to study the effects of fish oil. Points to Consider when Preparing Purified Diets One must account for species-specific requirements when purified diets are used. There is a greater risk of creating shortages of unknown essential nutrients when purified diets, rather than natural-ingredient diets, are used because unknown nutrients can be present as “natural contaminants” in natural-ingredient diets. Chromium and vanadium are examples of substances that are possibly essential nutrients for rodents that were not described until the latest revision of the recommendations for rodents by the National Research Council in 1995. The selection of certain refined ingredients can be critical. An example was the use of a shorttype cellulose fiber (Arbocel R B-00) in a purified diet, which caused intestinal obstruction and death in rats (Speijers 1987). The fiber content of the diet was 10.5%. Replacement of this short-type fiber by a longer type (Arbocel R B-200) resulted in the disappearance of the intestinal problems. Oils like fish oil oxidize readily (due to their high content of polyunsaturated fatty acids), and therefore diet mixtures containing such oils need to be prepared fresh each day. Oils should be stored under liquid nitrogen until the day the diet is mixed and fed. Antioxidants can be added to the diet as well.
84550.indb 327
11/11/10 6:26:09 PM
328
Handbook of Laboratory Animal Science
12
20FO
% Aorta Plaque Surface
10
10FO
8 7LN 6
0FO
1FO
4 40SF 2
0
0.6
0.8
1
1.2 1.4 1.6 Score Liver Pathology
1.8
2
2.2
Figure 11.6â•…Relationship between the amount of atherosclerosis in the aorta and liver pathology score in a rabbit study examining the influence of fish oil on atherosclerosis: correlation between group mean scores for liver pathology (X-axis) and relative aorta plaque area (Y-axis). N = 6, correlation coefficient = 0.96, P = 0.003. 20FO = 20 energy% fish oil diet; 10FO = 10 energy% fish oil diet; 1FO = 1 energy% fish oil diet; 0FO = 0 energy% fish oil diet; 7LN = 7 energy% linseed oil; 40€SF€= 40 energy% sunflower seed oil diet. All diets contained 40 energy% total fat. (From RitskesHoitinga, J. et al. 1998. Food and Chemical Toxicology 36:663–672. With permission.)
Pelletability and Feeding Devices Due to the composition of the purified diet, it is often difficult to pellet these diets properly; this frequently results in pellets with a structure that is “too loose,” leading to increased spillage. In most cases, purified diets are given in a powdered form. Special feeding devices (Figure€11.7) have been developed to facilitate the feeding of purified diets to groups of mice and rats in Macrolon cages without having too much spillage. Feeding purified diets in open food hoppers without somehow restricting access makes it possible for animals to play with the food and waste much of it. This also occurs when metabolic cages are used for collecting feces and urine. If the animal is relatively small in comparison to the hopper, the animal can hide inside the dark food hopper to avoid the brighter environment of the rest of the metabolic cage. Food from the hopper may stick to the animal’s fur, diminishing the reliability of the food intake measurements. Food that sticks to the fur may also contaminate urine and feces collections. Provision of adapted enrichment devices is one potential solution to this problem, promoting the well-being of the animals by supplying a darker environment and simultaneously increasing the precision of food intake measurements, and urine and fecal collections. The least spillage is achieved when a metabolic cage that is of the proper dimensions for the age and size of the particular species is used.
84550.indb 328
11/11/10 6:26:13 PM
Nutrient Requirements, Experimental Design, and Feeding
329
Figure 11.7â•…Examples of feeding devices suitable for feeding powdered diets to rats (left) and mice (right) housed in Macrolon cages. (Feeding devices kindly provided by Scanbur A/S, Denmark.)
Contaminants Contaminants may be defined as undesirable substances (usually foreign) that, when present at a sufficiently high concentration in the food, may affect the animal and therefore the outcome of the experiments (British Association of Research Quality Assurance 1992). Possible contaminants include industrial chemicals, pesticides (e.g., DDT), plant toxins, mycotoxins (e.g., aflatoxin), heavy metals, nitrates, nitrites, bacteria, and bacterial toxins (British Association of Research Quality Assurance 1992). Specific critical parameters will vary from user to user and will depend upon the objectives of the study in which the diet is used. Investigators must determine the critical contaminants for a particular study; it is not feasible to identify and analyze for every possible contaminant. On the basis of this information, researchers can establish the identities and maximum levels of contaminants that can be allowed in the diet. Several documents state general guidelines for maximum allowed concentrations of contaminants (British Association of Research Quality Assurance 1992; GV-Solas 1980). One of the guidelines that provides maximum limits and is referred to by toxicologists across the globe was issued by the U.S. Environmental Protection Agency (Environmental Protection Agency 1979). Dietary production firms have often developed their own maximum limits. In addition to contaminants, several nutrients can also be toxic (sometimes at concentrations that are not far above the dietary requirements). This information can be found in the National Research Council documents on nutrient requirements (http://dels.nas.edu/banr/nutrient_requirements_series.shtml). Since different guidelines may state different problematic levels for contaminants, it can be difficult to choose the “correct” maximum tolerated levels. First, it has to be decided which guidelines are most appropriate in the specific experimental setting, and then those guidelines must be chosen as the institutional policy. Specific institutional guidelines may also be developed. Second, for each group of experiments of similar type, a literature search must be performed to determine whether there are contaminants that will interfere with the specific purpose of the study. Thus, specific maximum levels of individual contaminants can be established and when diets are ordered for specific types of experiments. It should be clear that a batch analysis certificate report is necessary in order to be able to judge the levels of contaminants in a specific batch of diet. If particular contaminants are not
84550.indb 329
11/11/10 6:26:18 PM
330
Handbook of Laboratory Animal Science
included in routine analyses, it may be necessary to request separate analyses for these specified contaminants. If the level of any contaminant is above the acceptable level, then a different diet should be used or a different acceptance criteria should be adopted (British Association of Research Quality Assurance 1992). Strong justification will be needed for adopting altered acceptance criteria, especially in regulatory studies, where government inspectors are likely to investigate such alterations. Contaminants can be virtually avoided using purified ingredients and diets. In natural-ingredient diets, more contaminants in higher levels are likely to be found. Quality Considerations Storage Conditions Diets must be kept under suitable storage conditions (a cool, dry place impervious to entry by wild rodents) at all times, at the manufacturer and the user, to ensure that they remain within the specifications until the recommended expiration date. Information concerning proper storage conditions is provided by the manufacturer. Providing diets ad libitum means that diets are exposed to room temperature 24 h/day, potentially causing a temperature-associated decrease in some nutrients. Therefore, diet present in the cages at room temperature must be discarded regularly and replaced by fresh food (e.g., one to two times per week for “regular” natural-ingredient diets—the guidelines from the manufacturer should be followed). When diets that include highly unsaturated fats, like fish oil, are used, it may be necessary to provide fresh food every day. Preservatives like butyl hydroxytoluene (BHT) may be added in order to prevent oxidation. BHT is added only when highly oxidative oils are used. During transport, diets must remain dry and protected from damage, heat, and wild rodents. No chemicals should be transported or stored with the diets (British Association of Research Quality Assurance 1992). When diets arrive at a facility, they have to be inspected for damage or contamination, and discarded if unfit for use. Sterilization Diets can be sterilized by gamma irradiation or autoclaving in order to reduce microbiological contamination levels further. This is especially relevant for barrier units. In addition to reducing microbiological contamination, these processes reduce the contents of unstable nutrients like vitamins A and E as well. Manufacturers offer special diets that are meant to be sterilized. In these diets, the content of unstable nutrients has been increased to ascertain that these levels still live up to the requirements after the sterilization process. One example of the importance of such special diets was presented by Wyss-Spillmann and colleagues (1997). Briefly, alopecia, reduced weight gain, and decreased production developed in a breeding colony of rats in a barrier unit due to a malfunction of a valve of the autoclave at the entrance of the barrier unit. Moving the animals from the barrier unit to the conventional unit caused the problems to disappear. Food analyses revealed that vitamins A, B1, and B6 were present only in deficient trace amounts in the diet after autoclaving in the malfunctioning equipment. The problems disappeared once the autoclave was repaired. Pellet Hardness The feeding of pellets that were too hard was found to reduce the growth of preweaned mice (Koopman et al. 1989b). Hardness of pellets is measured as the amount of pressure, in kilopond (kp) weight, that is required to crush the pellet. In a single type of diet, pellet hardness can vary
84550.indb 330
11/11/10 6:26:18 PM
Nutrient Requirements, Experimental Design, and Feeding
331
from 4 to 50 kp/cm3 (Koopman et al. 1989b). A value of greater than 20 is considered problematic, making it difficult for preweaned mice to obtain enough food, potentially resulting in reduced growth. Part of this effect is suggested to be mediated through the effect on the mother (Koopman et al. 1989a). Feeding Schedules Ad Libitum Intake and Isocaloric Exchange A diet that is offered ad libitum is available at all times. As discussed earlier, voluntary food intake is, in principle, determined by the energy need of the animals. Fat has a much higher energy content than both protein and carbohydrates, so adding fat to the diet cannot be done without careful consideration. Only adding fat to a control diet, making it the test diet, will lead to an increase in the metabolizable energy content of the test diet (compared to the control diet). Because animals eat according to energy need, the food intake of the test diet (in grams) under ad libitum conditions will be lower than that of the control diet. This means that the intake of all nutrients in control and test groups will differ, making it impossible for a valid comparison of these two groups. It will not be possible to determine whether the observed effects are the result of the fat addition alone. To make a more valid interpretation of the effect of fat addition, isocaloric exchange can be executed. If a certain amount of fat is to be added to a diet, the amount of energy the fat presented is calculated, and then the same amount of energy is withdrawn from the diet by subtracting an appropriate amount of carbohydrates. If performed properly, this will result in diets with similar nutrient intake (grams) in control and test animals, except in terms of fat and carbohydrate (Table€11.12) (Beynen and Coates 2001). Restricted Feeding Restricted feeding refers to restricting the amount of food to which animals have access (in contrast to ad libitum feeding), while still ensuring nutritional adequacy (Hart et al. 1995). This implies that the amount of energy is restricted. With restricted feeding, it must be guaranteed that the feeding level provides sufficient essential nutrients. By feeding rodents restrictedly, instead of ad libitum, remarkable improvements in health and life expectancy have been achieved (Hubert et al. 2000; Masoro 2005). Feeding restrictedly has more positive effects on health than any changes in dietary composition—for instance, to a higher fiber content under ad libitum conditions (Hart et al. 1995). Ad libitum feeding is still considered “normal” practice for rodents; however, it is considered bad veterinary practice for pigs, monkeys, rabbits, and dogs because these species will become obese (Hart et al. 1995). The fact that rodents are still being fed ad libitum probably has more to do with economic and practical considerations than with scientific reasons. Ad libitum feeding, as opposed to moderately restricted feeding (75% of ad libitum intake), has a clearly negative impact on rodent health, shortening survival time, increasing cancer incidence, shortening cancer latency periods, and increasing the incidence of degenerative diseases in kidney and heart (Hart et al. 1995). These effects have been found to be very reproducible. Energy restriction resulted in a reduction in the incidence of mammary tumors by 55% as compared to ad libitum fed controls, as analyzed in a meta-analysis by Dirx et al. (2003). Dietary restriction leads to a reduced incidence not only of “spontaneous” tumors, but also of induced tumors (Zhu et al. 1997). It has been shown that the higher the degree of food restriction is, the longer the latency period and the lower the final incidence of MNU (1-methyl-1-nitrosourea)-induced tumors will be. In addition to influencing tumor development, dietary restriction has also been found to attenuate amyloid-beta deposition in the brains of transgenic mice modeling Alzheimer’s disease
84550.indb 331
11/11/10 6:26:18 PM
332
Handbook of Laboratory Animal Science
Table€11.12╅Examples of Expected Results of Low- and High-Fat Diet Formulations Fed to Rats Diet 1 Low Fat
Diet 2
Diet 3
Diet 4
High Fat
High Fat, Adjusted
High Fat, Adjusted
Diet ingredient Protein (g) Carbohydrate (g) Fat (g) Fiber (g) Mineral mix (g) Vitamin mix (g) Test compounds (g) “Inert” compound (g) Total (g) Energy value (kcal/g)
20 60 10 4 4 1 1 — 100 4.10
Energy (kcal/day) Food (g/day) Protein (g/day) Carbohydrate (g/day) Fat (g/day) Fiber (g) Mineral mix (g/day) Vitamin mix (g/day) Test compound (g/day) “Inert” compound (g/day)
82 20 4 12 2 0.8 0.8 0.2 0.2 —
20 40 30 4 4 1 1 — 100 5.10
20 15 30 4 4 1 1 — 75 5.47
20 15 30 4 4 1 1 25 100 4.10
82 15 4 3 6 0.8 0.8 0.2 0.2 —
82 20 4 3 6 0.8 0.8 0.2 0.2 5
Expected Intake 82 16 3.2 6.4 4.8 0.64 0.64 0.16 0.16 —
Source: From Beynen, A. C. and M. E. Coates. 2001. Nutrition and experimental results. In Principles of laboratory animal science, ed. L. F. M. Van Zutphen, V. Baumans, and A. C. Beynen. Amsterdam: Elsevier Scientific Publishers. With permission.
(Patel et€al. 2005). Moreover, beneficial effects of food restriction on diabetes pathologies (Bluher et al. 2003) and age-related muscle pathologies (Molon-Noblot et al. 2005) have been reported. In addition to having negative health consequences, ad libitum feeding necessitates an increase in the number of animals at the start of long-term toxicological studies, since many such studies require that 25 animals per sex survive a 2-year period. This increase is in conflict with at least one of the three Rs (reduction) and the compromised health of experimental rodents is in conflict with another (refinement). Therefore, the data suggest that moderate food restriction should become the new standard in laboratory facilities. Except for increased longevity, increased uniformity is expected to result when the amount of food per individual becomes standardized when animals are fed restrictedly, instead of ad libitum (M. Ritskes-Hoitinga et al. 2006). Power calculations on data from a study by Leakey, Seng, and Allaben (2003) and Leakey, Seng, Latendresse, et al. (2003) indicated that a 92% reduction in the numbers of mice could be achieved in order to prove a statistically significant difference in the response of one liver enzyme to chloral hydrate between test and control groups, when feeding restrictedly instead of ad libitum (Savenije et al. 2010). In addition, a dose-dependent increase in liver tumors by chloral hydrate was found in the restrictedly fed mice only, as in the ad libitum fed groups, liver neoplasm incidence was already increased at lower doses (Leakey et al. 2003a). Dietary restriction is generally associated with a higher sensitivity of animal models. The method of feeding also has an influence on the next
84550.indb 332
11/11/10 6:26:18 PM
Nutrient Requirements, Experimental Design, and Feeding
333
generation. Carney et al. (2004) measured the effects of the amount of diet fed to lactating CD rats. Dietary restriction levels of 90 and 70% of ad libitum led to a reduction in the variation of the rat pup weights as compared to ad libitum feeding. When lactating dams were fed at 50% of ad libitum, the variation in pup body weights increased, which may indicate that deficiencies had occurred. Keenan and colleagues (1999) state that ad libitum feeding (overfeeding) of rodents is, at present, one of the most poorly controlled variables affecting the current rodent bioassay. In past decades, variation in results has been shown to increase under ad libitum conditions and lead to the use of a higher number of animals than is truly necessary. Moderate dietary restriction (70–75% of adult ad libitum food intake) is advised as a method to improve uniformity, increase exposure time, and increase statistical sensitivity of chronic bioassays to detect true treatment effects (Keenan et al. 1999). However, moderate dietary restriction will only improve uniformity in individually housed animals, where there is control over individual food intake. A restricted amount of food in grouphoused animals is expected to increase variation due to differences in individual food intakes that are dependent on the social hierarchy in the group. It will be a challenge to develop restricted feeding methods for group-housed rodents to fulfill the social needs of the animals as well. Feeding systems that are designed to control the intake of individual animals within a social group have been used for quite some time in agricultural settings, so adaptation of these systems to the laboratory animal setting should not present insurmountable problems. Meal Feeding For some animal species, it is normal practice to meal feed. For example, Göttingen minipigs usually get a meal in the morning and another in the afternoon. Adult dogs are often fed one meal per day at a time that is chosen by the researcher (i.e., typically so as to interfere the least with experimental procedures and results). In order to conduct studies that mimic postprandial factors in humans, rats can be trained to eat meals (J. Ritskes-Hoitinga et al. 1995). Rats learn within a week that they are fed meals and will be able to eat what they need during these periods. Of course, this depends on the use of meal times that are sufficient in relation to the dietary composition; that is, if palatability is low or diet composition is such that it needs more time for chewing, food intake can be limited when food is presented during a relatively short period. Crossover studies cannot be done reliably in rats because a small change in dietary composition will be noticed by the rats, who will consequently reduce food intake immediately (J. Ritskes-Hoitinga et al. 1995). Pair Feeding In case a test substance has a bad taste or gives negative health effects, the consequence usually is that food intake is reduced as compared to the control group. In order to be able to judge the effects of the test substance, this necessitates that food intake in the control group be reduced to the same level as that in the test group. Through measuring the amount of food the test group eats, the control group receives a similar amount of food the next day or the next week; this method is defined as “pair feeding” (feeding in pairs). One can do this on an individual basis or on a group basis. These decisions are to be made by the responsible scientist, judging the specific circumstances of a particular study. Although measuring food intake takes time, it is sometimes a necessary step in order to obtain reliable results. Influence of Feeding Schedules in Pharmacological Studies The effect and pharmacokinetics of pharmacological substances (e.g., oral antibiotics) are largely dependent on the time of administration in relation to the time of feeding. An empty stomach is often
84550.indb 333
11/11/10 6:26:18 PM
334
Handbook of Laboratory Animal Science
required when applying substances per gavage, such as in toxicological or pharmacological studies. This practice is considered necessary in order to avoid mixing the test substance with food, which could dilute or interact with it. How long animals need to be fasted before the “bare” effect of pharmacological substances tested can be judged is an important animal welfare issue (Claassen€1994). A rat will have an empty stomach after 6 hours (Vermeulen et al. 1997). Fasting for longer periods led to increased locomotor and grooming behavior, and 18 h fasting caused a body weight loss of at least 10% (Vermeulen et al. 1997). Fasting for 2, 3, and 4 days led to a body weight loss of 11, 16, and 13%, respectively, in rats with an initial body weight of 75, 112, and 225 g (Chwalibog et al. 1998a). In case an empty intestinal tract is required, fasting up to 22 h is required in the rat (SGV Newsletter 2001). In that case, it is advised—if it does not interfere with the experiment—to give the rats sucrose cubes or a 10% glucose or maltose solution for the time required to empty the gastrointestinal tract. To avoid the intake of bedding, animals should be housed on grid floors during fasting. Depending on the type of study or which part of the intestinal tract is studied, duration of fasting should be adapted accordingly. In nutritional studies, fasting before blood collections should be avoided because one studies the effects of dietary composition rather than the effects of fasting. In some cases, short-term fasting may be necessary; for example, triacylglycerols in high concentrations may interfere with certain (colorimetrical) measurements. Feeding by gavage is considered a stressful event that may influence metabolism. Vachon et al. (1988) compared feeding a similar meal by gavage versus voluntary intake, which gave different results. The voluntary intake of the meal gave results similar to human studies, whereas giving the same meal by gavage did not (Vachon et al. 1988). Feeding by gavage leads to a reduced gastrointestinal passage time. Blood glucose and insulin peak faster after eating a meal voluntarily as compared to giving a similar meal by gavage, even though, in the latter case, the meal arrives in the stomach faster. Gavage bypasses the first part of the digestion in the mouth, which therefore avoids physical processing (chewing) and adding of saliva and enzymes. Influence of Feeding Schedules on Circadian Rhythms Circadian rhythms are biological functions with a certain periodicity. The typical diurnal frequency is one cycle per 24 h (±4 h). Circadian rhythms are generated endogenously in the brain in the circadian oscillator, located in the suprachiasmatic nuclei in the hypothalamus. In order to prevent obesity, rabbits, pigs, and dogs are fed restrictedly. Usually, the restricted amount of food is provided during normal working hours for personnel. In certain nutritional studies (e.g., postprandial studies), a restricted amount of food is given to rats, usually during the daytime as well. When fed ad libitum, nocturnally active animal species like the mouse, rat, hamster, and rabbit consume almost all their food during the hours of darkness. The natural behavior of these species in the wild also indicates that the dark period is used for foraging and eating (M.€RitskesHoitinga and Strubbe 2004). When a restricted amount of food is given during the normal working hours of the personnel, food-restricted animals start eating immediately because they are hungry; consequently, many biochemical and physiological functions become related to this event. This implies that feeding during daylight will lead to changes in natural rhythms (e.g., in nocturnal species like rabbits and rats). Some metabolic parameters (e.g., blood glucose and insulin) that are directly linked to the time of food ingestion will directly shift to the time of food access. Other activity rhythms that do not seem directly related to food intake (e.g., the 24 h rhythm of locomotor activity and core body temperature) will also be changed and influenced by the altered feeding schedule (Jilge and Hudson 2001). Although some activity rhythms are influenced by periodically restricted food access, this can coincide with an unaffected circadian oscillator at the same time; the effect is called “masking.” This implies that when the restricted feeding schedule is stopped again, the
84550.indb 334
11/11/10 6:26:19 PM
Nutrient Requirements, Experimental Design, and Feeding
335
rhythms are dictated solely by the internal oscillator again, which has not been influenced by the feeding schedule. It is also possible that external variables influence the circadian oscillator. This process is defined as “entrainment.” The time needed to reach a “stable” phase again can last up to 50–60€days (Jilge et al. 1987). Thus, by using periodic food access, some functions are rearranged immediately. Simultaneously, entrainment and masking are taking place. The masking of certain physiological functions is probably needed for maintaining vital functions during the time-consuming process of achieving a new homeostatic state for functions implying complete circadian reorganization. Many metabolic functions are brought out of phase when food access is restricted to some hours during the day. This accounts especially for nocturnal animals being fed during the light period. The process of reentrainment can require 50–60 days; during that time, physiological functions like locomotor activity, digestive functions, and urine excretion will be affected (Jilge and Stähle 1993). It is possible that the feeding of a restricted amount of food at a more “natural” time point (during the dark hours) may prevent the phase shifting of metabolic processes. It is outside the scope of this chapter to discuss this issue in depth, but further details and references can be found in J. RitskesHoitinga and Jilge (2001) and M. Ritskes-Hoitinga and Strubbe (2004).
Diet and Welfare Transport and Acclimatization Knowledge of the species is important when animals are transported. Before transport, getting specialist advice for each particular species is needed; for example, (mini)pigs will vomit when being fed just before transport, rabbits may develop stomach rupture when transported with a full stomach, rats and mice will acclimatize faster after transport when food and water have been provided during the transport (Van Ruiven et al. 1996). After transport of rats within one continent, (i.e., without a shift in light–dark rhythm), it was reported that an acclimatization period of at least 3 days was considered sufficient for nutritional studies (Van Ruiven et al. 1998). As a general rule, a 1-week acclimatization period is advised. Whether or not transport stress has an effect on some individuals will usually become clear during the first week after arrival. When animals are transported between continents, a longer acclimatization period of up to 3–4 weeks may be necessary due to the shift in the light–dark schedule (Van Ruiven et al. 1996). However, as mentioned under circadian rhythms, a 50- to 60-day period may be necessary (Jilge and Stahle 1993). Enrichment and Variation versus Need for Standardization From preference testing, it is known that rats prefer to work for food, instead of having it available at all times. For each species, there are certain species-specific, essential, basic needs connected to searching and finding food (e.g., rooting of pigs). If these essential needs are not fulfilled, abnormal behavior like stereotypies can occur. Pigs can develop sham chewing, mice can develop rotating behavior inside the cage, rabbits can make digging movements in the cage, dogs can pace back and forth, etc. (Krohn et al. 1999; Lawrence and Terlouw 1993; Broom and Johnson 2000; Garner 2005). The fact that some animals develop stereotyped behavior may increase the variation in results, as animals performing stereotypies related to movement may have lower body weights than the animals that have not developed stereotypies. The fact that stereotypies develop implies that an essential need of an animal has not been fulfilled (Broom and Johnson 2000), and therefore the focus should be on the prevention that stereotypies will occur (e.g., by providing enrichment).
84550.indb 335
11/11/10 6:26:19 PM
336
Handbook of Laboratory Animal Science
Possible ways of enriching the environment include letting the animals work and search for food. This aspect is well known and used in zoos, but not so much yet in laboratory animal facilities (Wolfensohn and Honess 2005). The hesitation within the laboratory setting is probably due to the fear that this will compromise standardization. However, by frustrating the animals’ basic needs, one is also compromising animal welfare and standardization. Therefore, the animals should be given the benefit of the doubt, and fulfilling species-specific needs related to foraging and feeding processes should be introduced whenever possible. Knowledge of the natural feeding time and behavior are important factors to consider. The time of day at which a restricted amount of food is given can be an important tool for improving welfare. For example, Krohn et al. (1999) and M. Ritskes-Hoitinga and Strubbe (2004) reported a significantly reduced frequency of stereotypies by feeding rabbits a restricted amount of food just before the dark period, instead of providing ad libitum or restricted amounts of food in the morning. Giving food rewards is an important tool for teaching and training animals. Monkeys have been trained to present their arm voluntarily for taking blood samples by giving them a banana as reward. These social activities with human beings are also important enrichment tools and will improve trust toward people handling them. That way, animals will be less stressed during experiments and the pain threshold goes up. Which food rewards are chosen and in what amounts need careful consideration to avoid interference with the experimental results and health of the animal. Certain dietary schedules require individual housing. Because individual housing compromises the well-being of social species, alternative ways of feeding need to be considered whenever possible. For example, the animals can be individually fed for a certain period each day and then socially housed for the remaining part of the 24 h period. A current challenge of modern laboratory animal science is to increase the efforts on enrichment related to feeding for all species because this has seldom been used. Further research will prove whether nutritional enrichment may compromise standardization and, if so, how and when. However, it may also prove that by enriching the feeding process, animal welfare increases, thereby reducing variation and improving standardization. Behavioral needs of individual animals, as well as social interaction in the group, are important factors codetermining experimental outcome. In group-housed S3 rats, the individual food intake pattern clearly varied between the cage where a clearly dominant animal was present versus the cage where no clear hierarchy was present in the group (M. Ritskes-Hoitinga and Strubbe 2004). Nutritional studies in groups of dogs clearly demonstrated that socialization and training improved group harmony, animal welfare, and the validity of experimental results (J. Ritskes-Hoitinga 2006). This illustrates that efforts to improve animal welfare that respect the essential needs of individuals and species can simultaneously lead to better experimental results. Conclusion This chapter has offered an overview of the importance of nutrition and feeding of laboratory animals. The composition of diets needs to fulfill species-specific needs in order to obtain healthy animals and to avoid undesirable interference with experimental results. The composition of the diet, the amount of diet fed, and the time of feeding are important experimental factors to consider because they influence the experimental results and well-being of the animals. There is still a great potential to be explored for stimulating welfare of laboratory animals through dietary enrichment programs such as the types of diets used, working and searching for food, and adapting the amount and timing of eating to the specific species.
84550.indb 336
11/11/10 6:26:19 PM
Nutrient Requirements, Experimental Design, and Feeding
337
References Anderson, J. W. 1985. Physiological and metabolic effects of dietary fiber. Federal Proceedings 44:2902–2906. Beynen, A. C. and M. E. Coates. 2001. Nutrition and experimental results. In Principles of laboratory animal science, ed. L. F. M. Van Zutphen, V. Baumans, and A. C. Beynen. Amsterdam: Elsevier Scientific Publishers. Bird, R., and R. Bruce. 1986. Effect of dietary fat levels on the susceptibility of colonic cells to nuclear-damaging agents. Nutrition and Cancer 8:93–100. Blaxter, K. 1989. Energy metabolism in farm animals, chap. 8. Cambridge, England: Cambridge University Press. Bluher, M., B. B. Kahn, and C. R. Kahn. 2003. Extended longevity in mice lacking the insulin receptor in adipose tissue. Science 299:572–574. Bollen, P. J. A., A. G. Lemmens, A. C. Beynen, O. Meyer, and J. Ritskes-Hoitinga. 2006. Bone composition in male and female Göttingen minipigs fed variously restrictedly and near ad libitum. Scandinavian Journal of Laboratory Animal Science 33 (3): 1–10. British Association of Research Quality Assurance (BARQA). 1992. Guidelines for the manufacture and supply of GLP animal diets. Broom, D. M., and K. G. Johnson. 2000. Stress and animal welfare. Dordrecht: Kluwer Academic Publishers. Bull, A., J. Bronstein, and N. Nigro. 1989. The essential fatty acid requirement for azoxymethane-induced intestinal carcinogenesis in rats. Lipids 24:340–346. Burr, G. O. and Burr, M. M. 1929. A new deficiency disease produced by the rigid exclusion of fat from the diet. Journal of Biological Chemistry 82:345–367. ———. 1930. On the nature and role of the fatty acids essential in nutrition. Journal of Biological Chemistry 86:587–621. Cañas, R., J. J. Romero, and R. L. Baldwin. 1982. Maintenance energy requirements during lactation in rats. Journal of Nutrition 112:1876–1880. Carney, E. W., C. L. Zablotny, M. S. Marty, J. W. Crissman, P. Anderson, M. Woolhiser, and M. Holsapple. 2004. The effects of feed restriction during in utero and postnatal development in rats. Toxicological Sciences 82:237–249. Christensen, K. 1985. Determination of linoleic acid requirements in slaughter pigs. National Institute of Animal Science Report 577, Copenhagen: Landhusholdningsselskabet. Chwalibog, A. 1989. [Nutrition of laboratory animals.] DSR Forlag, the Royal Veterinary and Agricultural University, Copenhagen. ———. 1991. Energetics of animal production. Acta Agriculturæ Scandinavica 41:147–160. ———. 2000. [Animal nutrition], chap. 4. DSR Forlag, the Royal Veterinary and Agricultural University Copenhagen. Chwalibog, A., K. Jakobsen, S. Henckel, and G. Thorbek. 1992. Oxidation and fat retention from carbohydrate, protein, and fat in growing pigs. Journal of Animal Physiology and Animal Nutrition 68:123–125. Chwalibog, A., K. Jakobsen, A-H. Tauson, and G. Thorbek. 1998a. Heat production and substrate oxidation in rats fed at maintenance level and during fasting. Comparative Biochemistry and Physiology 121:423–429. Chwalibog, A., A-H. Tauson, R. Fink, and G. Thorbek. 1998b. Oxidation of substrates and lipogenesis in pigs (Sus scrofa), mink (Mustela vison), and rats (Rattus norvegicus). Thermochimica Acta 309: 49–56. Chwalibog, A., and G. Thorbek. 2000. Estimation of net nutrient oxidation and lipogenesis in growing pigs. Archives of Animal Nutrition 53:253–272. Claassen, V. 1994. Neglected factors in pharmacology and neuroscience research: Biopharmaceutics, animal characteristics, maintenance, testing conditions, 1st ed. Amsterdam: Elsevier. Clarke, H. E., M. W. Coates, J. K. Eva, D. J. Ford, C. K. Milner, P. N. O’Donoghue, P. O. Scott, and R. J. Ward. 1977. Dietary standards for laboratory animals: Report of the Laboratory Animal Center Diets Advisory Committee. Lab Anim 11:1–28. Coates, M. E. 1976. The nutrition of laboratory animals. In The UFAW handbook on the care and management of laboratory animals, ed. C. W. Hume, chap. 3. London: UFAW, Churchill Livingstone. Crevel, R. W. R., J. V. Friend, B. F. Goodwin, and W. E. Parish. 1992. High-fat diets and the immune response of C57 B1 mice. British Journal of Nutrition 67:17–26. Cummings, J. H. 1978. Nutritional implications of dietary fiber. American Journal of Clinical Nutrition 31:S21–S29.
84550.indb 337
11/11/10 6:26:19 PM
338
Handbook of Laboratory Animal Science
Dean, J., and D. G. Edwards. 1985. The nutritional value of rat diets of differing energy and protein levels when subjected to physical processing. Laboratory Animals 19:311–319. Descholdt-Lanckman, M., P. Robberecht, J. Camus, C. Baya, and J. Christophe. 1974. Hormonal and dietary adaptation of rat pancreatic hydrolases before and after weaning. American Journal of Physiology 226:39–44. DeWille, J. W., K. Waddell, C. Steinmeyer, and S. T. Farmer. 1993. Dietary fat promotes mammary tumorigenesis in MMTV/v-Ha-ras transgenic mice. Cancer Letters 69:59–66. Dirx M. J., M. P. Zeegers, P. C. Dagnelie, and P. A. van den Brandt. 2003. Energy restriction and the risk of spontaneous mammary tumors in mice: A meta-analysis. International Journal of Cancer 106:766–770. Edwards, D. G., and J. Dean. 1985. The responses of rats to various combinations of energy and protein. II. Diets made from natural ingredients. Lab Anim 19:336–343. Edwards, D. G., P. D. Porter, and J. Dean. 1985. The responses of rats to various combinations of energy and protein. I. Diets made from purified ingredients. Lab Anim 19:328–335. Eggum, B. O. 1973. A study of certain factors influencing protein utilization in rats and pigs. Thesis. National Institute of Animal Science Report 406. ———. 1989. Biochemical and methodological principles. In Protein metabolism in farm animals, ed. H. D. Bock, B. O. Eggum, A. G. Low, O. Simon, and T. Zebrowska, chap. 1. Berlin: Deutscher Landwirtschaftsverlag, Oxford Scientific Publications. Eggum, B. O., Andersen, J. O., and Rotenberg, S. 1982. The effect of dietary fiber level and microbial activity in the digestive tract on fat metabolism in rats and pigs. Acta Agriculturæ Scandinavica 32:145–150. Eggum, B. O., K. E. Bach Knudsen, and I. Jacobsen. 1981. The effect of amino acid imbalance on nitrogen retention (biological value) in rats. British Journal of Nutrition 45:175–181. Eggum, B. O., and R. M. Beames. 1986. Use of laboratory animals as models for studies on nutrition of domestic animals. In Laboratory animals, ed. E. J. Ruitenberg and P. W. J. Peters, chap. 9. Amsterdam: Elsevier Science Publishers B.V. Eggum, B. O., and Chwalibog, A. 1983. A study on requirement for maintenance and growth in rats with normal or reduced gut flora. Zeitschrift für Tierphysiologie Tierernährung und Futtermittelkunde 49:104–114. Environmental Protection Agency. 1979. Proposed health effects test standards for toxic substances control act test rules. Good laboratory standards for health effects. Federal Register 44. Fisher, M. S., J. C. Claudio, M. Locniskar, M. A. Belury, R. E. Maldve, M. L. Lee, J. Leyton, T. J. Slaga, and D. Bechtel. 1992. The effect of dietary fat on the rapid development of mammary tumors induced by 7,12-dimethylbenz(a)anthracene in SENCAR mice. Cancer Research 52:662–666. Forbes, J. M. 1995. Voluntary food intake and diet selection in farm animals, chap. 11. Wallingford, England: CAB International. ———. 2007. The voluntary food intake and diet selection of farm animals. Wallingford, England: CAB International. Fuller, M. F., and T. C. Wang. 1987. Amino acid requirements of the growing pig. In Manipulation of pig production, ed. J. L. Barnett, 97–111. Werribee: Australasian Pig Science Association. Garner, J. P. 2005. Stereotypies and other abnormal repetitive behaviors: Potential impact on validity, reliability, and replicability of scientific outcomes. ILAR Journal 46 (2): 106–117. Glick, Z., A. Bray, and R. J. Teague. 1984. Effect of prandial glucose on brown fat thermogenesis in rats: Possible implications for dietary obesity. Journal of Nutrition 114:1934–1939. GV-Solas publication no. 9. 1980. Definition of terms and designations in laboratory animal nutrition (part 1): Use of feedstuffs and bedding materials for nonclinical laboratory studies (part 2). Harper, A. E., N. J. Benevenga, and R. M. Wohlhueter. 1970. Effects of ingestion of disproportionate amounts of amino acids. Physiological Reviews 50:428–558. Hart, R. W., D. A. Neumann, and R. T. Robertson. 1995. Dietary restriction: Implications for the design and interpretation of toxicity and carcinogenicity studies. Washington, D.C.: ILSI Press. Haseman, J. K., J. Huff, and G. A. Boorman. 1984. Use of historical control data in carcinogenicity studies in rodents. Toxicologic Pathology 12:126–135. Hawrylewicz, E. J., H. H. Huang, and W. H. Blair. 1991. Dietary soybean isolate and methionine supplementation affect mammary tumor progression in rats. Journal of Nutrition 121:1693–1698. Henning, S. J. 1981. Postnatal development: Coordination of feeding, digestion, and metabolism, American Journal of Physiology 241:G199–G214.
84550.indb 338
11/11/10 6:26:19 PM
Nutrient Requirements, Experimental Design, and Feeding
339
———. 1987. Functional development of the gastrointestinal tract. In Physiology of the gastrointestinal tract, ed. R. Johnson, chap. 9. New York: Raven Press. Herberg, L. 1991. Interrelationships between obesity and diabetes. Proceedings of the Nutrition Society 50:605–617. Holman, R. T. 1968. Essential fatty acid deficiency. In Progress in the chemistry of fats and other lipids, vol. 9, 619. Elmsford, NY: Pergamon Press. Hubert, M. F., P. Laroque, J-P. Gillet, and K. Keenan. 2000. The effects of diet, ad libitum feeding, and moderate and severe dietary restriction on body weight, survival, clinical pathology parameters, and cause of death in control Sprague–Dawley rats. Toxicological Sciences 58:195–207. Imai, K., M. Ohnaka, M. Ondani, and Y. Niiyama. 1986. Maintenance energy requirement in pregnant rats and net energetic efficiency for fetal growth during late pregnancy. Journal of Nutritional Science and Vitaminology 32:527–536. Innis, S. M. 1991. Essential fatty acids in growth and development. Progress in Lipid Research 30:39–103. Ip, C. 1987. Fat and essential fatty acid in mammary carcinogenesis. American Journal of Clinical Nutrition 45:218–224. Ip, C., C. A. Carter, and M. M. Ip. 1985. Requirement of essential fatty acid for mammary tumorigenesis in the rat. Cancer Research 45:155–159. Jefery, P., M. Burrows, and A. Bye. 1987. Does the rat have an empty stomach after an overnight fast? Lab Anim 21:330–334. Jilge, B., H. Hörnicke, and H. Stähle. 1987. Circadian rhythms of rabbits during restrictive feeding. American Journal of Physiology 253:R46–R54. Jilge, B., and R. Hudson. 2001. Diversity and development of circadian rhythms in the European rabbit: A review. Chronobiology International 18:1–26. Jilge, B., and H. Stähle. 1993. Restricted food access and light–dark: Impact of conflicting zeitgebers upon circadian rhythms of the rabbit. American Journal of Physiology 264:R708–R715. Keenan, K. P., G. C. Ballam, R. Dixit, K. Soper, P. Laroque, B. A. Mattson, S. P. Adams, and J. B. Coleman. 1997. The effects of diet, overfeeding and moderate dietary restriction on Sprague–Dawley rat survival, disease and toxicology. Journal of Nutrition 127:851–856. Keenan, K. P., G. C. Ballam, K. A. Soper, P. Laroque, J. B. Coleman, and R. Dixit. 1999. Diet, caloric restriction, and the rodent bioassay. Toxicological Sciences 52 (S2): 24–34. Keenan, K. P., P. Laroque, K. Soper, R. E. Morrissey, and R. Dixit. 1996. The effects of overfeeding and moderate dietary restriction on Sprague–Dawley rat survival, pathology, carcinogenicity, and the toxicity of pharmaceutical agents. Experimental and Toxicological Pathology 48:139–144. Keenan, K. P., P. E. Smith, P. Hertzog, K. Soper, G. C. Ballam, and R. L. Clark. 1994. The effects of overfeeding and dietary restriction on Sprague–Dawley rat survival and early pathology biomarkers of aging. Toxicologic Pathology 22:300–315. Keenan, K. P., K. Soper, P. R. Hertzog, L. A. Gumprecht, P. F. Smith, B. A. Mattson, G. C. Ballam, and R. L. Clark. 1995a. Diet, overfeeding and moderate dietary restriction in control Sprague–Dawley rats. II. Effects on age-related proliferative and degenerative lesions. Toxicologic Pathology 23:287–302. Keenan, K. P., K. Soper, P. F. Smith, G. C. Ballam, and R. L. Clark. 1995b. Diet, overfeeding and moderate dietary restriction in control Sprague–Dawley rats. I. Effects on spontaneous neoplasmas. Toxicologic Pathology 23:269–286. Kleiber, M. 1961. The fire of life, chap. 18. New York: John Wiley & Sons. Klein, M., and L. Hoffmann. 1989. Bioenergetics of protein retention. In Protein metabolism in farm animals, ed. H. D. Bock, B. O. Eggum, A. G. Low, O. Simon, and T. Zebrowska, chap. 11. Berlin: Oxford Scientific Publications, Deutscher Landwirtschaftsverlag. Koopman, J. P., P. M. Scholten, and A. C. Beynen. 1989a. Hardness of diet pellets and growth of preweaned mice: Separation of direct effects on the young and indirect effects mediated by the lactating females. Zeitschrift für Versuchstierkunde 32:257–260. Koopman, J. P., P. M. Scholten, P. C. Roeleveld, Y. W. M. Velthuizen, and A. C. Beynen. 1989b. Hardness of diet pellets and its influence on growth of preweaned and weaned mice. Zeitschrift für Versuchstierkunde 32:71–74. Krohn, T. C., J. Ritskes-Hoitinga, and P. Svendsen. 1999. The effects of feeding and housing on the behavior of the laboratory rabbit. Lab Anim 33 (2): 101–107.
84550.indb 339
11/11/10 6:26:19 PM
340
Handbook of Laboratory Animal Science
Kuratko, C., and B. Pence. 1992. Rat colonic antioxidant status: Interaction of dietary fats with 1,2 dimethylhydrazine challenge. Journal of Nutrition 122:278–282. Lang, C. M., and G. T. Harrell. 2000. Laboratory animal science in the future. Scandinavian Journal of Laboratory Animal Science 27:166–172. Lawrence, A. B., and E. M. Terlouw. 1993. A review of behavioral factors involved in the development and continued performance of stereotypic behaviors in pigs. Journal of Animal Science 71:2815–2825. Leakey, J. E., J. E. Seng, and W. T. Allaben. 2003a. Body weight considerations in the B6C3F1 mouse and the use of dietary control to standardize background tumor incidence in chronic bioassays. Toxicology and Applied Pharmacology 193:237–265. Leakey, J. E., J. E. Seng, J. R. Latendresse, N. Hussain, L. J. Allen, and W. T. Allaben. 2003b. Dietary controlled carcinogenicity study of chloral hydrate in male B6C3F1 mice. Toxicology and Applied Pharmacology 193:266–280. Madsen, C. 1989. Squamous-cell carcinoma and oral, pharyngeal and nasal lesions caused by foreign bodies in feed. Cases from a long-term study in rats. Lab Anim 23:241–247. Masoro, E. J. 2005. Overview of caloric restriction and ageing. Mechanisms of Ageing and Development 126 (9): 913–922. McDonald, P., R. A. Edwards, and J. F. D. Greenhalgh. 1981. Animal nutrition, chap. 16. Chicago: Longman. McDonald, P., R. A. Edwards, J. F. D. Greenhalgh, and C. A. Morgan. 2002. Animal nutrition. London: Addison Wesley Longman. Melander, A. 1978. Influence of food on the bioavailability of drugs. Clinical Pharmacokinetics 3:337–351. Molon-Noblot, S., M. F. Hubert, C. M. Hoe, K. Keenan, and P. Laroque. 2005. The effects of ad libitum feeding and marked dietary restriction on spontaneous skeletal muscle pathology in Sprague–Dawley rats. Toxicologic Pathology 33:600–608. National Research Council. 1987. Vitamin tolerance of animals. Washington, D.C.: National Academy Press. ———. 1995. Nutrient requirements of laboratory animals, 4th ed. Washington, D.C.: National Academy of Sciences. National Research Council Collection: Nutrient requirements of sheep 1985; of dogs 1985; of beef cattle 1984; of mink and foxes, 1982; of laboratory animals (rat, mouse, guinea pig, hamster, gerbil, vole), 1995; of poultry 1994; of fish 1993; of horses 1989; of dairy cattle 1989; of swine 1998; of cats 1986; of goats 1981; of nonhuman primates 1978; of rabbits 1977. Washington, D.C.: National Academy Press. Nygaard Jensen, M., and M. Ritskes-Hoitinga. 2007. How isoflavone levels in common rodent diets can interfere with the value of animal models and with experimental results. Laboratory Animals 41:1–18. Owens, F. N., and J. E. Pettigrew. 1989. Subdividing amino acid requirements into portions for maintenance and growth. In Absorption and utilization of amino acids, ed. M. Friedman, chap. 2. Boca Raton, FL: CRC Press, Inc. Patel, N. V., M. N. Gordon, K. E. Connor, R. A. Good, R. W. Engelman, J. Mason, D. G. Morgan, T. E. Morgan, and C. E. Finch. 2005. Caloric restriction attenuates Abeta-deposition in Alzheimer transgenic models. Neurobiology of Aging 26:995–1000. Prochazka, P., P. Hahn, O. Koldovsky, M. Nohynek, and J. Rokos. 1964. The activity of alpha-amylase in homogenates of the pancreas of rats during early postnatal development. Physiologia Bohemoslovaca 13:288–291. Pullar, J. D., and A. J. F. Webster. 1974. Heat loss and energy retention during growth in congenitally obese and lean rats. British Journal of Nutrition 31:377–392. Raczynski, G., B. O. Eggum, and A. Chwalibog. 1982. The effect of dietary composition on transit time in rats. Zeitschrift für Tierphysiologie Tierernährung und Futtermittelkunde 47:160–167. Rafecas, I., M. Esteve, X. Remesar, and M. Alemany. 1991. Plasma amino acids of lean and obese Zucker rats subjected to a cafeteria diet after weaning. Biochemistry International 25:797–806. Reeves, P. G., F. H. Nielsen, and G. C. Fahey, Jr. 1993 AIN-93 Purified diets for laboratory rodents: Final report of the American Institute of Nutrition Ad Hoc Writing Committee on the Reformulation of the AIN-76A Rodent Diet. Journal of Nutrition 123:1939–1951. Ritskes-Hoitinga, J. 1992. Nephrocalcinosis in the laboratory rat. Thesis, Utrecht University, the Netherlands. ———. 2006. Heeft een rat boeddhanatuur? Radboud University Nijmegen. http://www.umcn.nl/userfiles/ other/Heeft_een_rat_boeddhanatuur_.pdf (in Dutch).
84550.indb 340
11/11/10 6:26:19 PM
Nutrient Requirements, Experimental Design, and Feeding
341
Ritskes-Hoitinga, J., and P. Bollen. 1997. Nutrition of (Göttingen) minipigs: Facts, assumptions and mysteries. Pharmacology and Toxicology 80 (SII): 5–9. ———. 1998. The formulation of a test diet in establishing the nutrient requirements and optimum feeding schedules for minipigs. Scandinavian Journal of Laboratory Animal Science 25 (S1): 27–30. ———. 1999. The Göttingen minipig, a refined animal model? Der Tierschutzbeauftragte 2:87–90. Ritskes-Hoitinga, J. and B. Jilge. 2001. FELASA—Quick reference paper on laboratory animal feeding and nutrition (http://www.felasa.org). Ritskes-Hoitinga, J., J. N. J. J. Mathot, L. H. J. C. Danse, and A. C. Beynen. 1991. Commercial rodent diets and nephrocalcinosis in weanling female rats. Lab Anim 25:126–132. Ritskes-Hoitinga, J., M. Meijers, G. W. Meijer, and J. A. Weststrate. 1996. The influence of dietary linoleic acid on mammary tumor development in various animal models. Scandinavian Journal of Laboratory Animal Science 23 (1): 463–468. Ritskes-Hoitinga, J., K. H. van het Hof, W. J. Kloots, E. A. M. de Deckere, J. M. van Amelsvoort, and J. A. Weststrate. 1995. Rat as a model to study postprandial effects in man. Alternative methods in toxicology and the life sciences series, vol. 11. In Proceedings of the World Congress on Alternatives and Animal Use in the Life Sciences: Education, Research, Testing, ed. A. M. Goldberg and L. F. M. van Zutphen, 403–410. Ritskes-Hoitinga, J., P. M. Verschuren, G. W. Meijer, A. Wiersma, A. J. van de Kooij, W. G. Timmer, C. G. Blonk, and J. A. Weststrate. 1998. The association of increasing dietary concentrations of fish oil with hepatotoxic effects and a higher degree of aorta atherosclerosis in the ad lib.-fed rabbit. Food and Chemical Toxicology 36:663–672. Ritskes-Hoitinga, M., L. Bjoerndal Gravesen, and I. M. Jegstrup. 2006. Refinement benefits animal welfare and quality of science. NC3Rs #6 (http://www.nc3rs.org.uk/news.asp?id=212). Ritskes-Hoitinga, M., and J. H. Strubbe. 2004. Nutrition and animal welfare. In The welfare of laboratory animals, ed. E. Kaliste, 51–80. Dordrecht: Kluwer Academic Publishers. Robinson, M., D. Hart, and G. H. Pigott. 1991. The effects of diet on the incidence of periodontitis in rats. Lab Anim 25:247–253. Roe, F. C. J. 1994. Historical histopathological control data for laboratory rodents: Valuable treasure or worthless trash? Lab Anim 28:148–154. Rogers, Q. R., and P. M. B. Leung. 1973. The influence of amino acids on the neuroregulation of food intake. Federal Proceedings 32:1709–1719. Ross, M. H., and G. Brass. 1965. Influence of protein under- and overnutrition on spontaneous tumor prevalence in the rat. Journal of Nutrition 103:944–963. Rotenberg, S., and J. O. Andersen. 1982. The effect of antibiotica on some lipid parameters in rats receiving cornstarch, potato flour or pectin in the diet. Acta Agriculturæ Scandinavica 32:321–340. Rotenberg, S., B. O. Eggum, M. Hegedüs, and I. Jacobsen. 1982. The effect of pectin and microbial activity in the digestive tract on fecal excretion of amino acids, fatty acids, thiamin, riboflavin and niacin in young rats. Acta Agriculturæ Scandinavica 32:309–319. Rubino, A., F. Zimbalatti, and S. Auricchio. 1964. Intestinal disaccharidase activities in adult and suckling rats. Biochimica et Biophysica Acta 92:305–311. Savenije, B., J. Strubbe, and M. Ritskes-Hoitinga. 2010. Nutrition, feeding and animal welfare. In The UFAW handbook on the care and management of laboratory and other research animals, 8th ed., ed. R. Hubrecht and J. Kirkwood, 183–193. Oxford, England: Wiley–Blackwell. Schiemann, R., K. Nehring, L. Hoffmann, W. Jentsch, and A. Chudy. 1971. Energetische Futterbewertung und Energinormen. Berlin: VEB Deutscher. Landwirtschaftsverlag. SGV Newsletter, no. 24 (Spring 2001): http://www.sgv.org/Newsletter/news-24.htm Shin, I-S. 1990. Subdividing amino acid requirements for maintenance from requirements for growth. Thesis, Oklahoma State University. Solleveld, H. A., P. McAnulty, J. Ford, P. W. J. Peters, and J. Tesh. 1986. Breeding, housing and care for laboratory animals. In Laboratory animals, ed. E. J. Ruitenberg and P. W. J. Peters, chap. 1. Amsterdam: Elsevier Science Publishers B.V. Speijers, G. J. A. 1987. Voedingsvezel en haarballen. NVP Symposium Proceedings, “Voeding en kwaliteit van proef en dier.” Suzuki, T., S. Tana-Ami, H. Fujiwara, and T. Ishibashi. 1991. Effect of the energy density of nonpurified diets on reproduction, obesity, alopecia and aging in mice. Experimental Animals 40:499–510.
84550.indb 341
11/11/10 6:26:19 PM
342
Handbook of Laboratory Animal Science
Szelényi-Galàntai, M., I. Jacobsen, and B. O. Eggum. 1981. The influence of dietary energy density on protein utilization in rats. Acta Agriculturæ Scandinavica 31:204–206. Tauson, A-H., A. Chwalibog, K. Jakobsen, and G. Thorbek. 1998. Pattern of protein retention in growing boars of different breeds, and estimation of maximum protein retention. Archives of Animal Nutrition 51:253–272. Thigpen, J. E., K. D. R. Setchell, K. B. Alhmark, J. Locklear, T. Spahr, G. F. Caviness, M. F. Goelz, J. K. Haseman, R. R. Newbold, and D. B. Forsythe. 1999. Phytoestrogen content of purified, open- and closedformula laboratory animal diets. Laboratory Animal Science 49:530–536. Thorbek, G., and A. Chwalibog. 1981. [Growth, nitrogen and energy metabolism in growing rabbits measured at different feed combinations.] National Institute of Animal Science Report 510:1–21. Thorbek, G., A. Chwalibog, B. O. Eggum, and K. Christensen. 1982. Studies on growth, nitrogen, and energy metabolism in rats. Archives of Tierernährung 32:827–840. Thorbek, G., A. Chwalibog, and S. Henckel. 1983. Energetics of growth in pigs from 20 to 120 kg live weight. Zeitschrift für Tierphysiologie Tierernährung und Futtermittelkunde 49:238–249. Vachon, C., J. D. Jones, A. Nadeau, and L. Savoie. 1988. A rat model to study postprandial glucose and insulin responses to dietary fibers. Nutrition Reports International 37:1339–1348. Van Ruiven, R., G. W. Meijer, L. F. M. van Zutphen, and J. Ritskes-Hoitinga. 1996. Adaptation period of laboratory animals after transport: A review. Scandinavian Journal of Laboratory Animal Science 23:185–190. Van Ruiven, R., G. W. Meijer, A. Wiersma, V. Baumans, L. F. M. van Zutphen, and J. Ritskes-Hoitinga. 1998. The influence of transportation stress on selected nutritional parameters to establish the necessary minimum period for adaptation in rat feeding studies. Lab Anim 32:446–456. Vermeulen, J. K., A. de Vries, F. Schlingmann, and R. Remie. 1997. Food deprivation: Common sense or nonsense? Animal Technology 48:45–54. Ward, J. D. 2008. A manual for laboratory animal management. Singapore: World Scientific Publishing Co. Wolfensohn, S., and P. Honess. 2005. Handbook of primate husbandry and welfare. Oxford, England: Blackwell Publishing. Wyss-Spillmann, S. K., F. R. Homberger, G. Lott-Stolz, R. Jörg, and P. E. Thomann. 1997. Alopecia in rats due to nutritional deficiency. In Proceedings of the Sixth FELASA Symposium, June 19–21, 1996, ed. P. N. O’Donoghue, Basel, Switzerland. Yoshida, A., A. E. Harper, and C. A. Elvehjem. 1957. Effects of protein per calorie ratio and dietary level of fat on calorie and protein utilization. Journal of Nutrition 63:555–570. Zhao, X., H. Jørgensen, and B. O. Eggum. 1995. The influence of dietary fiber on body composition, visceral organ weight, digestibility and energy balance in rats housed in different thermal environments. British Journal of Nutrition 73:687–699. Zhu, Z., A. D. Haegele, and H. J. Thompson. 1997. Effect of caloric restriction on premalignant and malignant stages of mammary carcinogenesis. Carcinogenesis 18:1007–1012.
84550.indb 342
11/11/10 6:26:20 PM
Chapter 12
Impact of the Biotic and Abiotic Environment on Animal Experiments
Nancy A. Johnston and Timo Nevalainen Contents Introduction..................................................................................................................................... 343 Biotic Factors..................................................................................................................................344 Abiotic Factors................................................................................................................................ 347 Bedding...................................................................................................................................... 347 Cage Complexity........................................................................................................................ 351 Cage Material............................................................................................................................. 353 Relative Humidity...................................................................................................................... 354 Light........................................................................................................................................... 354 Sound......................................................................................................................................... 355 Temperature............................................................................................................................... 357 Summary......................................................................................................................................... 358 References....................................................................................................................................... 358 Introduction Laboratory animal science is concerned with standardization of all the factors that may have an impact on animals and, consequently, on experimental results, as nicely presented half a century ago by Biggers, McLaren, and Michie (1958). Unnecessarily large variation is the enemy of scientists; hence, all approaches and methods to control variation should be utilized. This is a key element in animal experiments with the aim to operate with best practice and with relatively low numbers of animals. Recently, the original definition of alternative methods by Russell and Burch (1959) has been reiterated and updated in the Declaration of Bologna. A closer look at refinement and reduction alternative methods shows that we are dealing with the same topics. In essence, the refinement alternative is any activity for improvement of animal welfare through housing or procedure practices, and the reduction alternative is any activity toward lowest possible number of animals. In practical terms, this means that every scientist can and should apply these alternative methods. How should one define the optimal number of laboratory animals in terms of refinement and reduction? Refinement and reduction are not independent of each other; hence, an understanding 343
84550.indb 343
11/11/10 6:26:20 PM
344
Handbook of Laboratory Animal Science
Refined welfare Operation area Fewer animals
More animals
Poor welfare Figure 12.1â•…Schematic presentation of refinement (y-axis) and reduction (x-axis) interplay illustrating the preferred operation area (shaded) for animal studies and the preferred direction (arrows) to which the “loop” should be drawn. The refinement axis is straightforward, while the reduction axis has a window for appropriate number of animals, below which the statistical power is too low to reach significance and above which animals are “wasted.”
of their interplay as depicted in Figure€12.1 is crucial. On the reduction axis, there is a window of appropriate numbers of animals, below and above which the experiment becomes less meaningful and potentially even unethical. Experiments that utilize too few animals have limited statistical power and are likely to fail to be able to draw conclusions, and experiments that involve too many subjects are potentially guilty of unnecessarily large group size. The refinement axis is more straightforward: the higher the refinement value is, the better. These two axes establish the operation area in animal studies, and arrows demonstrate the preferred direction. Environmental biotic and abiotic factors can interfere with a study and study results in essentially two ways. If such a factor alters the mean of determination, this may be the lesser of the evils because it should alter values in both the experimental and control groups. A factor that results in a change in variation is far more troublesome because this is likely to affect the appropriate number of animals required for the experiment, resulting in some of the consequences mentioned before. Unfortunately, relatively few data are available to assess the consequences of environmental biotic and abiotic factors on refinement and reduction. There are lists of factors that have an impact on animals, and they should be considered thoroughly while an animal study is designed and planned. These lists provide a summary of variable environmental parameters, which should be carefully described in all scientific publications (Ellery 1985; Hooijmans et al. 2010). This is fundamentally important to facilitate the reader’s understanding of the findings and interpretation (and potential replication) of the study. Biotic Factors One of the major sources of variation in laboratory animal science is contamination or infection of the research animals with microbial agents. Controlling or eliminating the agents of infection contributes to the standardization of experiments using animals. The renewed call for the establishment of alternative methods in laboratory animal science necessitates the appropriate control of these biological agents. The refinement of experimental techniques requires the elimination of the unknown variables that are associated with biotic agents. Reduction of the number of animals is only possible if unexpected deaths of animals, clinical illnesses, and physiological alterations are minimized or eliminated.
84550.indb 344
11/11/10 6:26:21 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
345
Rodent pathogens remain prevalent and widespread; recent surveys in Asia, North America, and Europe (Schoondermark-van de Ven et al. 2006; Clifford and Watson 2008; Liang et al. 2009; Mähler and KÖhl 2009; Pritchett-Corning et al. 2009) document the continued presence of many viral, bacterial, parasitic, and fungal pathogens. Newly identified pathogens are found in a great percentage of samples, indicating that the desire for standardization of microbial flora remains an ongoing quest (Besselsen et al. 2008). Whether or not these infectious agents produce clinical disease, laboratory animals’ exposure to and infection from such agents can influence the outcome of experiments. The effects of these agents may lead to false conclusions or misinterpretation of data. The introduction of this type of biological variability may also make it difficult to perform reproducible studies (Baker 1998; GV-SOLAS 1999; Nicklas et al. 2002; Baker 2003). The benefits of eliminating clinical disease seem obvious, and the benefits of eliminating subclinical infections through creating a biologically controlled environment are becoming more widely recognized. Certainly, the elimination of rodent pathogens is the goal; at the least, defining the microbial flora present would help standardize or harmonize research findings throughout the scientific community (Yamamoto et al. 2001; Nicklas 2008). The environment of laboratory animals may be classified based on the presence of biotic factors: “conventional,” specific pathogen free (SPF), or sterile (axenic). Many research animals are housed in conventional facilities, where there are opportunities for animals to come into contact with many known and unknown microbes (e.g., bacteria, fungi, viruses, and parasites) with little or no control over exposure. Wild rodent populations continue to carry significant pathogen loads, posing a very real threat of exposure to laboratory populations (Becker et al. 2007; Easterbrook et al. 2008). Such interactions with microbes may produce myriad host responses, including clinical or subclinical disease, immune responses, behavioral or physiological changes, and decreases in reproductive capabilities, among others. These microbes create variability in the experimental environment and add a layer of uncertainty to any data collected. Conventional facilities suffering from widespread disease and infection are becoming increasingly rare as researchers understand how these biotic factors may adversely influence research parameters. In SPF facilities, a select number of bacteria, viruses, and/or parasites are excluded from the facility. The animals are not sterile, and they are not free from all infections or exposure to microbes, but animals typically remain clinically healthy. The selection of specific agents excluded is determined by the individual facility, usually by selecting the specific microbes that cause the greatest illness or harm to the animals, may result in human illness, or that cause physiological or metabolic changes severe enough to alter research results. In this environment, animals are tested for infection or exposure to the selected viruses, bacteria, and/or parasites, but carry an unknown microbial flora that may differ substantially across facilities or even among individual animals in the same facility (O’Rourke et al. 1988). For example, unintentional infection with Helicobacter species may cause tremendous variations in response to experimental variables in SPF mouse models (Chichlowski and Hale 2009). The animals housed in the most controlled environments are classified as axenic or gnotobiotic. Axenic animals are kept totally free from association with microbes, and gnotobiotic animals have only a few known, nonpathogenic microorganisms as normal flora (Frost and Hamm 1990). These animals, although more uniform in health status than conventionally reared animals, may still display different nutritional needs or demonstrate different responses to drug metabolism than animals with normal flora (Rahija 2007). Interfacility collaborations among investigators have increased the risk of contamination to animals, especially mice, transported between institutions. As genetically engineered animals are becoming increasingly used and shared, the spread of pathogens may also be shared across institutions. Sources of animals have shifted from primarily commercial breeders and vendors to individual laboratories, returning to “the status of ‘cottage industries’ much like the early 1900s” (Barthold 2004). Pathogenic contamination of germ line cells (spermatozoa and oocytes), embryos,
84550.indb 345
11/11/10 6:26:21 PM
346
Handbook of Laboratory Animal Science
or embryonic stem cells may also inadvertently spread viruses among mouse colonies (Agca et al. 2007; Mahabir, Bauer, and Schmidt 2008; Mahabir et al. 2009). Many of the microbes that infect laboratory animals do result in clinical illness. Due to modern husbandry techniques, more controlled environments, and health monitoring programs, the incidence of many of these diseases is low (Pritchett-Corning et al. 2009). Clinical illness in rodents may manifest itself as poor body condition, weight loss, poor hair coat, decreases in reproductive measures, or death. Larger animals may display signs of illness as anorexia or lethargy, among others. Clinical illness often alters the physiologic and behavioral parameters of the affected animals. The effect on research due to clinical illness in a colony of animals may be devastating. Fortunately, most large laboratory animals, such as dogs, cats, and pigs, have been vaccinated against the major causes of illness (Rehbinder et al. 1998). Rodent diseases with high morbidity and mortality rates have not been controlled by vaccines but rather by strict control of husbandry, sanitation, animal purchase, and health surveillance procedures. Clinical illness may cause other serious problems for research groups, resulting in unreliable data, loss of study animals, or poor reproducibility. In some immunodeficient animals or genetically modified animals, the disease may persist or become amplified in the colony (Percy and Barthold 2007) because the individual animals are incapable of clearing the infectious agent. Subclinical infections pose a more serious risk to disruption of research than do clinical infections, because there are no visible manifestations or warning signs of disease. Researchers may not even be aware that the colony is infected or of the serious complications that the infection may create for the research. Pathogens may become enzootic within a colony or population. Factors contributing to this may include genetic resistance (Janus et al. 2008) and maternal transfer of antibodies to the neonate (Barthold 2004). The immunological effects of disease may persist for months after the infection has occurred (Frost and Hamm 1990; GV-SOLAS 1999; Lipman and Perkins 2002). Gross pathological or histopathological changes that are discovered at the end point of a study may have been caused by the pathogen or host response to the infection, rather than by the experimental procedure or compound being tested, thus leading to false conclusions. The lesions associated with subclinical Helicobacter spp. infection in mice are an example. Infection with these bacteria may produce chronic proliferative hepatitis (Fox et al. 1996; Baker 1998; Elsaied et al. 2009; Huang et al. 2009) or inflammatory large bowel disease (Ward et al. 1996; Cahill et al. 1997; Shomer et al. 1997; GV-SOLAS 1999; Nicklas et al. 2002), both of which may confound research objectives (Truett et al. 2000; Whary et al. 2000; Chichlowski and Hale 2009). Changes in behavior, activity level, memory, or learning are known to occur in rodents with endoparasitic infections (McNair and Timmons 1977; Mohn and Philipp 1981; Webster 1994; Braithwaite et al. 1998; Agersborg et al. 2001). Endoparasite infection has been reported to alter nonopioidmediated analgesia (Kavaliers and Colwell 1994), inhibit growth rate (Wagner 1988), and alter the immune response (Sato et al. 1995). With many strains of rodents, the environment in which they live can influence the expression of phenotype. Several studies compared strains of laboratory rodents for differing susceptibility to various pathogens, such as Mycoplasma pulmonis (Reyes et al. 2000) or Sendai virus (Parker et al. 1978). The differences are attributed to host-specific factors and immune system differences among the strains. The nonobese diabetic (NOD) mouse strain develops insulin-dependent diabetes mellitus, but at different frequencies, depending on the microbial environment in which they live. After mouse hepatitis virus exposure, the diabetes incidence in one colony decreased significantly (Wilberz et al. 1991). Other parasitic and microbial factors have influenced the incidence of diabetes in this model (Oldstone 1990; Cooke et al. 1999; Yoon and Jun 2006; van der Werf et al. 2007). The microbial flora of an individual animal can influence development of autoimmunity. Several studies have compared the incidence of autoimmune diseases in conventionally reared animals to specific pathogenfree reared animals and have found significant differences between the two groups with all other
84550.indb 346
11/11/10 6:26:21 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
347
factors remaining constant (Penhale and Young 1988; Brabb et al. 1997). Stimulation of the immune system, chronic infection, or chronic inflammation may contribute to the variation documented. With expanding populations of transgenic animals with untested and unknown phenotypes in laboratory animal facilities, the standardization of microbial flora is crucial in determining the true characteristics of any transgenic strain. The elimination of biological variation may become even more important as toxicological studies, immunological studies, or aging studies become more common. These experimental categories rely heavily on standardized animal models for valid and reliable results (Sebesteny 1991; van der Logt 1991). The end point assays often measure more specialized functions of cells or organs of laboratory animals, requiring more uniformity among the animals. Any long-term study depends on the consistency of the animal and experimental variables throughout the entire length of the study. Any change in the health status of animals used in a study may disrupt the final interpretation of the data and may render the study not reproducible. Table€12.1 illustrates some of the research complications associated with certain murine pathogens. This table includes only a small fraction of the possible ways in which microbial factors can alter research data in mice. There are at least two other major benefits of controlling the microbial status of laboratory animals: reducing the risk of zoonotic disease exposure to humans and increasing the welfare of the animals (Fray et al. 2008). Several animal pathogens have zoonotic potential. Some, like herpes B virus or lymphocytic choriomeningitis virus, have devastating effects in people infected with these agents (Muchmore 1987; Fox et al. 2002; Troan et al. 2007; Charrel and de Lamballerie 2010). When the risk of zoonotic disease exposure is high, the quality and scope of research is compromised due to concern for human safety. With the increasing concerns for animal welfare and the adoption of refinement and reduction techniques, eliminating animal disease will benefit the quality of life for the laboratory animals, as well as the quality of research.
Abiotic Factors The primary emphasis of this section of the chapter is on the abiotic factors that operate inside the cage, pen, or tank housing the experimental animals. The major components that comprise these factors include the materials used for the walls, bottom, and top of the cage; the diet; the bedding; the water bottle or nipple; and enrichment items like nestlets or aspen blocks. The optimal situation would, of course, be that anything introduced into the cage would be made of material already present or of inert materials, such as glass and stainless steel. The question is not only whether the item is toxic or not, but also whether it could potentially interfere with the study. It may not be a problem in every study, but it is easy to conceive studies in which new materials could be problematic. For the sake of certainty, it may be necessary to adhere to a “no new materials” approach. This approach has often been ignored when designing environmental enrichment programs. Bedding A variety of bedding materials, such as cedar, pine, aspen, corncob, and shredded paper, are in common use. Laboratory animals are all the time in direct contact with the bedding—wire mesh floors are not allowed, for example, in Europe—so it is a crucial environmental factor to the animals. Bedding properties that may have an impact on animals and results fall into two categories: endogenous factors, such as enzyme induction, carcinogenicity, and ureolytic factors exogenous factors, such as chemical residues, microbes, and dirty environment (Vesell et al. 1973)
84550.indb 347
11/11/10 6:26:21 PM
348
Handbook of Laboratory Animal Science
Table€12.1╅References to Selected Murine Pathogens Causing Research Complications Virus Reo-3
MHV
LCM
Produces Signs of Lesions Stanley, Dorman, and Ponsford 1953; Walters et al. 1963; Walters et al. 1973 Piazza 1969
Mims and Tosolini 1969; Hotchin 1971
LDH-E
MVM
Sendai
Robinson, Cureton, and Heath 1968; Appell et al. 1971
CMV
Alters Immunity
Mycoplasma
Mycoplasma arthritidis
Cross and Parker 1967
Stanley, Dorman, and Ponsford 1953; Walters et al. 1963
de Souza and Smith 1991; Lardans et al. 1996; Schijns et al. 1998; RodriguezCuesta et al. 2005 Bro-Jorgensen and Volkert 1974; Jacobs and Cole 1976 Notkins 1965, 1971; Rowson and Mahy 1975; Hayashi, Hashimoto, and Kawashima 1994; Verdonck et al. 1994 Tattersall and Cotmore 1986; Takei et al. 1992
Cross and Parker 1967; Piazza 1969; Barthold 1986; Rodriguez-Cuesta et al. 2005 Cross and Parker 1967; Hotchin 1971; Kohler et al. 1990 Cross and Parker 1967; Notkins 1971; Rowson and Mahy 1975
Piazza 1969
Kay 1978
Cross and Parker 1967; Parker et al. 1970 Cross and Parker 1967
Lussier 1975; Selgrade et al. 1976
Bro-Jorgensen and Volkert 1974 Notkins 1965, 1971; Rowson and Mahy 1975; Takei et al. 1992
Yunis and Salazar 1993 Lussier 1975
Cross and Parker 1967 Cohen, Cross, and Mosier 1975 Lindsey et al. 1971
Cohen, Cross, and Mosier 1975 Rodriguez-Cuesta et al. 2005
Rodriguez-Cuesta et al. 2005; Schoeb et al. 2009
Ventura and Domaradzki 1967; Green 1970
Cole and Cassell 1979 Huang et al. 2009; Schmitz et al. 2009
Helicobacter spp.
Lemke et al. 2009; Takeshima et al. 2009
Ward et al. 2006; Lencioni et al. 2008 Murine norovirus
Alters Metabolism
Walters et al. 1973
Polyoma Thymic agent
Alters Neoplasia
Chichlowski et al. 2008; Chichlowski and Hale 2009; Hatakeyama 2009; Nagamine et al. 2009
Henderson 2008; Kastenmayer, Perdue, and Elkins 2008; Lencioni et al. 2008
Sources: Adapted from Jacoby, R. O., and S. W. Barthold. 1981. In Scientific Considerations in Monitoring and Evaluating Toxicologic Research, ed. E. J. Gralla, 27–55. Washington, D.C.: Hemisphere; Loew, F. M., and J. G. Fox. 1983. In The Mouse in Biomedical Research, vol. III, ed. H. L. Foster, J. D. Small, and J. G. Fox, 69–82. New York: Academic Press.
84550.indb 348
11/11/10 6:26:22 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
349
A study by Vesell (1967) clearly demonstrated a causal relationship between softwood bedding and increased liver microsomal enzyme activity. This was later attributed to cedrol or α-cedrene in cedar wood (Bang and Ourisson 1975) and to α-pinene in spruce and pine wood (Nielsen et al. 1984). Other effects of pine beddings on animals include endocytosis, gastrointestinal mucosal immune response, and tissue antioxidant concentration (Pick and Little 1965; Sabine 1975; Cunliffe-Beamer et al. 1981; Nielsen et al. 1986; Weichbrod et al. 1988; Sanford et al. 2002; Buddaraju and Van Dyke 2003; Davey et al. 2003). Despite ample time since the study of Vesell (1967) for implementation, a study on volatile compounds in selected, commonly used European beddings showed that many are virtually loaded with harmful compounds and autoclaving is a way to circumvent the problem (Nevalainen and Vartiainen 1996) (Figure€ 12.2). Because heat treatments (like drying) used by the manufacturer partially lower these concentrations, facilities may be using beddings with highly variable profiles of volatile substances. Such variability is clearly undesirable in attempts to conduct reliable and valid experiments. 2500
Concentration (µg/g) Limonene 3-carene
2000
Camphene Betapinene Alfapinene
1500
Toluene Heptanal Hexanal
1000
Pentanal Propanal 500
0
A
A1
D
D1
E
E1
G
G1
(a) Figure 12.2â•…Volatile compounds in commonly used beddings before and after autoclaving. (a) Illustrates summative concentrations of 10 volatile compounds of beddings with the highest sum; (b) illustrates those with the lowest sum. Capital letters by x-axis depict bedding code, when followed by a, the same when autoclaved. (Reprinted with permission from Nevalainen, T., and T. Vartiainen. 1996. Scandinavian Journal of Laboratory Animal Science 23:101–104.)
84550.indb 349
11/11/10 6:26:25 PM
350
Handbook of Laboratory Animal Science
Concentration (µg/g)
50
45
40
Limonene 3-carene
35
Camphene Betapinene
30
Alfapinene Toluene
25
Heptanal Hexanal
20
Pentanal 15
Propanal
10
5
0
B
Ba
C
Ca
F
Fa
H
Ha
(b) Figure 12.2â•…(Continued)
It has been suggested that certain of the bedding materials used may contain carcinogenic compounds. Some studies demonstrate an association between the use of cedar material and an increased incidence of spontaneous tumors in rodents (Schoental 1973, 1974). More specifically, Sabine (1975) showed that cedar bedding was associated with a higher frequency of hepatic and mammary tumors in some mouse strains; however, in some studies in mice no effect could be seen (Vlahakis 1977; Jacobs and Dieter 1978; Tennekes et al. 1981). Urease is an enzyme commonly found in plants and hence also in bedding materials of plant origin. A study comparing various beddings found three- to sevenfold amounts of urease activity in heat-treated hardwood as compared to pelleted alfalfa or pelleted corncobs, with large variation between batches. Furthermore, a heat-stable activator of bacterial urease was found in hardwood chips and crushed corncobs (Gale and Smith 1981). Because enteric bacteria contain urease that is
84550.indb 350
11/11/10 6:26:36 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
351
capable of increasing cage ammonia levels, it is unclear whether the urease activity in bedding has any practical consequences. When beddings of plant origin are used, accumulation of residues such as heavy metals and herbicides can be a problem. Although there are no widely agreed upon recommendations for maximally allowable concentrations of these residues, it might make sense to apply similar recommendations to those that apply to diets (Fox 1979). In countries with an active timber industry, it must be remembered that antifungal agents are commonly used to prevent the discoloring of wood. Practical experience has shown that rodent pups do not thrive on bedding containing antifungal agents; effects on adult animals have yet to be assessed. Low concentrations of N-nitrosoamines have also been detected in bedding, but the significance of this finding is not yet clear (Silverman and Adams 1983). Microbial accumulation in bedding material before use seems to be rare. Adverse health effects attributable to microbial growth in bedding, such as fungal rhinitis (Royals et al. 1999) and tracheobronchial disease (Hubbs et al. 1991) have been reported in rats. Apparently, this is the result of poor drying or damage to packages during transport and the consequent introduction of excessive moisture. Such excessive moisture buildup could be possible in static filter-top cages and provides an optimal microenvironment for bedding fungi in the cage. However, a study to assess this with rats, with and without Penicillium added, shows the opposite (Pernu et al. 2000). The criteria used to determine when a change of bedding is required are variable. Some facilities change bedding before the fur of the animals becomes wet or soiled or when the bedding looks dirty or is smelly. Another potential criterion to determine when a bedding change is necessary could be bacterial and fungal accumulation in the bedding during use. Bedding should be changed before an exponential increase of microbial burden occurs (Haataja et al. 1989). The species and age of the animal, the density of animals within the cage, and the type of bedding are additional factors that will influence the frequency with which bedding is changed. Normally, bedding is changed once a week for mice and twice a week for rats and rabbits. New ventilation solutions, like individually ventilated cages (IVCs), enhance ventilation inside the cage to high numbers of air changes per hour. IVCs are drier, leading facilities to reduce cage change frequency considerably. A study focusing on cage microenvironment and animal health suggested that a cage change every 2 weeks with 60 air changes per hour provided optimal conditions for mice in IVCs (Reeb-Whitaker et al. 2001). However, it remains to be seen whether the environment in IVCs becomes dirty enough to cause impairment of hepatic microsomal enzymes, as has been reported in standard cages by Vesell et al. (1973). Beddings that consist of relatively small particles (smaller than 1.2 × 1.6 mm) are generally avoided by the animals, and beddings composed of large particles are preferred by rats and mice, seemingly for their manipulability (Blom et al. 1996). Ago and co-workers (2002) concluded that, out of four different paper-bedding materials, ICR male mice preferred soft materials (like newspaper) that allowed them to hide and build nests. However, Bohonowych and colleagues (2008) have shown that newspaper, printed or virgin, may be toxic. Rats and mice prefer aspen shavings over other tested materials (Mulder 1975; Ras et al. 2002). When given a choice between corncob and aspen chip bedding and a mixture of the two, both mice and rats rejected cages with pure corncob bedding during the day, and none of the animals preferred the mixture (Krohn and Hansen 2008). Additionally, Burn and colleagues (2006) have shown that aspen bedding is relatively inert compared with other wood beddings. Additional information on choosing bedding and nesting material can be found in an article by Gonder and Laber (2007). Cage Complexity Cages for laboratory animals should contain elements that increase complexity and enrich the environment. The Council of Europe expert group emphasized the need for cage complexity for
84550.indb 351
11/11/10 6:26:36 PM
352
Handbook of Laboratory Animal Science
all laboratory species, unless there are scientific or veterinary reasons not to apply it (Hansen et al. 1999), and this is now mandated by the European guidelines (Council of Europe 2006; European Union 2007). Interference with experimental outcomes might be an example of a scientific reason for not increasing complexity, and fighting between incompatible animals might be a veterinary reason. Beneficial effects of cage complexity were shown half a century ago by Hebb (1947), who reported changes in physiology, behavior, and brain anatomy. Since then, several comprehensive reports have been published that contain numerous and detailed examples of furnished environments (Olsson and Dahlborn 2002), housing refinements for mice (Jennings et al. 1998), and housing refinements for rabbits (BVAAWF/FRAME/RSPCA/UFAW 1993). In most studies, the items used to increase cage complexity have been referred to as “enrichment,” a term that can be defined in many ways. According to Newberry (1995), enrichment is “an improvement in the biological functioning of animals resulting from modifications to their environment” (p. 230). Improvements in biological functioning include increased lifetime for reproduction, increased inclusive fitness, and improved health. Chamove’s (1989) definition of enrichment, on the other hand, was based on behavior: The aim of enrichment is to support “desirable” behaviors (e.g., foraging) and reduce “undesirable” behaviors (e.g., stereotypies or hair-pulling)—in other words, to allow the animals to exhibit species-specific behaviors. Finally, Purves (1997) stated that, in addition to improving the life of animals, enrichment could reduce variability in study results, thereby reducing the number of animal subjects needed. Newberry (1995) saw a variety of problems in studies of enrichment; for instance, the control animals may have been on wire-bottom cages or alone, unlike the study groups, and added objects may have ranged from one item to diverse combinations of items. However, many of these are experimental design issues that are not necessarily relevant to studies of enrichment only. Furthermore, the term enrichment refers to an improvement, but according to Newberry (1995), the term is applied to environmental changes, rather than to the outcomes or effects of the manipulations. Although the vast majority of studies show positive refinement outcomes that can be attributed to cage complexity, negative refinement outcomes cannot be totally ruled out. Some furniture combinations have been associated with an increase in aggression, particularly in mice (Haemisch et al. 1994), resulting in the facility in question abandoning that approach. Assessment of the effects of cage complexity is laborious because of the vast number of item combinations possible. Furthermore, different mouse strains respond differently to manipulations of cage complexity (van de Weerd et al. 1994), decreasing the generalizability of the studies. It may well be that the refinement value is specific to item, strain, parameter, and facility, thus complicating the establishment of general enrichment recommendations. Whether cage complexity has a negative refinement outcome remains a controversial issue, and most likely, the last word has not been said on this topic. Many complexity regimens are originally designed to provide animals with enhanced capabilities to cope with various challenges— manipulations that are assumed to lead to less variation in study results (Broom 1986). Augustsson et al. (2003) looked at variation of behavior attributable to cage complexity in two common mouse strains and reported that strain had a greater impact on behavior than did changes in complexity. Another mouse study of the effects of cage complexity on a number of behavioral and physiological parameters in potency testing for tetanus vaccine and stress-induced hyperthermia found no effects on variability for any of the parameters measured (Van de Weerd et al. 2002). A few studies have even demonstrated that enrichment can increase variation (Haemisch and Gartner 1994; Tsai and Hackbarth 1999; Kaliste et al. 2006). Weighing both “refinement” and “reduction” outcomes simultaneously has the greatest potential for assessing the enrichment value of cage furniture. A study by Eskola et al. (1999b) looked at arbitrarily chosen physiological and serum chemistry parameters from animals housed with and without
84550.indb 352
11/11/10 6:26:36 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
353
two enrichment items made from aspen wood (a no-new-materials approach). Rats used one of the items extensively: verification of its refinement value. Eskola and colleagues applied power analyses to yield estimates of the minimum number of animals needed in each group to obtain significant results. Both increased and decreased numbers of subjects would be required, depending on the specific parameters studied. (Eskola et al. 1999a; Mering et al. 2001). If cage complexity changes variation, the scientists—some of whom may be totally unaware of which enrichment regimen is used for their study animals—may experience altered probability errors, affecting the statistical interpretations of their research. Cage furniture is unlikely to be the first factor considered when attempting to explain interindividual variation. Implementation of cage complexity should not be a hobby, but rather a systematic program of the animal facility based on scientific data. In essence, both the refinement and reduction values of specific items (or their combinations) should be assessed in order to determine which items to implement systematically. Cage Material Pertinent recommendations and guidelines on laboratory animal housing commonly contain specifications on space allocation and “enrichment” approaches (Hebb 1947; Brain et al. 1993; ILAR 1996; Hansen et al. 1999; Council of Europe 2006; European Union 2007). Only a few of these deal with the cage materials as such (Hebb 1947). Traditionally, two new and two old cage materials are used in laboratory rodent housing: polysulfone, polyetherimide, polycarbonate, and stainless steel. Because stainless steel cages have mostly been grid floor and polycarbonate cages solid bottom enclosures, all possible choices have not been utilized. Truly valid comparisons of cage materials should be made with the same floor type. The choice of cage materials has typically been based on practical considerations. Steel cages are perceived as durable and can be autoclaved, but with nontransparent walls, they make animal observations difficult. Polysulfone and polycarbonate cages provide insulation, are lighter to handle, and are transparent, thus making observations easier. The Berlin report and Rodent Refinement Working Party recommend plastic as the preferred rodent cage material (Brain et al. 1993; Jennings et al. 1998). However, especially for albino animals, the light in plastic cages may be excessive (Schlingmann et al. 1993b). Furthermore, old and used polycarbonate cages have been shown to leach bisphenol A, a compound with estrogenic effects (Howdeshell et al. 2003; Timms et al. 2005). Even though plastic cages are more popular with laboratory facility managers, the true refinement value of these housing alternatives necessitates assessment of the animals’ preferences for cages made of different materials. Comparisons of cages with solid bottoms to cages with grid floors have been performed that suggest that, overall, rats chose to dwell on solid floors rather than grids while resting and that that rats would work as hard to reach a solid floor to rest on as they would to explore a novel environment (Manser et al. 1995, 1996). When rats were given choice of a variety of beddings and wire mesh floor, the cages with sawdust and wire mesh floor were avoided. However, rats preferred different flooring during day and night, so maybe the concept of housing rats on one type of flooring should be abandoned (van de Weerd et al. 1996). The effect of cage type has also been studied in combination with the use of aspen blocks as enrichment objects. The rats housed on solid bottom were lighter, had smaller adrenals, and displayed higher serum corticosterone levels than rats on grid floor; provision of the blocks resulted in more active and less timid animals, especially with grid flooring (Eskola and Kaliste-Korhonen 1998). In another study, Kaliste-Korhonen, Kelloniemi, and Harri (1996) compared rats’ preferences
84550.indb 353
11/11/10 6:26:36 PM
354
Handbook of Laboratory Animal Science
for polycarbonate or stainless steel cages, with no clear preference demonstrated for either of the materials. In more recent studies, rats were allowed to choose between stainless steel and polycarbonate cage halves. For most cage variations presented, the rats seemed to prefer stainless steel, regardless of the cage material in which they were born and raised. However, the position of a food hopper and illumination levels seemed to have significant effects on the choice (Heikkila et al. 2001). A later study showed that cage material may be of less importance (Voipio et al. 2008). Contrary to general opinion, the empirical data suggest that rats do not find steel less attractive than polycarbonate as cage material. Furthermore, animals living in polycarbonate cages of varying degrees of transparency may be exposed to more variable lighting, a potential cause of increased variation. Relative Humidity An optimal range for relative humidity (RH) in laboratory cages exists at which no adverse effects can be seen in rodents and rabbits. ILAR suggests a range of between 30 and 70%, while European guidelines call for a more narrow range of RH (45–65%) (ILAR 2010; Council of Europe 2006; European Union 2007). Although both ranges may seem wide, together with strict temperature control, these ranges are not always easy to achieve. Due to decreased ventilation, relative humidity can be excessive in filter-top cages (Simmons et al. 1968; Murakami 1971; Corning and Lipman 1991; Lipman et al. 1992; Perkins and Lipman 1995; Baer et al. 1997; Serrano 1971; Keller et al. 1989). Consequently, waste gases, such as carbon dioxide and ammonia, may accumulate inside the cage (Serrano 1971; Corning and Lipman 1991; Lipman et al. 1992; Perkins and Lipman 1995). These conditions may negatively affect the animals and their health. Better ventilation helps in lowering humidity and waste gases in the animal room and within the cage (Fujita et al. 1981; Hasenau et al. 1993; Choi et al. 1994; Huerkamp and Lehner 1994; Reeb et al. 1997; Ooms et al. 2008). However, individually ventilated cages seem to have higher within-cage RH when the incoming air is taken from the room (Kemppinen et al. 2008). When RH is maintained between 50 and 60%, virus transmission and ammonia production are lower (Anderson and Cox 1967; Clough 1982) and puberty is earlier in mice (Drickamer 1990). Higher RH (60–70%) has been shown to increase weaning survival and body weight gain in mice (Ellendorff et al. 1970; Drickamer 1990; Jones et al. 1995). Relative humidities that are too low, which can occur during the heating season in cold climates, are associated with a specific disease called “ringtail” in rats; necrotic belts are seen around the tails, potentially resulting in partial tail amputation (Njaa et al. 1957). From an occupational health and animal health view, it is noteworthy that, at about 50% RH, viability of microbes in the air is lowest (Anderson and Cox 1967). However, low RH increases the concentration of dust particles and thus animal allergens in the room (Jones et al. 1995). Light Due to major differences between human and rodent vision, rodent room lighting designed for personnel may not be suitable for the animals. Furthermore, most of the common laboratory animals are nocturnal or crepuscular and are typically in their resting phase during most human working hours. Some commonly used laboratory animals are albinos, which are even more sensitive to light than pigmented animals. Albino animals thrive in relatively low-light-intensity conditions (about 25 lux); albino rats avoid areas in the cage above 25 lux. The same is true for pigmented rats with light intensities above 60 lux (Schlingmann et al. 1993). Albino rats develop retinal degeneration within
84550.indb 354
11/11/10 6:26:36 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
355
3 months when they are exposed to illumination of 60 lux (Stotzer et al. 1970; Williams, Howard, and Williams 1985). The use of a light spectrum in animal rooms that mimics as closely as possible natural outdoor illumination is desirable, but is difficult to accomplish, given the limitations of artificial lighting systems. Furthermore, shading systems and cage walls are obstacles that influence the intensity and spectrum of light experienced by laboratory animals. To accomplish an optimal light intensity in the cage, the animals must be provided with shade, and overall illumination in the animal room should not be too bright. Introduction of items such as shelters and tunnels to the cage provides shaded places where animals can alter their exposure to light. This is especially true in transparent cages. Common transparent cage body materials are known to cut off wavelengths below 400 nm and the animals’ preferred location in such cages is often under the hopper, which not only contains food and the water bottle (Heikkila et al. 2001) but also blocks light. Exposure to light is not equivalent for all cages in an animal room; both rack location with respect to the light source and the location of the cage in the rack (Kemppinen et al. 2008) influence the amount of light experienced by the animals in the cages. Cages on the top shelves of racks may need supplemental protection from high light intensity. Even light intensities that are well below those that lead to retinal damage have been associated with reproductive and behavioral changes (Donnelly and Saibaba 1993). Young rats play more in dim- as opposed to high-intensity illumination (Vanderschuren et al. 1995); in similarly high light intensity, rats change their sleeping position (Vanderschuren et al. 1995). Fluorescent tubes are most commonly used for animal room lighting, and the frequency of their discharge may be a welfare issue. Although Sherwin (2007) concluded that conventional tubes are not problematic in this respect, an older study showed that low-frequency flickering at 80 lux for half an hour was stressful to albino rats (Lalitha et al. 1988). The length of the daily photoperiod seems to affect reproductive cycles strongly. When lights are on less than 8 h a day, estrous cycles in rats cease, and when the photoperiod exceeds 14 h a day, rats may display constant estrus, devoid of ovulation (Clough 1982). A 1-week disturbance in light cycle affects stress and aggressive behaviors in mice (Van der Meer et al. 2004). In general, relatively equal lengths of dark and light periods during the 24 h “day” are recommended for laboratory animals (Jennings et al. 1998). Requirements for rodent housing, especially albino strains, call for a dim environment, while considerably higher light intensities are necessary for animal house staff. Schlingmann and colleagues (Schlingmann 1993) suggest a minimum room intensity of 210 lux at working height. Distance and angle to the light source, cage material and type, and the presence or absence of nontransparent shelves in the cage rack may all result in highly variable illumination levels across cage rows. Considerable effort should be devoted to making illumination as even as possible across rows and cages because only then will potential reactions to light introduce the minimum amount of unwanted variation. If even illumination cannot be accomplished, then randomizing cage positions in the rack is the next best option. Investigators must then simply live with the possibility of increased variation in the results. Burn (2008) and Castelhano-Carlos and Baumans (2009) have produced two recent comprehensive reviews on rat sensory perception that include discussion on effects of light on the animals. Sound There are two philosophical approaches to sounds and laboratory animals. The “silence is golden” attitude aims to achieve the exclusion of all sounds. The opposite attitude regards the total exclusion of sounds as deprivation of the acoustic environment for the animals and supports the provision of purposely made sounds (e.g., a radio music program). In addition to this refinement
84550.indb 355
11/11/10 6:26:37 PM
356
Handbook of Laboratory Animal Science
A-weighting
H-weighting
R-weighting
Sound Pressure Level (dB)
10 0
–10 –20 –30 –40 –50
0 80 0 12 50 20 00 31 50 50 00 80 00 12 50 0 20 00 0 31 50 0 50 00 0 80 00 0
5
50
0
31
5
20
80
12
50
20 31 .5
–60
One-Third-Octave Midband Frequency (Hz) Figure 12.3â•…Illustration of novel sound pressure weighting, called R-weighting (rat), which takes account of the rat’s hearing sensitivity, with comparison to similarly processed, new H-weighting (human) and most commonly used A-weighting (human). (Reprinted with permission from Björk, E. et al. 2000. Laboratory Animals 34:136–144.)
dimension, sounds can cause a wide variety of physiological responses, with the possibility of interfering with study results. Reactions to sounds depend on the hearing capabilities of the animals and on the quality and type of sound. Unlike humans, many animals hear high-frequency sounds (ultrasounds, >20 kHz). Mice can hear frequencies between 2 and 80 kHz and have the greatest sensitivity to sounds between 8 and 18 kHz (Heffner and Masterton 1980). For albino rats, the hearing range is from 250 Hz to 80 kHz at 70 dB, and greatest sensitivity is between 8 and 38 kHz (Gourevitch and Hack 1966; Kelly and Masterton 1977); there seems to be no difference in the audiograms of albino and pigmented rats (Heffner et al. 1994). Strain differences are common in mice, and some strains are considered to have quite a poor sense of hearing (Willott et al. 1995; Zheng et al. 1999). Because of differences in rodent and human hearing ranges, the human ear is unlikely to be the best tool to assess sound disturbance to the animals. BjÖrk and colleagues (2000) calculated a new “R-weighting” to illustrate how rats hear and analogically calculated “H-weighting” for humans. These were both then compared to the commonly used “A-weighting” for humans. The comparisons showed that studies dealing with rat sound studies require different instrumentation than that used in humans and that R-weighting provides a new and important tool for rat hearing assessment (Figure€12.3). Different sound types result in a variety of behavioral and physiological responses in animals. Auditory cortex development is delayed if there is a continuous moderate level of noise (Chang and Merzenich 2003). Changes attributable to sounds have been seen in several organs, in growth, and in the reproductive and circulatory systems (Zondek and Tamari 1964; Algers et al. 1978; Soldani et al. 1999). Sounds can also induce specific behavioral reactions, such as seizures, acoustic startle, freezing, and orientation (Hoffman and Fleshler 1963; Fleshler 1965; Kelly et al. 1987; Brudzynski and Chiu 1995). These reactions have been organized into a classification system by Voipio (1997). Noise-type sounds result in fear reactions like startle, flight, and freezing, even when the sound pressure is low. Wave-type sounds (like whistling resembling rats’ own vocalization) cause movement and listening
84550.indb 356
11/11/10 6:26:38 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
357
or no reaction at low-pressure sound levels. Adaptation to sounds occurs rapidly, but memory is short (Voipio 1997). Furthermore, this study showed that animals are able to adapt to sounds. Animal audition and the effects of sounds on laboratory animals have been reviewed extensively on at least three occasions (Turner et al. 2007; Burn 2008; Castelhano-Carlos and Baumans 2009). Animal care routines and equipment, study apparatus, doors, sinks, and ventilation systems are the principal sources of external sounds in an animal facility. More than one-half of the sources screened in an animal facility were shown to have ultrasound components, while overall frequencies and sound pressures varied considerably (Sales et al. 1988). As an example of a sound effect of an animal care routine, pouring food pellets into a hopper above the rat’s head is a considerable sound source, causing 15 dB (R) higher sound exposure levels than does pouring food into the hopper of an adjacent cage. Because of differences in hearing between humans and rats, we hear these sounds about 10–15 dB louder than rats (Voipio et al. 2006). It would seem logical to carry out experiments when the levels of noise-producing activities in the laboratory are lowest—that is, during weekends. At these times, many sound disturbances, like cage changes and adding diet to hoppers, would be absent. If it is not possible to conduct experiments on weekends, it is advisable to determine on which day or days major animal care routines in the room are done and then allow the longest possible time to elapse before commencing the study procedures. Temperature The effects of ambient temperature on animals and on experimental results have been assessed thoroughly; consequently, temperature control is in use in most facilities. There are guidelines for animal room temperature (Council of Europe 2006; European Union 2007) giving species-specific ranges. Recommendations for rodents (mice, rats, hamsters, gerbils, guinea pigs) call for temperatures between 20 and 24°C, while the range for rabbits is between 16 and 21°C. Within these ranges, it is necessary to maintain tight temperature control. However, the temperature in the room does not necessarily match the temperature inside the cage, where animals may experience different temperatures depending on a variety of factors, including cage size and type, animal and group size, ventilation inside the cage, cage change interval, and type and amount of bedding. When IVC racks were kept in the same room with open cages and ventilation drew air from the room, recorded temperatures were 1–4°C higher in IVCs, suggesting that the ventilation system is incapable of dissipating the heat produced inside the cage (Kemppinen et al. 2008). The thermoneutral zone is defined as the ambient temperature range that causes the least thermal stress to the animal (Romanovsky et al. 2002). Even though the thermoneutral zone depends on the physical environment and the setup of studies, it appears to be considerably higher than the recommended housing temperatures. The animals are compelled to make adjustments to maintain body temperature (Gonder and Laber 2007). Temperature has a variety of effects on core temperature, weight gain, delivery rate, litter size, food and water intake, organ weights, hematological values, circulatory parameters, activity, O2 consumption, and CO2 elimination in both mice and rats (Poole and Stephenson 1977; Yamauchi et al. 1981; Gwosdow and Besch 1985; Swoap et al. 2004). An elevation in ambient temperature decreases activity in male Wistar rats (Poole and Stephenson 1977), and mean arterial pressure (MAP) and HR in female SD rats and NIH Swiss mice (Swoap et al. 2004). In high temperatures, inactivity is the animal’s first attempt to reduce heat production, and thereby counteract rising body temperature. Ambient temperatures above 30°C decrease delivery rate, litter size, and weaning rate in Wistar rats (Yamauchi et al. 1981). Ideally, the temperature should be maintained at the set value ± 1°C. If this is successful, temperature is unlikely to change metabolic rate in rodents; this may cause considerable variation in
84550.indb 357
11/11/10 6:26:38 PM
358
Handbook of Laboratory Animal Science
results (Clough 1982). Conversely, 4°C changes in ambient temperature, possible within recommended temperature ranges, can result in a magnitude change of toxicity (Harri 1976). As compared to standard caging, newer housing solutions like cubicles, ventilated cabinets, isolators, and individually ventilated cages increase ventilation efficiency inside the cage, especially when no recirculation is allowed. This increases the chance for more uniformity in temperatures between the cages. If ambient temperature is important in the study, then this should result in less variation in results. Summary Like all scientific research, studies using laboratory animals must minimize variables to produce valid, reproducible data. To achieve these ends, uniform animal subjects and a controlled environment are necessary. This chapter has highlighted some of the many factors that may influence research outcomes. While high-quality animal health and husbandry may not be the subject of a research study, they are critical to ensuring its success.
References Agca, Y., B. A. Bauer, D. K. Johnson, J. K. Critser, and L. K. Riley. 2007. Detection of mouse parvovirus in Mus musculus gametes, embryos, and ovarian tissues by polymerase chain reaction assay. Comparative Medicine 57:51–56. Agersborg, S. S., K. M. Garza, and K. S. Tung. 2001. Intestinal parasitism terminates self-tolerance and enhances neonatal induction of autoimmune disease and memory. European Journal of Immunology 31:851–859. Ago, A., T. Gonda, M. Takechi, T. Takeuchi, and K. Kawakami. 2002. Preferences for paper bedding material of the laboratory mice. Experimental Animals 51:157–161. Algers, B., I. Ekesbo, and S. Stromberg. 1978. The impact of continuous noise on animal health. Acta Veterinaria Scandinavica. Supplementum 68:1–26. Anderson, J. D., and C. S. Cox. 1967. Microbial survival. In Airborne Microbes, 17th Symposium of the Society for General Microbiology, Cambridge, U.K. Appell, L. H., R. M. Kovatch, J. M. Reddecliff, and P. J. Gerone. 1971. Pathogenesis of Sendai virus infection in mice. American Journal of Veterinary Research 32:1835–1841. Augustsson, H., H. A. van de Weerd, C. L. Kruitwagen, and V. Baumans. 2003. Effect of enrichment on variation and results in the light/dark test. Laboratory Animals 37:328–340. Baer, L. A., B. J. Corbin, M. F. Vasques, and R. E. Grindeland. 1997. Effects of the use of filtered microisolator tops on cage microenvironment and growth rate of mice. Laboratory Animal Science 47:327–329. Baker, D. G. 1998. Natural pathogens of laboratory mice, rats, and rabbits and their effects on research. Clinical Microbiology Reviews 11:231–266. ———. 2003. Natural pathogens of laboratory animals: Their effects on research. Washington, D.C.: ASM Press. Bang, L., and G. Ourisson. 1975. Hydroxylation of cedrol by rabbits. Tetrahedron Letters 24:1881–1884. Barthold, S.W. 1986. Research complications and state of knowledge of rodent coronaviruses. In Complications of viral and mycoplasmal infections in rodents to toxicology research testing, ed. T. F. Hamm, 53–89. Washington, D.C.: Hemisphere. ———. 2004. Intercurrent infections in genetically engineered mice. In Mouse models of human cancer, ed. E. C. Holland, 31–41. New York: Wiley–Liss. Becker, S. D., M. Bennett, J. P. Stewart, and J. L. Hurst. 2007. Serological survey of virus infection among wild house mice (Mus domesticus) in the UK. Laboratory Animals 41:229–238. Besselsen, D. G., C. L. Franklin, R. S. Livingston, and L. K. Riley. 2008. Lurking in the shadows: Emerging rodent infectious diseases. ILAR Journal 49:277–290. Biggers, J. D., A. McLaren, and D. Michie. 1958. Variance control in the animal house. Nature 182:77–80. BjÖrk, E., T. Nevalainen, M. Hakumäki, and H. M. Voipio. 2000. R-weighting provides better estimation for rat hearing sensitivity. Laboratory Animals 34:136–144.
84550.indb 358
11/11/10 6:26:39 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
359
Blom, H. J., G. Van Tintelen, C. J. Van Vorstenbosch, V. Baumans, and A. C. Beynen. 1996. Preferences of mice and rats for types of bedding material. Laboratory Animals 30:234–244. Bohonowych, J. E., B. Zhao, A. Timme-Laragy, D. Jung, R. T. Di Giulio, and M. S. Denison. 2008. Newspapers and newspaper ink contain agonists for the ah receptor. Toxicological Sciences 102:278–290. Brabb, T., A. W. Goldrath, P. von Dassow, A. Paez, H. D. Liggitt, and J. Goverman. 1997. Triggers of autoimmune disease in a murine TCR-transgenic model for multiple sclerosis. Journal of Immunology (Baltimore, MD: 1950) 159:497–507. Brain, P. F., D. Buttner, P. Costa, J. A. Gregory, W. O. P. Heine, J. Koolhaas, K. Militzer, F. O. Ödberg, W. Scharmann, and M. Stauffacher. 1993. Rodents. Paper presented at The Accommodation of Laboratory Animals in Accordance with Animal Welfare Requirements, Berlin. Braithwaite, V. A., D. J. Salkeld, H. M. McAdam, C. G. Hockings, A. M. Ludlow, and A. F. Read. 1998. Spatial and discrimination learning in rodents infected with the nematode Strongyloides ratti. Parasitology 117:145–154. Bro-Jorgensen, K., and M. Volkert. 1974. Defects in the immune system of mice infected with lymphocytic choriomeningitis virus. Infection and Immunity 9:605–614. Broom, D. M. 1986. Indicators of poor welfare. British Veterinary Journal 142:524–526. Brudzynski, S. M., and E. M. Chiu. 1995. Behavioral responses of laboratory rats to playback of 22 kHz ultrasonic calls. Physiology & Behavior 57:1039–1044. Buddaraju, A. K., and R. W. Van Dyke. 2003. Effect of animal bedding on rat liver endosome acidification. Comparative Medicine 53:616–621. Burn, C. C. 2008. What is it like to be a rat? Rat sensory perception and its implications for experimental design and rat welfare. Applied Animal Behavior Science 112:1–32. Burn, C. C., A. Peters, M. J. Day, and G. J. Mason. 2006. Long-term effects of cage-cleaning frequency and bedding type on laboratory rat health, welfare, and handleability: A cross-laboratory study. Laboratory Animals 40:353–370. BVAAWF/FRAME/RSPCA/UFAW. 1993. Refinements in rabbit husbandry. Second report of the BVAAWF/ FRAME/RSPCA/UFAW Joint Working Group on Refinement. British Veterinary Association Animal Welfare Foundation. Fund for the Replacement of Animals in Medical Experiments. Royal Society for the Prevention of Cruelty to Animals. Universities Federation for Animal Welfare. Laboratory Animals 27:301–329. Cahill, R. J., C. J. Foltz, J. G. Fox, C. A. Dangler, F. Powrie, and D. B. Schauer. 1997. Inflammatory bowel disease: An immunity-mediated condition triggered by bacterial infection with Helicobacter hepaticus. Infection and Immunity 65:3126–3131. Castelhano-Carlos, M. J., and V. Baumans. 2009. The impact of light, noise, cage cleaning and in-house transport on welfare and stress of laboratory rats. Laboratory Animals 43:311–327. Chamove, A. S. 1989. Environmental enrichment: A review. Animal Technology 40:155–178. Chang, E. F., and M. M. Merzenich. 2003. Environmental noise retards auditory cortical development. Science (New York) 300:498–502. Charrel, R. N., and X. de Lamballerie. 2010. Zoonotic aspects of arenavirus infections. Veterinary Microbiology 140:213–220 Chichlowski, M., and L. P. Hale. 2009. Effects of Helicobacter infection on research: The case for eradication of Helicobacter from rodent research colonies. Comparative Medicine 59:10–17. Chichlowski, M., J. M. Sharp, D. A. Vanderford, M. H. Myles, and L. P. Hale. 2008. Helicobacter typhlonius and Helicobacter rodentium differentially affect the severity of colon inflammation and inflammationassociated neoplasia in IL 10-deficient mice. Comparative Medicine 58:534–541. Choi, G. C., J. S. McQuinn, B. L. Jennings, D. J. Hassett, and S. E. Michaels. 1994. Effect of population size on humidity and ammonia levels in individually ventilated microisolation rodent caging. Contemporary Topics in Laboratory Animal Science 33:77–81. Clifford, C. B., and J. Watson. 2008. Old enemies, still with us after all these years. ILAR Journal 49:291–302. Clough, G. 1982. Environmental effects on animals used in biomedical research. Biological Reviews of the Cambridge Philosophical Society 57:487–523. Cohen, P. L., S. S. Cross, and D. E. Mosier. 1975. Immunologic effects of neonatal infection with mouse thymic virus. Journal of Immunology (Baltimore, MD: 1950) 115:706–710. Cole, B. C., and G. H. Cassell. 1979. Mycoplasma infections as models of chronic joint inflammation. Arthritis and Rheumatism 22:1375–1381.
84550.indb 359
11/11/10 6:26:39 PM
360
Handbook of Laboratory Animal Science
Cooke, A., P. Tonks, F. M. Jones, H. O’Shea, P. Hutchings, A. J. Fulford, and D. W. Dunne. 1999. Infection with Schistosoma mansoni prevents insulin dependent diabetes mellitus in nonobese diabetic mice. Parasite Immunology 21:169–176. Corning, B. F., and N. S. Lipman. 1991. A comparison of rodent caging systems based on microenvironmental parameters. Laboratory Animal Science 41:498–503. Council of Europe. 2006. Appendix of the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (ETS123). Guidelines for accommodation and care of animals (Article 5 of the convention) approved by the multilateral consultation. Trans. Council of Europe. Vol. Cons 123 (2006) 3. Strasbourg. Cross, S. S., and J. C. Parker. 1967. Viral contaminants of mouse tumor system. Abstract. Bacteriology Proceedings. Cunliffe-Beamer, T. L., L. C. Freeman, and D. D. Myers. 1981. Barbiturate sleep time in mice exposed to autoclaved or unautoclaved wood beddings. Laboratory Animal Science 31:672–675. Davey, A. K., J. P. Fawcett, S. E. Lee, K. K. Chan, and J. C. Schofield. 2003. Decrease in hepatic drug-metabolizing enzyme activities after removal of rats from pine bedding. Comparative Medicine 53:299–302. de Souza, M. S., and A. L. Smith. 1991. Characterization of accessory cell function during acute infection of BALB/cByJ mice with mouse hepatitis virus (MHV), strain JHM. Laboratory Animal Science 41:112–118. Donnelly, H., and P. Saibaba. 1993. Light intensity and the estrous cycle in albino and normally pigmented mice. Laboratory Animals 27:385–390. Drickamer, L. C. 1990. Environmental factors and age of puberty in female house mice. Developmental Psychobiology 23:63–73. Easterbrook, J. D., J. B. Kaplan, G. E. Glass, J. Watson, and S. L. Klein. 2008. A survey of rodent-borne pathogens carried by wild-caught Norway rats: A potential threat to laboratory rodent colonies. Laboratory Animals 42:92–98. Ellendorff, F., D. Smidt, M. Monzavifar, and P. de Boer. 1970. Temperature, humidity and reproduction in mice. II. Influence of 35°C, 45 and 91% relative humidity on body weight and genital tract. International Journal of Biometeorology 14:85–93. Ellery, A. W. 1985. Guidelines for specification of animals and husbandry methods when reporting the results of animal experiments. Working Committee for the Biological Characterization of Laboratory Animals/ GV-SOLAS. Laboratory Animals 19:106–108. Elsaied, N. A., A. Y. Abbas, S. N. El-Beshbishi, and H. M. Elsheikha. 2009. Increased Helicobacter pylori-associated pathology in outbred mice coinfected with Schistosomiasis. Parasitology Research 105:297–299. Eskola, S., and E. Kaliste-Korhonen. 1998. Effects of cage type and gnawing blocks on weight gain, organ weights and open-field behavior in Wistar rats. Scandinavian Journal of Laboratory Animal Science 25:180–193. Eskola, S., M. Lauhikari, H. M. Voipio, M. Laitinen, and T. Nevalainen. 1999a. Environmental enrichment may alter the number of rats needed to achieve statistical significance. Scandinavian Journal of Laboratory Animal Science 26:134–144. Eskola, S., M. Lauhikari, H. M. Voipio, and T. Nevalainen. 1999b. The use of aspen blocks and tubes to enrich the cage environment of laboratory rats. Scandinavian Journal of Laboratory Animal Science 26:1–10. European Union. 2007. Commission recommendation of June 18, 2007 on guidelines for the accommodation and care of animals used for experimental and other scientific purposes (2007/526/EC). Fleshler, M. 1965. Adequate acoustic stimulus for startle reaction in the rat. Journal of Comparative and Physiological Psychology 60:200–207. Fox, J. 1979. Selected aspects of animal husbandry and good laboratory practices. Clinical Toxicology 15:539–553. Fox, J. G., X. Li, L. Yan, R. J. Cahill, R. Hurley, R. Lewis, and J. C. Murphy. 1996. Chronic proliferative hepatitis in A/JCr mice associated with persistent Helicobacter hepaticus infection: A model of Helicobacterinduced carcinogenesis. Infection and Immunity 64:1548–1558. Fox, J. G., C. E. Newcomer, and H. Rozmiarek. 2002. Selected zoonoses. In Laboratory animal medicine, 2nd ed., ed. J. G. Fox, L. C. Anderson, F. M. Loew, and F. W. Quimby, 1060–1105. San Diego, CA: Academic Press. Fray, M. D., A. R. Pickard, M. Harrison, and M. T. Cheeseman. 2008. Upgrading mouse health and welfare: Direct benefits of a large-scale rederivation program. Laboratory Animals 42:127–139. Frost, W. W., and T. E. Hamm, Jr. 1990. Prevention and control of animal disease. In The experimental animal in biomedical research, ed. B. E. Rollin and M. L. Kesel, 133–152. Boca Raton, FL: CRC Press.
84550.indb 360
11/11/10 6:26:39 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
361
Fujita, S., T. Obara, I. Tanaka, and C. Yamauchi. 1981. Distribution of environmental temperature and relative humidity according to the number of conditioned air changes in laboratory animals rooms. Experimental Animals 30:21–29. Gale, G. R., and A. B. Smith. 1981. Ureolytic and urease-activating properties of commercial laboratory animal bedding. Laboratory Animal Science 31:56–59. Gonder, J. C., and K. Laber. 2007. A renewed look at laboratory rodent housing and management. ILAR Journal 48:29–36. Gourevitch, G., and M. H. Hack. 1966. Audibility in the rat. Journal of Comparative and Physiological Psychology 62:289–291. Green, G. M. 1970. The J. Burns Amberson lecture—In defense of the lung. American Review of Respiratory Disease 102:691–703. GV-SOLAS. 1999. Implications of infectious agents on results of animal experiments. Report of the Working Group on Hygiene of the Gesellschaft für Versuchstierkunde—Society for Laboratory Animal Science (GV-SOLAS). Laboratory Animals 33 (Suppl 1): S39–S87. Gwosdow, A. R., and E. L. Besch. 1985. Effect of thermal history on the rat’s response to varying environmental temperature. Journal of Applied Physiology (Bethesda, MD: 1985) 59:413–419. Haataja, H., H. M. Voipio, and T. Nevalainen. 1989. Bedding bacterial counts as indicators of hygiene in rats. Scandinavian Journal of Laboratory Animal Science 16:123–126. Haemisch, A., and K. Gartner. 1994. The cage design affects intermale aggression in small groups of male laboratory mice: Strain specific consequences on social organization, and endocrine activations in two inbred strains (DBA/2J and CBA/J). Journal of Experimental Animal Science 36:101–116. Haemisch, A., T. Voss, and K. Gartner. 1994. Effects of environmental enrichment on aggressive behavior, dominance hierarchies, and endocrine states in male DBA/2J mice. Physiology & Behavior 56:1041–1048. Hansen, A. K., V. Baumans, H. Elliot, R. Francis, B. Holgate, R. Hubrecht, M. Jennings, A. Peters, and M. Stauffacher. 1999. Future principles for housing and care of laboratory rodents and rabbits. Report concerning revision of the Council of Europe Convention ETS 123 Appendix A concerning questions related to rodents and rabbits issued by the Council’s Working Group for Rodents and Rabbits. Part A, actions and proposals of the working group. Harri, M. N. 1976. Effect of body temperature on cardiotoxicity of isoprenaline in rats. Acta Pharmacologica et Toxicologica 39:214–224. Hasenau, J. J., R. B. Baggs, and A. L. Kraus. 1993. Microenvironments in microisolation cages using BALB/c and CD-1 mice. Contemporary Topics in Laboratory Animal Science 33:11–16. Hatakeyama, M. 2009. Helicobacter pylori and gastric carcinogenesis. Journal of Gastroenterology 44:239–248. Hayashi, T., S. Hashimoto, and A. Kawashima. 1994. Effect of infection by lactic dehydrogenase virus on expression of intercellular adhesion molecule-1 on vascular endothelial cells of pancreatic islets in streptozotocin-induced insulitis of CD-1 mice. International Journal of Experimental Pathology 75:211–217. Hebb, D. O. 1947. The effects of early experience on problem-solving at maturity. American Journal of Psychology 2:306–307. Heffner, H., and B. Masterton. 1980. Hearing in glires—Domestic rabbit, cotton rat, feral house mouse, and kangaroo rat. Journal of the Acoustical Society of America 68:1584–1599. Heffner, H. E., R. S. Heffner, C. Contos, and T. Ott. 1994. Audiogram of the hooded Norway rat. Hearing Research 73:244–247. Heikkila, M., R. Sarkanen, H. M. Voipio, S. Mering, and T. Nevalainen. 2001. Cage position preferences of rats. Scandinavian Journal of Laboratory Animal Science 28:65–74. Henderson, K. S. 2008. Murine norovirus, a recently discovered and highly prevalent viral agent of mice. Lab Anim (NY) 37:314–320. Hoffman, H. S., and M. Fleshler. 1963. Startle reaction: Modification by background acoustic stimulation. Science (New York) 141:928–930. Hooijmans, C. R., M. Leenaars, and M. Ritskes-Hoitinga. 2010. A gold standard publication checklist to improve the quality of animal studies, to fully integrate the three Rs, and to make systematic reviews more feasible. ATLA-Alternatives to Laboratory Animals 38:167–182. Hotchin, J. 1971. The contamination of laboratory animals with lymphocytic choriomeningitis virus. American Journal of Pathology 64:747–769.
84550.indb 361
11/11/10 6:26:39 PM
362
Handbook of Laboratory Animal Science
Howdeshell, K. L., P. H. Peterman, B. M. Judy, J. A. Taylor, C. E. Orazio, R. L. Ruhlen, F. S. Vom Saal, and W. V. Welshons. 2003. Bisphenol A is released from used polycarbonate animal cages into water at room temperature. Environmental Health Perspectives 111:1180–1187. Huang, Y., X. F. Tian, X. F. Fan, C. Y. Fu, and C. Zhu. 2009. The pathological effect of Helicobacter pylori infection on liver tissues in mice. Clinical Microbiology and Infection 15:843–849. Hubbs, A. F., F. F. Hahn, and D. L. Lundgren. 1991. Invasive tracheobronchial aspergillosis in an F344/N rat. Laboratory Animal Science 41:521–524. Huerkamp, M. J., and N. D. Lehner. 1994. Comparative effects of forced-air, individual cage ventilation or an absorbent bedding additive on mouse isolator cage microenvironment. Contemporary Topics in Laboratory Animal Science 33:58–61. ILAR. 2010. Guide for the care and use of laboratory animals. Washington, D.C.: National Academy Press. Jacobs, B. B., and D. K. Dieter. 1978. Spontaneous hepatomas in mice inbred from ha:ICR Swiss stock: Effects of sex, cedar shavings in bedding, and immunization with fetal liver or hepatoma cells. Journal of the National Cancer Institute 61:1531–1534. Jacobs, R. P., and G. A. Cole. 1976. Lymphocytic choriomeningitis virus-induced immunosuppression: A virusinduced macrophage defect. Journal of Immunology (Baltimore, MD: 1950) 117:1004–1009. Jacoby, R. O., and S. W. Barthold. 1981. Quality assurance for rodents used in toxicological research and testing. In Scientific considerations in monitoring and evaluating toxicologic research, ed. E. J. Gralla, 27–55. Washington, D.C.: Hemisphere. Janus, L. M., M. Mähler, W. Köhl, A. Smoczek, H. J. Hedrich, and A. Bleich. 2008. Minute virus of mice: Antibody response, viral shedding, and persistence of viral DNA in multiple strains of mice. Comparative Medicine 58:360–368. Jennings, M., G. R. Batchelor, P. F. Brain, A. Dick, H. Elliott, R. J. Francis, R. C. Hubrecht, et al. 1998. Refining rodent husbandry: The mouse. Report of the Rodent Rrefinement Working Party. Laboratory Animals 32:233–259. Jones, R. B., J. B. Kacergis, M. R. MacDonald, F. T. McKnight, W. A. Turner, J. L. Ohman, and B. Paigen. 1995. The effect of relative humidity on mouse allergen levels in an environmentally controlled mouse room. American Industrial Hygiene Association Journal 56:398–401. Kaliste, E. K., S. M. Mering, and H. K. Huuskonen. 2006. Environmental modification and agonistic behavior in NIH/S male mice: Nesting material enhances fighting but shelters prevent it. Comparative Medicine 56:202–208. Kaliste-Korhonen, E., J. Kelloniemi, and M. Harri. 1996. Cage material and rat behavior. Scandinavian Journal of Laboratory Animal Science 23:125–128. Kastenmayer, R. J., K. A. Perdue, and W. R. Elkins. 2008. Eradication of murine norovirus from a mouse barrier facility. Journal of American Association for Laboratory Animal Science 47:26–30. Kavaliers, M., and D. D. Colwell. 1994. Parasite infection attenuates nonopioid mediated predator-induced analgesia in mice. Physiology & Behavior 55:505–510. Kay, M. M. 1978. Long-term subclinical effects of parainfluenza (Sendai) infection on immune cells of aging mice. Proceedings of the Society for Experimental Biology and Medicine 158:326–331. Keller, L. S., W. J. White, M. T. Snider, and C. M. Lang. 1989. An evaluation of intracage ventilation in three animal caging systems. Laboratory Animal Science 39:237–242. Kelly, J. B., P. W. Judge, and I. H. Fraser. 1987. Development of the auditory orientation response in the albino rat (Rattus norvegicus). Journal of Comparative Psychology (Washington, D.C.: 1983) 101:60–66. Kelly, J. B., and B. Masterton. 1977. Auditory sensitivity of the albino rat. Journal of Comparative and Physiological Psychology 91:930–936. Kemppinen, N., A. Meller, E. Bjork, T. Kohila, and T. Nevalainen. 2008. Exposure in the shoebox: Comparison of physical environment of IVCs and open rat cages. Scandinavian Journal of Laboratory Animal Science 35:97–103. Kohler, M., B. Ruttner, S. Cooper, H. Hengartner, and R. M. Zinkernagel. 1990. Enhanced tumor susceptibility of immunocompetent mice infected with lymphocytic choriomeningitis virus. Cancer Immunology, Immunotherapy: CII 32:117–124. Krohn, T. C., and A. K. Hansen. 2008. Evaluation of corncob as bedding for rodents. Scandinavian Journal of Laboratory Animal Science 35:231–236. Lalitha, R., N. Suthanthirarajan, and A. Namasivayam. 1988. Effect of flickering light stress on certain biochemical parameters in rats. Indian Journal of Physiology and Pharmacology 32:182–186.
84550.indb 362
11/11/10 6:26:39 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
363
Lardans, V., C. Godfraind, J. T. van der Logt, W. A. Heessen, M. D. Gonzalez, and J. P. Coutelier. 1996. Polyclonal B lymphocyte activation induced by mouse hepatitis virus A59 infection. Journal of General Virology 77:1005–1009. Lemke, L. B., Z. Ge, M. T. Whary, Y. Feng, A. B. Rogers, S. Muthupalani, and J. G. Fox. 2009. Concurrent Helicobacter bilis infection in C57BL/6 mice attenuates proinflammatory H. pylori-induced gastric pathology. Infection and Immunity 77:2147–2158. Lencioni, K. C., A. Seamons, P. M. Treuting, L. Maggio-Price, and T. Brabb. 2008. Murine norovirus: An intercurrent variable in a mouse model of bacteria-induced inflammatory bowel disease. Comparative Medicine 58:522–533. Liang, C. T., A. Shih, Y. H. Chang, C. W. Liu, Y. T. Lee, W. C. Hsieh, Y. L. Huang, et al. 2009. Microbial contaminations of laboratory mice and rats in Taiwan from 2004 to 2007. Journal of American Association for Laboratory Animal Science 48:381–386. Lindsey, J. R., H. J. Baker, R. G. Overcash, G. H. Cassell, and C. E. Hunt. 1971. Murine chronic respiratory disease. Significance as a research complication and experimental production with Mycoplasma pulmonis. American Journal of Pathology 64:675–708. Lipman, N. S., B. F. Corning, and M. A. Coiro. 1992. The effects of intracage ventilation on microenvironmental conditions in filter-top cages. Laboratory Animals 26:206–210. Lipman, N. S., and S. E. Perkins. 2002. Factors that may influence animal research. In Laboratory animal medicine, 2nd ed., ed. J. G. Fox, L. C. Anderson, F. M. Loew, and F. W. Quimby, 1143–1184. San Diego, CA: Academic Press. Loew, F. M. and J. G. Fox. 1983. In The mouse in biomedical research, vol. III, ed. H. L. Foster, J. D. Small, and J. G. Fox, 69–82. New York: Academic Press. Lussier, G. 1975. Murine cytomegalovirus (MCMV). Advances in Veterinary Science and Comparative Medicine 19:223–247. Mahabir, E., B. Bauer, and J. Schmidt. 2008. Rodent and germplasm trafficking: Risks of microbial contamination in a high-tech biomedical world. ILAR Journal 49:347–355. Mahabir, E., K. Reindl, J. Mysliwietz, J. Needham, D. Bulian, K. Markoullis, H. Scherb, and J. Schmidt. 2009. Impairment of germ line transmission after blastocyst injection with murine embryonic stem cells cultured with mouse hepatitis virus and mouse minute virus. Transgenic Research 18:45–57. Mähler, M., and W. Köhl. 2009. A serological survey to evaluate contemporary prevalence of viral agents and Mycoplasma pulmonis in laboratory mice and rats in Western Europe. Lab Animal (NY) 38:161–165. Manser, C. E., H. Elliot, T. H. Morris, and D. M. Broom. 1996. The use of a novel operant test to determine the strength of preference for flooring in laboratory rats. Laboratory Animals 30:1–6. Manser, C. E., T. H. Morris, and D. M. Broom. 1995. An investigation into the effects of solid or grid cage flooring on the welfare of laboratory rats. Laboratory Animals 29:353–363. McNair, D. M., and E. H. Timmons. 1977. Effects of Aspiculuris tetraptera and Syphacia obvelata on exploratory behavior of an inbred mouse strain. Laboratory Animal Science 27:38–42. Mering, S., E. Kaliste-Korhonen, and T. Nevalainen. 2001. Estimates of appropriate number of rats: Interaction with housing environment. Laboratory Animals 35:80–90. Mims, C. A., and F. A. Tosolini. 1969. Pathogenesis of lesions in lymphoid tissue of mice infected with lymphocytic choriomeningitis (LCM) virus. British Journal of Experimental Pathology 50:584–592. Mohn, G., and E. M. Philipp. 1981. Effects of Syphacia muris and the anthelmintic fenbendazole on the microsomal monooxygenase system in mouse liver. Laboratory Animals 15:89–95. Muchmore, E. 1987. An overview of biohazards associated with nonhuman primates. Journal of Medical Primatology 16:55–82. Mulder, J. B. 1975. Bedding preferences of pregnant laboratory-reared mice. Behavior Research Methods & Instrumentation 7:21–22. Murakami, H. 1971. Differences between internal and external environments of the mouse cage. Laboratory Animal Science 21:680–684. Nagamine, C. M., J. J. Sohn, B. H. Rickman, A. B. Rogers, J. G. Fox, and D. B. Schauer. 2009. Helicobacter hepaticus infection promotes colon tumorigenesis in the BALB/c-Rag2(–/–) apc(min/+) mouse. Infection and Immunity 76:2758–2766. Nevalainen, T., and T. Vartiainen. 1996. Volatile organic compounds in commonly used beddings before and after autoclaving. Scandinavian Journal of Laboratory Animal Science 23:101–104.
84550.indb 363
11/11/10 6:26:39 PM
364
Handbook of Laboratory Animal Science
Newberry, R. C. 1995. Environmental enrichment—Increasing the biological relevance of captive environments. Applied Animal Behavior Science 44:229–243. Nicklas, W. 2008. International harmonization of health monitoring. ILAR Journal 49:338–346. Nicklas, W., P. Baneux, R. Boot, T. Decelle, A. A. Deeny, M. Fumanelli, B. Illgen-Wilcke, and FELASA (Federation of European Laboratory Animal Science Associations Working Group on Health Monitoring of Rodent and Rabbit Colonies). 2002. Recommendations for the health monitoring of rodent and rabbit colonies in breeding and experimental units. Laboratory Animals 36:20–42. Nielsen, J. B., O. Andersen, and P. Svendsen. 1984. Effekt af stroelse på leverens cytochrome P-450 i mus. ScandLAS Nyt 11:7–13. ———. 1986. Hepatic O-deethylase activity in mice on different types of bedding. Zeitschrift für Versuchstierkunde 28:69–75. Njaa, L. R., F. Utne, and O. R. Braekkan. 1957. Effect of relative humidity on rat breeding and ringtail. Nature 180:290–291. Notkins, A. L. 1971. Enzymatic and immunologic alterations in mice infected with lactic dehydrogenase virus. American Journal of Pathology 64:733–746. ———. 1965. Lactic dehydrogenase virus. Bacteriological Reviews 29:143–160. Oldstone, M. B. 1990. Viruses as therapeutic agents. I. treatment of nonobese insulin-dependent diabetes mice with virus prevents insulin-dependent diabetes mellitus while maintaining general immune competence. Journal of Experimental Medicine 171:2077–2089. Olsson, I. A., and K. Dahlborn. 2002. Improving housing conditions for laboratory mice: A review of “environmental enrichment.” Laboratory Animals 36:243–270. Ooms, T. G., J. E. Artwohl, L. M. Conroy, T. M. Schoonover, and J. D. Fortman. 2008. Concentration and emission of airborne contaminants in a laboratory animal facility housing rabbits. Journal of the American Association for Laboratory Animal Science 47:39–48. O’Rourke, J., A. Lee, and J. McNeill. 1988. Differences in the gastrointestinal microbiota of specific pathogen free mice: An often unknown variable in biomedical research. Laboratory Animals 22:297–303. Parker, J. C., M. J. Collins, Jr., S. S. Cross, and W. P. Rowe. 1970. Minute virus of mice. II. Prevalence, epidemiology, and occurrence as a contaminant of transplanted tumors. Journal of the National Cancer Institute 45:305–310. Parker, J. C., M. D. Whiteman, and C. B. Richter. 1978. Susceptibility of inbred and outbred mouse strains to Sendai virus and prevalence of infection in laboratory rodents. Infection and Immunity 19:123–130. Penhale, W. J., and P. R. Young. 1988. The influence of the normal microbial flora on the susceptibility of rats to experimental autoimmune thyroiditis. Clinical and Experimental Immunology 72:288–292. Percy, D. H., and S. W. Barthold. 2007. Pathology of laboratory rodents and rabbits, 3rd ed. Ames, IA: Blackwell Publishing. Perkins, S. E., and N. S. Lipman. 1995. Characterization and quantification of microenvironmental contaminants in isolator cages with a variety of contact beddings. Contemporary Topics in Laboratory Animal Science 34:93–98. Pernu, N., A. Hyvarinen, M. Toivola, A. Nevalainen, M. Harri, and T. Nevalainen. 2000. Filter-top cages: Moldy homes for rodents? Scandinavian Journal of Laboratory Animal Science 27:1–12. Piazza, M. 1969. Hepatitis in mice. In Experimental viral hepatitis, 13–140. Springfield, IL: Charles C Thomas. Pick, J. R., and J. M. Little. 1965. Effect of type of bedding material on thresholds of pentylenetetrazol convulsions in mice. Laboratory Animal Care 15:29–33. Poole, S., and J. D. Stephenson. 1977. Body temperature regulation and thermoneutrality in rats. Quarterly Journal of Experimental Physiology and Cognate Medical Sciences 62:143–149. Pritchett-Corning, K. R., J. Cosentino, and C. B. Clifford. 2009. Contemporary prevalence of infectious agents in laboratory mice and rats. Laboratory Animals 43:165–173. Purves, K. E. 1997. Rat and mice enrichment. In Animal alternatives, welfare and ethics, ed. L. F. M. van Zutphen and M. Balls, 199–207. New York: Elsevier Science B.V. Rahija, R. J. 2007. Gnotobiotics. In The mouse in biomedical research, vol. III, 2nd ed., ed. J. G. Fox, S. W. Barthold, M. T. Davisson, C. E. Newcomer, F. W. Quimby, and A. L. Smith, 217–233. San Diego, CA: Academic Press. Ras, T., M. van de Ven, E. G. Patterson-Kane, and K. Nelson. 2002. Rats’ preferences for corn versus woodbased bedding and nesting materials. Laboratory Animals 36:420–425.
84550.indb 364
11/11/10 6:26:39 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
365
Reeb, C. K., R. B. Jones, D. W. Bearg, H. Bedigian, and B. Paigen. 1997. Impact of room ventilation rates on mouse cage ventilation and microenvironment. Contemporary Topics in Laboratory Animal Science 36:74–79. Reeb-Whitaker, C. K., B. Paigen, W. G. Beamer, R. T. Bronson, G. A. Churchill, I. B. Schweitzer, and D. D. Myers. 2001. The impact of reduced frequency of cage changes on the health of mice housed in ventilated cages. Laboratory Animals 35:58–73. Rehbinder, C., P. Baneux, D. Forbes, H. van Herck, W. Nicklas, Z. Rugaya, and G. Winkler. 1998. FELASA recommendations for the health monitoring of breeding colonies and experimental units of cats, dogs and pigs. Report of the Federation of European Laboratory Animal Science Associations (FELASA) Working Group on Animal Health. Laboratory Animals 32:1–17. Reyes, L., D. A. Steiner, J. Hutchison, B. Crenshaw, and M. B. Brown. 2000. Mycoplasma pulmonis genital disease: Effect of rat strain on pregnancy outcome. Comparative Medicine 50:622–627. Robinson, T. W., R. J. Cureton, and R. B. Heath. 1968. The pathogenesis of Sendai virus infection in the mouse lung. Journal of Medical Microbiology 1:89–95. Rodriguez-Cuesta, J., F. Vidal-Vanaclocha, L. Mendoza, M. Valcarcel, N. Gallot, and G. Martinez de Tejada. 2005. Effect of asymptomatic natural infections due to common mouse pathogens on the metastatic progression of B16 murine melanoma in C57BL/6 mice. Clinical & Experimental Metastasis 22:549–558. Romanovsky, A. A., A. I. Ivanov, and Y. P. Shimansky. 2002. Selected contribution: Ambient temperature for experiments in rats: A new method for determining the zone of thermal neutrality. Journal of Applied Physiology (Bethesda, MD: 1985) 92:2667–2679. Rowson, K. E., and B. W. Mahy. 1975. Lactic dehydrogenase virus. Virology Monographs. Die Virusforschung in Einzeldarstellungen 13:1–121. Royals, M. A., D. M. Getzy, and S. VandeWoude. 1999. High fungal spore load in corncob bedding associated with fungal-induced rhinitis in two rats. Contemporary Topics in Laboratory Animal Science 38:64–66. Russell, W. M. S., and R. L. Burch. 1959. The principles of humane experimental technique. London: Methuen. Sabine, J. R. 1975. Exposure to an environment containing the aromatic red cedar, Juniperus virginiana: Procarcinogenic, enzyme-inducing and insecticidal effects. Toxicology 5:221–235. Sales, G. D., K. J. Wilson, K. E. Spencer, and S. R. Milligan. 1988. Environmental ultrasound in laboratories and animal houses: A possible cause for concern in the welfare and use of laboratory animals. Laboratory Animals 22:369–375. Sanford, A. N., S. E. Clark, G. Talham, M. G. Sidelsky, and S. E. Coffin. 2002. Influence of bedding type on mucosal immune responses. Comparative Medicine 52:429–432. Sato, Y., H. K. Ooi, N. Nonaka, Y. Oku, and M. Kamiya. 1995. Antibody production in Syphacia obvelata infected mice. Journal of Parasitology 81:559–562. Schijns, V. E., B. L. Haagmans, C. M. Wierda, B. Kruithof, I. A. Heijnen, G. Alber, and M. C. Horzinek. 1998. Mice lacking IL-12 develop polarized Th1 cells during viral infection. Journal of Immunology (Baltimore, MD: 1950) 160:3958–3964. Schlingmann, F. S. 1993. Light intensity in animal rooms and cages in relation to the care and management of albino rats. Animal Technology 44:97–107. Schlingmann, F. S., H. L. M. De Rijk, W. J. Pereboom, and R. Remie. 1993a. Avoidance as a behavioral parameter in the determination of distress among albino and pigmented rats at various light intensities. Animal Technology 44:87–96. Schlingmann, F., W. J. Pereboom, and R. Remie. 1993b. The sensitivity of albino and pigmented rats to light: A mini review. Animal Technology 44:71–86. Schmitz, J. M., C. G. Durham, S. B. Ho, and R. G. Lorenz. 2009. Gastric mucus alterations associated with murine Helicobacter infection. Journal of Histochemistry and Cytochemistry 57:457–467. Schoeb, T., E. McConnell, M. Juliana, J. Davis, M. Davidson, and J. Lindsey. 2009. Commentary: Mycoplasma pulmonis and lymphoma in bioassays in rats. Veterinary Pathology 46:952–959. Schoental, R. 1973. Carcinogenicity of wood shavings. Lab Animal 7:47–49. ———. 1974. Letter: Role of podophyllotoxin in the bedding and dietary zearalenone on incidence of spontaneous tumors in laboratory animals. Cancer Research 34:2419–2420. Schoondermark-van de Ven, E. M., I. M. Philipse-Bergmann, and J. T. van der Logt. 2006. Prevalence of naturally occurring viral infections, Mycoplasma pulmonis and Clostridium piliforme, in laboratory rodents in Western Europe screened from 2000 to 2003. Laboratory Animals 40:137–143.
84550.indb 365
11/11/10 6:26:40 PM
366
Handbook of Laboratory Animal Science
Sebesteny, A. 1991. Necessity of a more standardized microbiological characterization of rodents for aging studies. Neurobiology of Aging 12:663–668. Selgrade, M. K., A. Ahmed, K. W. Sell, M. E. Gershwin, and A. D. Steinberg. 1976. Effect of murine cytomegalovirus on the in vitro responses of T- and B-cells to mitogens. Journal of Immunology (Baltimore, MD: 1950) 116:1459–1465. Serrano, L. J. 1971. Carbon dioxide and ammonia in mouse cages: Effect of cage covers, population, and activity. Laboratory Animal Science 21:75–85. Sherwin, C. M. Validating refinements to laboratory housing: Asking the animals. 2007. In NCR3Rs [database online; cited 09/5 2009]. Available from http://www.nc3rs.org.uk/news.asp?id=759. Shomer, N. H., C. A. Dangler, M. D. Schrenzel, and J. G. Fox. 1997. Helicobacter bilis-induced inflammatory bowel disease in SCID mice with defined flora. Infection and Immunity 65:4858–4864. Silverman, J., and J. D. Adams. 1983. N-nitrosamines in laboratory animal feed and bedding. Laboratory Animal Science 33:161–164. Simmons, M. I., D. M. Robie, J. B. Jones, and L. J. Serrano. 1968. Effect of filter cover on temperature and humidity in a mouse cage. Laboratory Animals 2:113–120. Soldani, P., M. Gesi, P. Lenzi, G. Natale, F. Fornai, A. Pellegrini, M. P. Ricciardi, and A. Paparelli. 1999. Long-term exposure to noise modifies rat adrenal cortex ultrastructure and corticosterone plasma levels. Journal of Submicroscopic Cytology and Pathology 31:441–448. Stanley, N. F., D. C. Dorman, and J. Ponsford. 1953. Studies on the pathogenesis of a hitherto undescribed virus (hepato-encephalomyelitis) producing unusual symptoms in suckling mice. Australian Journal of Experimental Biology and Medical Science 31:147–159. Stotzer, H., I. Weisse, F. Knappen, and R. Seitz. 1970. Retinal degeneration in rat. Its dependence on intensity of illumination and vitamin A in food. Arzneimittel-Forschung 20:811–817. Swoap, S. J., J. M. Overton, and G. Garber. 2004. Effect of ambient temperature on cardiovascular parameters in rats and mice: A comparative approach. American Journal of Physiology, Regulatory, Integrative and Comparative Physiology 287:R391–R396. Takei, I., Y. Asaba, T. Kasatani, T. Maruyama, K. Watanabe, T. Yanagawa, T. Saruta, and T. Ishii. 1992. Suppression of development of diabetes in NOD mice by lactate dehydrogenase virus infection. Journal of Autoimmunity 5:665–673. Takeshima, E., K. Tomimori, H. Teruya, C. Ishikawa, M. Senba, D. D’Ambrosio, F. Kinjo, et al. 2009. Helicobacter pylori-induced interleukin-12 p40 expression. Infection and Immunity 77:1337–1348. Tattersall, P., and S. F. Cotmore. 1986. The rodent parvoviruses. In Viral and mycoplasmal infections of laboratory rodents: Effects on biomedical research, ed. P. N. Bhatt, R. O. Jacoby, and H. C. Morse, 305–348. San Diego, CA: Academic Press. Tennekes, H. A., A. S. Wright, K. M. Dix, and J. H. Koeman. 1981. Effects of dieldrin, diet, and bedding on enzyme function and tumor incidence in livers of male CF-1 mice. Cancer Research 41:3615–3620. Timms, B. G., K. L. Howdeshell, L. Barton, S. Bradley, C. A. Richter, and F. S. vom Saal. 2005. Estrogenic chemicals in plastic and oral contraceptives disrupt development of the fetal mouse prostate and urethra. Proceedings of the National Academy of Sciences USA 102:7014–7019. Truett, G. E., J. A. Walker, and D. G. Baker. 2000. Eradication of infection with Helicobacter spp. by use of neonatal transfer. Comparative Medicine 50:444–451. Troan, B. V., L. Perelygina, I. Patrusheva, A. J. Wettere, J. K. Hilliard, M. R. Loomis, and R. S. Voe. 2007. Naturally transmitted herpesvirus papio-2 infection in a black and white Colobus monkey. Journal of the American Veterinary Medical Association 231:1878–1883. Tsai, P. P., and H. Hackbarth. 1999. Environmental enrichment in mice; is it suitable for every experiment? Abstract. Abstracts of Scientific Papers, ICLAS, FELASA, 112. Turner, J. G., C. A. Bauer, and L. P. Rybak. 2007. Noise in animal facilities: Why it matters. Journal of the American Association for Laboratory Animal Science 46:10–13. Van der Logt, J. T. 1991. Necessity of a more standardized virological characterization of rodents for aging studies. Neurobiology of Aging 12:669–672. Van der Meer, E., P. L. Van Loo, and V. Baumans. 2004. Short-term effects of a disturbed light–dark cycle and environmental enrichment on aggression and stress-related parameters in male mice. Laboratory Animals 38:376–383.
84550.indb 366
11/11/10 6:26:40 PM
Impact of the Biotic and Abiotic Environment on Animal Experiments
367
Vanderschuren, L. J., R. J. Niesink, B. M. Spruijt, and J. M. Van Ree. 1995. Influence of environmental factors on social play behavior of juvenile rats. Physiology & Behavior 58:119–123. Van der Werf, N., F. G. Kroese, J. Rozing, and J. L. Hillebrands. 2007. Viral infections as potential triggers of type 1 diabetes. Diabetes/Metabolism Research and Reviews 23:169–183. Van de Weerd, H. A., V. Baumans, J. M. Koolhaas, and L. F. van Zutphen. 1994. Strain-specific behavioral response to environmental enrichment in the mouse. Journal of Experimental Animal Science 36:117–127. Van de Weerd, H. A., F. A. R. van den Broek, and F. Baumans. 1996. Preference for different types of flooring in two rat strains. Applied Animal Behavior Science 46:251–261. Van de Weerd, H. A., E. L. Aarsen, A. Mulder, C. L. Kruitwagen, C. F. Hendriksen, and V. Baumans. 2002. Effects of environmental enrichment for mice: Variation in experimental results. Journal of Applied Animal Welfare Science 5:87–109. Ventura, J., and M. Domaradzki. 1967. Role of Mycoplasma infection in the development of experimental bronchiectasis in the rat. Journal of Pathology and Bacteriology 93:342–348. Verdonck, E., C. J. Pfau, M. D. Gonzalez, P. L. Masson, and J. P. Coutelier. 1994. Influence of viral infection on anti-erythrocyte autoantibody response after immunization of mice with rat red blood cells. Autoimmunity 17:73–81. Vesell, E. S. 1967. Induction of drug-metabolizing enzymes in liver microsomes of mice and rats by softwood bedding. Science (New York) 157:1057–1058. Vesell, E. S., C. M. Lang, W. J. White, G. T. Passananti, and S. L. Tripp. 1973. Hepatic drug metabolism in rats: Impairment in a dirty environment. Science (New York) 179:896–897. Vlahakis, G. 1977. Possible carcinogenic effects of cedar shavings in bedding of C3h-avy fb mice. Journal of the National Cancer Institute 58:149–150. Voipio, H. M. 1997. How do rats react to sound? Scandinavian Journal of Laboratory Animal Science 24:1–80. Voipio, H-M, A-M Maatta, H. Honkanen, R. Haapakoski, M. Heikkila, K. Mauranen, S. Mering, and T. Nevalainen. 2008. Cage material and food hopper as determinants in rat preference tests. Scandinavian Journal of Laboratory Animal Science 35:69–77. Voipio, H. M., T. Nevalainen, P. Halonen, M. Hakumaki, and E. Bjork. 2006. Role of cage material, working style and hearing sensitivity in perception of animal care noise. Laboratory Animals 40:400–409. Wagner, M. 1988. The effect of infection with the pinworm (Syphacia muris) on rat growth. Laboratory Animal Science 38:476–478. Walters, M. L., N. F. Stanley, R. L. Dawkins, and M. P. Alpers. 1973. Immunological assessment of mice with chronic jaundice and runting induced by reovirus 3. British Journal of Experimental Pathology 54:329–345. Walters, M. N. I., R. A. Joske, P. J. Leak, and N. F. Stanley. 1963. Murine infection with retrovirus. I. Pathology of the acute phase. British Journal of Experimental Pathology 44:427. Ward, J. M., M. R. Anver, D. C. Haines, J. M. Melhorn, P. Gorelick, L. Yan, and J. G. Fox. 1996. Inflammatory large bowel disease in immunodeficient mice naturally infected with Helicobacter hepaticus. Laboratory Animal Science 46:15–20. Ward, J. M., C. E. Wobus, L. B. Thackray, C. R. Erexson, L. J. Faucette, G. Belliot, E. L. Barron, S. V. Sosnovtsev, and K. Y. Green. 2006. Pathology of immunodeficient mice with naturally occurring murine norovirus infection. Toxicologic Pathology 34:708–715. Webster, J. P. 1994. The effect of Toxoplasma gondii and other parasites on activity levels in wild and hybrid Rattus norvegicus. Parasitology 109:583–589. Weichbrod, R. H., C. F. Cisar, J. G. Miller, R. C. Simmonds, A. P. Alvares, and T. H. Ueng. 1988. Effects of cage beddings on microsomal oxidative enzymes in rat liver. Laboratory Animal Science 38:296–298. Whary, M. T., J. H. Cline, A. E. King, C. A. Corcoran, S. Xu, and J. G. Fox. 2000. Containment of Helicobacter hepaticus by use of husbandry practices. Comparative Medicine 50:78–81. Wilberz, S., H. J. Partke, F. Dagnaes-Hansen, and L. Herberg. 1991. Persistent MHV (mouse hepatitis virus) infection reduces the incidence of diabetes mellitus in nonobese diabetic mice. Diabetologia 34:2–5. Williams, R. A., A. G. Howard, and T. P. Williams. 1985. Retinal damage in pigmented and albino rats exposed to low levels of cyclic light following a single mydriatic treatment. Current Eye Research 4:97–102.
84550.indb 367
11/11/10 6:26:40 PM
368
Handbook of Laboratory Animal Science
Willott, J. F., L. C. Erway, J. R. Archer, and D. E. Harrison. 1995. Genetics of age-related hearing loss in mice. II. Strain differences and effects of caloric restriction on cochlear pathology and evoked response thresholds. Hearing Research 88:143–155. Yamamoto, H., H. Sato, K. Yagami, J. Arikawa, M. Furuya, T. Kurosawa, K. Mannen, et al. 2001. Microbiological contamination in genetically modified animals and proposals for a microbiological test standard for national universities in Japan. Experimental Animals 50:397–407. Yamauchi, C., S. Fujita, T. Obara, and T. Ueda. 1981. Effects of room temperature on reproduction, body and organ weights, food and water intake, and hematology in rats. Laboratory Animal Science 31:251–258. Yoon, J. W., and H. S. Jun. 2006. Viruses cause type 1 diabetes in animals. Annals of the New York Academy of Sciences 1079:138–146. Yunis, E. J., and M. Salazar. 1993. Genetics of life span in mice. Genetica 91:211–223. Zheng, Q. Y., K. R. Johnson, and L. C. Erway. 1999. Assessment of hearing in 80 inbred strains of mice by ABR threshold analyses. Hearing Research 130:94–107. Zondek, B., and I. Tamari. 1964. Effect of audiogenic stimulation on genital function and reproduction. iii. Infertility induced by auditory stimuli prior to mating. Acta Endocrinologica 45 (Suppl. 90): 227–233.
84550.indb 368
11/11/10 6:26:40 PM
Chapter 13
Experimental Design and Statistical Analysis
Michael F. W. Festing Contents Introduction..................................................................................................................................... 370 The Need for Improved Experimental Design............................................................................... 371 Basic Statistical Considerations...................................................................................................... 371 First Steps: Ethical Considerations............................................................................................ 371 Observational and Epidemiologic Studies................................................................................. 372 Randomized Controlled Experiments........................................................................................ 372 Basic Principles of Statistical Inference.................................................................................... 373 Types of Data............................................................................................................................. 374 Data Distributions................................................................................................................. 375 Computer Software.................................................................................................................... 375 Laboratory Animals as Models of Humans or Other Species................................................... 376 Experimental Design...................................................................................................................... 377 The Experimental Unit.............................................................................................................. 377 Formal Experimental Designs................................................................................................... 377 Factorial Designs or Treatment Arrangements.......................................................................... 378 A Well-Designed Experiment.................................................................................................... 378 Avoid Bias............................................................................................................................. 378 Have High Power................................................................................................................... 379 Have a Wide Range of Applicability: The Factorial Experimental Design.......................... 380 Be Simple.............................................................................................................................. 380 Be Capable of Being Analyzed Statistically......................................................................... 380 Randomization........................................................................................................................... 380 Sample Size................................................................................................................................ 381 Power Analysis...................................................................................................................... 381 The Resource Equation Method............................................................................................ 382 Formal Experimental Designs................................................................................................... 384 The Completely Randomized Design................................................................................... 384 The Randomized Block Design............................................................................................ 385 Within-Subjects Designs....................................................................................................... 386 Other Designs........................................................................................................................ 387 Statistical Analysis.......................................................................................................................... 388 Data Screening........................................................................................................................... 388 369
84550.indb 369
11/11/10 6:26:40 PM
370
Handbook of Laboratory Animal Science
The Analysis of Variance........................................................................................................... 388 Transformations......................................................................................................................... 391 Post hoc Comparisons and Orthogonal Contrasts..................................................................... 391 Two- or Three-Way ANOVA..................................................................................................... 392 Multiple Dependent Variables.................................................................................................... 395 Linear Regression and Correlation............................................................................................ 395 Presentation of Data........................................................................................................................ 397 References....................................................................................................................................... 398 Introduction It is not easy to design a perfect animal experiment (Festing et al. 2002). It needs to be just the right size: too small and it may not be able to detect important biological effects; too large and scientific resources and animals will be wasted. The appropriate size depends on the variability of the experimental subjects, so ways of reducing variability need to be considered. Heterogeneity will reduce the ability of the experiment to reveal any differences in treatment effects. Decisions need to be made about the choice of animals, including species, strain, and sex. If the animals are being used as models of some aspect of human biology, then whether or not they represent a “good” model needs to be considered. Diet, housing, and other environmental factors need to be specified. Treatments must be chosen, which may involve deciding on suitable dose levels, routes of administration, and time factors. This often involves a certain amount of guess work. If it is important to know whether males and females or different strains respond in the same way, then the experiment may need to include both sexes and/or several strains. The characters to be measured must be chosen and the importance of measurement errors taken into account. Finally, the methods used for statistical analysis of the resulting data must be considered in order to maximize the effective use of all of the information available to the scientist. In practice, experiments can range from those that are deemed very good and properly address the research questions economically and efficiently to those that provide the correct answers, but are quite inefficient, to those so poorly designed that they lead to erroneous conclusions and waste resources. On ethical grounds, it is difficult to justify experiments that are poorly designed if the mistakes have arisen from poor understanding of statistical principles by the investigator. Fortunately, many experiments do give the correct results, if only because the effects of many treatments are so dramatic that they are obvious even with a badly designed experiment. However, when the effects are more subtle, there is a danger that real biological differences will be obscured by “noise” or uncontrolled variation in the study, which can lead to false negative findings (known as type II errors). In other cases, experiment noise can act in such a way as to produce false positive results (or type I errors). Usually, this is caused by bias arising when treated and control animals differ in other ways, such as having different environments or being of different ages. Both types of errors can be minimized through proper planning in the design and analysis phase of the research. The aim of this chapter is to give a brief introduction to the principles of experimental design and the statistical analysis of animal experiments in hope of guiding research workers in planning and executing their own simpler studies and to facilitate their communications with statisticians for more advanced and complicated ones. It is a chapter about rather than a description of experimental design and statistics. The concepts are illustrated with numerical examples of real data that have been analyzed using the MINITAB statistical package (version 13, MINITAB Inc., State College, PA), though other statistical software would have given very similar results. The chapter is not a substitute for a textbook, such as one of those listed in the references. Software manuals are also a useful source of information. Readers are encouraged to work with persons knowledgeable in this area whenever necessary to ensure proper use of the methods presented here.
84550.indb 370
11/11/10 6:26:40 PM
Experimental Design and Statistical Analysis
371
The Need for Improved Experimental Design There are good reasons for concern about misuse of statistical methods within the laboratory animal science literature (Festing 1992, 1994). A survey of papers published in two academic toxicology journals found that roughly one-third of the 48 experiments described were about twice as large as they needed to be; none of them used blocking, a technique for improving precision; and more than 60% used incorrect statistical methods, such as failure to analyze a factorial design (see later discussion) using a two-way analysis of variance (ANOVA; described later) (Festing 1996). A similar survey of papers published in a veterinary journal, commissioned by its editors, found mistakes that included failure to use randomization, potential bias, inappropriate statistical methods, and a few cases where the conclusions were not supported by the data (McCance 1995). These problems are not unique to animal experiments. Surveys of published papers in human medical journals also show that there is room for improvement (Altman 1982a, 1982b), with some authors suggesting that the misuse of statistics can be viewed as a breach of ethical principles in science. More recently, a survey of 271 papers selected at random from academic institutions in the UK and United States in which mice, rats, or nonhuman primates were used (Kilkenny et al. 2009) found serious defects in the reporting of animal research, with evidence that the research was not always well conducted. The papers involved live animals, which might suffer pain, distress, or lasting harm. Among this sample, 5% did not clearly state the purpose of the study, making it much more difficult for the reader to interpret the results; 6% did not indicate how many separate experiments were being reported, making it difficult to reconcile the materials and methods with the result section; and 13% did not identify the experimental unit, which is the unit of randomization and of the statistical analysis. Some basic information, such as the sex and age or weight of the animals, was not given in 26 and 24% of the papers, respectively. Four percent failed to state the number of animals used and none of them attempted to justify the sample sizes used. In 35% of those that reported the numbers used, these differed in the materials and methods and the results sections, implying some inaccuracies in the handling of the numerical data. Two recent studies involving meta-analysis of animal experiments have also commented on the poor quality of the reporting and of the conduct of some of the studies. In a meta-analysis of 44 controlled experiments involving fluid resuscitation (Roberts et al. 2002), only two said that the animals had been allocated to the treatments using randomization. None of them was sufficiently large to be able to detect reliably a halving of the risk of death. This would be a clinically important end point. There was substantial heterogeneity of the results, so it was impossible to calculate an odds ratio. The authors questioned whether these studies made any contribution to human medicine. This was followed by another paper (Perel et al. 2007) that identified six interventions that had a known outcome in humans and then did meta-analyses of all the animal papers that they could find to see whether these predicted the human outcome. Very briefly, in three of the interventions there was evidence that the animal studies predicted accurately the response in humans, but the other three failed to do so. However, the authors were unable to determine whether this was because the animals were a bad model of humans or because the animal studies had been badly done. Basic Statistical Considerations First Steps: Ethical Considerations The first step in designing an investigation is to specify clearly the objective of the study and consider whether it can be achieved by means other than an animal experiment. The “three Rs,” introduced by Russell and Burch (1959), provide a framework for assessing each experiment. When possible, the use of animals should be replaced by less or nonsentient alternatives such as insects or cell
84550.indb 371
11/11/10 6:26:41 PM
372
Handbook of Laboratory Animal Science
or organ culture. If that is not possible, then experiments should be refined in order to minimize pain, distress, or lasting harm. There is now a substantial literature on methods of improving the welfare of experimental animals (Latham 2010). These are discussed in detail elsewhere in this book. Finally, if there is no alternative to the use of animals, then the numbers used should be reduced to ensure that the minimum numbers of animals are used, consistent with achieving the objectives of the study. The use of too many animals clearly leads to a waste of animals and other scientific resources, while the use of too few animals may lead to a failure to identify important biological or clinical effects. Generally, the numbers of animals can be reduced by careful control of interindividual variation, the choice of an appropriate experimental design, and use of an objective method of determining sample size. It may also be possible to use more sensitive material. For example, animals of some genotypes may be more responsive to a particular intervention than others. If the animals are to be used in a series of experiments it may be worthwhile testing a few strains to find the one which is most suitable for the project. Observational and Epidemiologic Studies This chapter is primarily concerned with randomized controlled experiments, but data used to explore a scientific hypothesis can be acquired in several different ways. Population-based investigations involving surveys, cross-sectional studies, case-control studies, and cohort studies are widely used for epidemiologic investigations. These are reviewed elsewhere (Weigler 2001). Epidemiologic work also includes methods to obtain unbiased measures of disease incidence and prevalence that are valuable for comparing trends and the impact of prevention efforts. Whether prospective or retrospective, epidemiologic studies can be designed to explore exposure–disease relationships in situations where formal controlled experiments are impossible, unethical, or impractical, such as in assessing whether smoking causes lung cancer in human beings. Hypothesis-based observational investigations can shed light onto possible risk factors for disease and thereby aid in the development of intervention strategies. Relative risks, odds ratios, and various other measures of association are used to quantify the importance of study factors on disease development in groups of individuals or animals. The study factors can include natural events and exposures that are not controlled by the researcher and can address agent variability, dose, route, host genetics, and important covariables more typical of the spectrum of real-world conditions. Randomized Controlled Experiments In contrast to purely observational studies, randomized, controlled animal experiments are used to investigate possible causal relationships between two or more treatments and one or more observed responses, with the scientist in control of many sources of variation in the experimental subjects. There are three main types of experiments: • Pilot studies are small experiments, the main aim of which is to test the logistics and feasibility of some larger study and, in some cases, to obtain preliminary information, such as whether dose levels seem appropriate. This information can be used in planning the larger study. Generally, the results of such studies will not be published. • Exploratory experiments are used to generate new hypotheses, usually by studying patterns of response to some treatment or other intervention. Typically, such studies will involve looking at many outcomes. Statistical tests of those that look “interesting” may be seriously biased as a result of chance interindividual variation, leading to false positive results. Thus, the results of an exploratory experiment need to be tested using a confirmatory experiment. • Confirmatory experiments are used to test a preferably simple hypothesis that is clearly stated before starting the experiment. If the experiment is well designed and the interindividual variation is well controlled, then the number of false negative and false positive results should also be minimized.
84550.indb 372
11/11/10 6:26:41 PM
Experimental Design and Statistical Analysis
373
The most important features of the controlled experiment are that there will be one or more independent variables or treatments, such as the dose of a compound (often including a zero dose, designated the control), which can be manipulated by the experimenter, and one or more dependent or response variables, which can be measured, counted, or otherwise assessed to determine whether they are altered by the treatment. The experiment must be replicated using several animals (or other subjects) in order to assess the extent of the inter-individuals variation. This information is needed in order to be able to make appropriate statistical tests and set confidence limits. Experimental subjects must be assigned to the treatment groups at random in order to minimize the chance of some systematic influence affecting one or more groups because this would lead to biased results. Randomization should extend to the housing of the animals and assessment of outcomes during the course of the experiment if bias is to be avoided. Basic Principles of Statistical Inference Once the data from an experiment have been collected, any difference between the means (or some other parameters, such as the medians) has to be interpreted. This involves one or more formal statistical tests. This is easiest to explain in the situation where there are just two groups and some character, such as body weight, is measured. Table€13.1 shows the result of a hypothetical experiment to compare the effect of two diets on body weight in mice. The mean body weights of the two groups were 40.64 and 44.18 g, a difference of 3.54 g. Within each group, individuals varied in weight from 34 to 46 g in the control group and from 41 to 51 g in the treated group. The variation is quantified in the sample standard deviations (SDs) of 4.50 and 2.96 g, respectively. Some or all of the differences in mean body weight between the two groups could be due to chance sampling variation and some could be due to the effect of the treatment. The estimates of the SDs are also subject to chance variation. Knowing the variation within each group and assuming that it will be the same between groups because animals were initially assigned to groups at random, it is possible to work out the probability that the difference between the two groups (3.54 g in this case) or a greater difference could have arisen by chance. A statistical test is used to estimate this probability. If it is very low, then it is assumed that the difference must be due to the effect of the treatment. Table€13.1╅Body Weights of Control and Treated Mice Control 34 46 35 42 42 42 44 43 39 34 46
Treated 42 42 51 48 44 43 41 45 42 44 44
Notes: Mean: 40.64; 44.18 Difference: 3.54 Std. Dev.: 4.5; 2.96 Pooled Std. Dev.: 3.73 Signal/noise ratio: 0.95 95% CI for difference between means (–6.93; –0.16) t-Value = –2.18; p-value = 0.041; DF = 20
84550.indb 373
11/11/10 6:26:41 PM
374
Handbook of Laboratory Animal Science
Several different statistical tests are available, some of which are discussed later. One common test is Student’s t-test. If this test is used, then the p-value, or probability, of such a difference (or a greater one) occurring by chance is estimated to be p = 0.041. Usually, if this probability is less than 5%, then the differences are said to be statistically significant at the 5% level. The test is often cast in terms of a null hypothesis, which states that there are no differences between the means (or other parameters) being tested, and an alternative hypothesis, which states that such differences exist. In this particular case, the null hypothesis would be rejected at the 5% level of probability because the p-value is less than 0.05, or 5%. Therefore, the alternative hypothesis—that there is a difference between the means—would be accepted. In many cases, it is important to know the size of the difference, particularly in a clinical experiment. The best estimate is clearly the difference between the two means, but it may be greater or smaller depending on sampling variation. The uncertainty can be quantified by estimating the 95% confidence interval for the difference, which in this case is –6.93 to –0.16 (the treated mean is subtracted from the control mean, resulting in negative values in this case). If there were no significant difference between the two groups, then this confidence interval would span zero. Therefore, although the estimated mean difference is –3.54 g, it would not be entirely surprising if it were as low as –0.16 g or as high as –6.93 g. One aim in planning an experiment is to have a high signal/noise ratio. With low noise, the signal is easier to detect. The signal is the response (in this case, 3.54 g) and the noise is the standard deviation. In this case, it is the average of the SDs in each group or (4.50 + 2.96)/2 = 3.73. Thus, the signal/noise ratio in this case is 0.95. This ratio can be used in planning future experiments because it is directly related to the required sample size, as described later. This example involves only two groups, and different tests are used with more complex designs. But very often an experiment will result in estimates of means, standard deviations, and p-values assessing the probability that various comparisons are likely to have arisen by chance. Of course, this does imply that in about 5% of all comparisons, the null hypothesis will be rejected when in fact it is true (i.e., there will be a type I error, or a false positive result). False negative results arise if the experimental material is excessively variable (high noise) or if the sample size for the experiment is too small, so that quite large differences between group responses could have occurred by chance. The power of an experiment is equal to one minus the type II error rate and is usually expressed as a percentage. Clearly, it is important to devise powerful experiments capable of detecting treatment effects of magnitudes likely to be of biological or clinical relevance. This is discussed in more detail when considering sample sizes later. Statistical significance and biological significance are not always identical. Differences can be statistically significant because the interindividual variation is very well controlled, but so small that they are of little biological interest. When statistical differences are found between treatment groups, it is important to look at their magnitude. Large differences are likely to be of more biological importance than small ones, although both may be statistically significant. Types of Data There are two major classifications of data: • Categorical data, such as dead/alive, male/female, can usually be summarized by counts, proportions and percentages. When there are just two groups, this is known as binary data. There can, of course, be more than two categories such as “small,” “medium,” and “large,” or 0, +, ++, and +++. This might be summarized as the counts or proportions in each category. • Measurement data (also known as interval or continuous type data) will usually be summarized using means or medians, sometimes with some estimate of the variation among subjects.
84550.indb 374
11/11/10 6:26:41 PM
Experimental Design and Statistical Analysis
375
Data Distributions In order to calculate the required probabilities for a statistical test, it is also necessary to know something about the distribution of the data for the outcome variable or variables of interest. Two of the most important distributions with categorical data are the Poisson and the binomial distributions. The Poisson distribution may be seen with counts in which the mean number counted in each sample is low—say, less than about five—but there is no upper limit. The classical example is the counts of cells in a hemocytometer. Each square might have none, one, two, three, etc. cells. If such counts are collected across many squares, then the number of squares with zero, one, two, three, etc. cells per square may follow a Poisson distribution. Data with a Poisson distribution can often be analyzed using parametric tests, such as the t-test or analysis of variance, if the data are first transformed to a different scale. With the Poisson distribution, if each observation is replaced by its square root, then the resulting distribution should approximately resemble the normal bell-shaped distribution (see later discussion). If the counts are much higher—say, with a mean of 20 or more—then, although the data are still discrete, the distribution will often resemble that of a normal or Gaussian distribution (discussed later). In contrast, the outcome data may be the proportion of animals in a test group showing some adverse effect. In this case, the data may follow a binomial distribution where, like all proportions, the outcome is bounded by zero and one. With measurement data, the most common distribution encountered is the normal distribution, which has a bell-shaped curve that is symmetrical about the mean. However, other distributions are commonly encountered, and scatter plots of the raw data can be used to explore this feature during the analysis phase of experiments. Most biological characteristics are not actually normally distributed; however, according to the central limit theorem, the mean values of samples drawn from populations of any form will tend toward normality, if the sample size is sufficiently large. Moreover, it is not the data, as such, that need to have a normal distribution for many types of statistical analysis, but rather the residuals or deviations from the group means that must be approximately normally distributed. This is reasonably common or can be made so by an appropriate data transformation, accounting for the widespread use of statistical methods based on the assumption that the data have a normal distribution. When the data are concentrations of some substance, say in body fluids, the underlying distribution may instead be log normal, which is asymmetrical with a long tail to the right due to a small number of very high values. This can be converted to a normal distribution by taking the logarithms of the individual values prior to a statistical analysis. Computer Software The calculations necessary to work out the means, standard deviations, and desired probabilities and associated statistical information are tedious if there is a lot of data. Fortunately, a wide range of statistical software is available for doing the calculations. Readers are strongly urged to use a dedicated statistical software package such as MINITAB®, SAS®, SPSS®, STATISTICA®, StatsDirect®, Stata®, Genstat®, GLIM®, or R, rather than a spreadsheet. A Web search for “statistical software” will give the Web sites of numerous such packages, in some cases with reviews. The R package is of particular interest because it is free and widely used by professional statisticians. However, it is driven by command, rather than menu, which means that it is more difficult to learn than a menu-driven package. It would be suitable as part of a formal statistics course using textbooks by P. Dalgaard (2003) or M. J. Crawley (2005). A menu-driven front end called “R Commander” is also available. Though useful for many purposes, spreadsheet software does not have the complete range of statistical techniques needed for most analysts and its output is often
84550.indb 375
11/11/10 6:26:41 PM
376
Handbook of Laboratory Animal Science
nonstandard. Spreadsheets also do not have the range of graphical output necessary to describe more complex data sets. It would be extremely tedious, for example, to produce many of the graphs in this chapter using a spreadsheet program. Laboratory Animals as Models of Humans or Other Species Laboratory animals are often used to model human responses or as models of some human disease. Models used in biomedical research have five key features:
84550.indb 376
1. There can be substantial asymmetry in the number of similarities and differences between the model and the target. In theory, the model and the target only need to have a single feature in common, but there can be any number of differences. This means that useful models can sometimes be highly abstract, such as mathematical equations or computer simulations. 2. There must be some differences between the model and the target; otherwise, the model would not be a model. Some of the differences are extremely important because they allow us to do things with the model that would not be possible with a human. Mice and rats are widely used because they are like humans in anatomy, biochemistry, immunology, and genetics but unlike humans in being small and prolific, and we can dose them with chemicals and sacrifice them when necessary. However, their small size makes them unsuitable for other applications, such as heart surgery. Toxicologists sometimes claim that because humans are genetically variable, it is necessary to use genetically variable, outbred stocks, rather than inbred strains in toxicological screening. However, a good toxicological model should be one that has the best chance of detecting the effects of a toxic chemical. Toxicologists already use bacteria and cell mutation assays as models of humans, so they should recognize that good models can often be quite different from their target. The question they should be asking themselves is whether inbred strains or outbred stocks are more capable of distinguishing between toxic and nontoxic substances and which type of stock will make it easiest to explore mechanisms. Inbred strains are more uniform in their responses and have many other useful properties, so they will usually be more sensitive than outbred stocks in detecting toxicity. However, whether using inbred strains or outbred stocks, it is necessary to use more than one strain because strains can differ substantially in their response to toxic agents (Kacew and Festing 1996; Festing 1999; Festing, Diamonti, and Turton 2001). This can be done without increasing total numbers and has the added advantage that differences between strains indicate genetic variation in response. One of the main criticisms of the use of outbred stocks in current toxicity tests is that genetic variation can only be detected if related animals respond similarly. But in practice, relationships among a group of outbred rats or mice are never known, so genetic variation in response is never detected in current toxicity tests, even though it is extremely important in humans. 3. Models are highly specific to a particular study. Strains of mice and rats that develop cancer, heart disease, diabetes, or neurological diseases could be of interest in the study of these diseases, but would probably be unsuitable for regulatory toxicology, where long-lived strains are usually required. Thus, it is impossible to say whether the rat is a good or bad model of humans without specifying the context of the proposed study. 4. Models need to be validated. Research using animal models usually aims to predict a response in humans. When a new treatment for a particular disease or condition is developed in animals, clinical trials will normally show whether or not the model was valid. One exception is in testing industrial and environmental chemicals for their toxicity, where deliberate exposure of humans is not ethically acceptable. 5. Models are subject to improvement through further research. A lot of animal research is aimed at understanding the animal as a potential model for particular human conditions, without being too precise as to what those conditions might be. Models are not just found; they need to be developed. This requires an understanding of the biology of the species and the effects of various interventions, such as inactivating specific genes or manipulating the environment. As our understanding increases, so does the chance of choosing the most appropriate models for a specific disease.
11/11/10 6:26:42 PM
Experimental Design and Statistical Analysis
377
Experimental Design The Experimental Unit The experimental unit is the entity that can be assigned at random to one of the treatments, independently of all other experimental units. It is also the unit of statistical analysis. Any two experimental units must be able to be assigned to different treatments. Incorrect identification of the experimental unit is a common error. It leads to an incorrect estimate of the number of experimental subjects and incorrect statistical tests. Some examples of experimental units include: • Mice or rats are individually or group housed, but individually treated—for instance, by injection or by gavage. In this case, the animal is the experimental unit. There may be cage differences, but these can often be regarded as a “blocking factor” that is removed in the statistical analysis (see later discussion of randomized block designs). • The effects of several anesthetics are studied sequentially in a single animal, with a rest period between tests. In this case, the experimental unit is the animal for a period of time because the animal can have different anesthetics in different periods. Several animals will normally need to be used, but the “animal effect” can be removed using a randomized block type of statistical analysis. • In a teratogenesis experiment, the pregnant female is assigned to a treatment group, but all the measurements are done on the fetuses. The pregnant female is the experimental unit. Some form of aggregate score for the measurements of the pup will be the metric that is used in the statistical analysis. • If there are three mice in a cage and the treatment is given in the diet or water, then the cage of mice is the experimental unit. If there are 10 cages of controls and 10 cages of treated animals, then “n” is 10, not 30. • If the animal has its back shaved and different compounds are applied topically to different patches of skin chosen at random, with some measurement being made on each patch during the experiment, then the patch of skin is the experimental unit. Differences between the animals can be removed using a randomized block statistical analysis.
Formal Experimental Designs Experiments can range in size and complexity from the use of a single animal in an uncontrolled experiment (where no statistical analysis is possible) to large, formal, controlled experiments involving hundreds of animals. A controlled experiment is one in which different treatment groups are compared, with the treatments decided by the experimenter. In many cases, one group is called the “control” group, although this is not always the case. Virtually all controlled experiments should conform to one of the standard types of design, such as “completely randomized,” “randomized block,” “Latin square,” “split plot,” “repeated measures,” “sequential,” “factorial,” or other, more advanced designs (Festing et al. 2002; Crawley 2005; Ruxton and Cograve 2006). Some of these are considered separately later. The main aim of these designs is to minimize interindividual variation so as to make the experiment more powerful (i.e., capable of detecting any real differences between treatments). Factorial designs are used to increase the generality of the results. For example, a factorial design might compare two treatments on both males and females. This will show whether responses are the same in both sexes (see later discussion) Randomized block and Latin square designs can be used to account for a mini-natural structure among the experimental units or to split up the whole experiment into smaller mini-experiments that are subsequently combined. For example, studies of neonatal animals need to account for the fact that they come in litters. A lack of availability of the required number of animals all at one time or some logistical bottleneck in making the observations can be taken into account using randomized block designs.
84550.indb 377
11/11/10 6:26:42 PM
378
Handbook of Laboratory Animal Science
Factorial Designs or Treatment Arrangements Superimposed on these design options, there may be a specific structure to the treatments. For example, a single-factor design will vary one independent variable, such as the administration of compounds A, B, C, or D, or various levels of one of these. In contrast, a factorial arrangement of treatments might vary two independent variables at a time. For example, if the four compounds are to be given to both males and females, this will be a 4 (compounds) × 2 (sexes) factorial design with eight treatments. It is analyzed as a single experiment. Factorial designs are often highly efficient because they can produce extra information at little or no additional cost. According to R. A. Fisher (1960): If the investigatorâ•›…â•›confines his attention to any single factor we may infer either that he is the unfortunate victim of a doctrinaire theory as to how experimentation should proceed, or that the time, material or equipment at his disposal is too limited to allow him to give attention to more than one aspect of his problem. Indeed in a wide class of cases (by using factorial designs) an experimental investigation, at the same time as it is made more comprehensive, may also be made more efficient if by more efficient we mean that more knowledge and a higher degree of precision are obtainable by the same number of observations.
A Well-Designed Experiment The aim of the experiment needs to be clearly specified and it needs to be designed to take the following into account. Avoid Bias Bias (systematic error) can arise, for example, when subjects in different treatment groups have different environments or when they are otherwise not comparable. For example, if the treated animals are older, heavier, or housed in a different environment from the control animals, the results may be biased because these effects could be mistaken for the effects of the treatment. Bias can be minimized, first, by ensuring that the experimental subjects are allocated at random to the different treatment groups and in all subsequent housing and related manipulations. It would not be sufficient to allocate the animals to the treatments at random and then process all the controls on the first day and all those in the treated group on the next day, because environmental factors may influence results differently on different days. Investigators may become more skilled over time, for example. If there is any subjective element in collecting the data, this should be done “blind,” using coded samples, so that the person collecting the data does not know the treatment group of an individual subject. With clinical trials, double blinding should be used whenever possible, with neither the patient nor the doctor knowing the group to which the patient belongs. Some variables cannot be randomized, potentially leading to bias. For example, the strain and sex of an animal are determined at conception and cannot be assigned at random. This means that if an experiment involves comparing two or more strains or sexes, the investigator must ensure that they are, as far as possible, comparable in all other ways. For example, if the animals also differed in age or source or had been subjected to different environments, then any differences between the strains could be due to genetics or to one of these factors, or both. In such cases, the factors “genotype” and “age” are said to be “confounded” and cannot be separated. When bias is known or suspected after a study has been completed, the results should include discussions of the implications. There is good evidence that a failure to randomize and/or use blinding can lead to seriously biased results. For example, in a review involving 290 papers, those that did
84550.indb 378
11/11/10 6:26:42 PM
Experimental Design and Statistical Analysis
379
130 120 110 100 Sample size
90 80 70 60 50 40 30 20 10 0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
Signal to noise ratio Figure 13.1â•…Relationship between the signal/noise ratio (effect size/standard deviation) and required number per group for a two-sided Student’s t-test assuming a 5% significance level and an 80 (triangles) or 90% (circles) power. Note that large group sizes are needed to detect a signal/noise ratio of less than about 0.5.
not use randomization, blinding, or both were more than five times more likely to claim a positive result (Bebarta et al. 2003). Have High Power The power of an experiment is the probability of being able to detect a specified treatment effect given certain other assumptions (discussed later when considering sample size). A powerful experiment will have a high signal/noise ratio for any given sample size. The signal is the response to the treatment (e.g., difference between treatment means) and the noise is the variation among individuals, quantified as the standard deviation. As already noted, the signal/noise ratio in Table€13.1 is 0.95. The higher the signal/noise ratio is, the smaller will be the required sample size. The relationship between the signal/noise ratio and sample size is shown in Figure€13.1. For example, if the signal/noise ratio is 0.5, with a power of 90%, a total of 86 animals would be required in each group. If the signal/noise ratio is 1.0, again with 90% power, only 23 animals would be required per group and only 17 animals would be needed per group with power reduced to 80%. The relationships among these variables are discussed in more detail in the section on sample size determination. High power can be achieved by using uniform material, such as isogenic strains of mice or rats that are free of disease, of uniform weight or age, the same sex, and housed under optimum conditions. If measurement error is likely to be large, it might be possible to reduce it by using multiple determinations of outcomes. For example, enzyme activity could be measured in duplicate samples, or open field activity could be recorded for a longer period or on more than one occasion in each animal. Other sources of variation can often be controlled by using a randomized block or Latin square design. If within-subjects designs, such as crossover designs, are possible, they will often lead to a useful increase in the power of the experiment. Increasing the signal can sometimes be accomplished by using higher dose levels or by choosing a more sensitive strain of animal. The problem with the former is that it may increase suffering,
84550.indb 379
11/11/10 6:26:42 PM
380
Handbook of Laboratory Animal Science
while choosing a more sensitive strain may not be possible if strains have not already been characterized for sensitivity. However, using small numbers of several strains will average out any strain differences and should minimize the chance of using a resistant strain (see later discussion). Power can also be increased by increasing the sample size (see later discussion). However, that will cost time, money, and animals, so it is best to concentrate on the other things first and then use the minimum number of subjects consistent with achieving the scientific objectives of the study. Have a Wide Range of Applicability: The Factorial Experimental Design A large number of variables need to be controlled by the scientist in order to do a sensible experiment. These include age, sex, strain, diet, caging, physical and social environment, and methods of measuring the characters of interest. Every experiment is done using a particular set of these fixed effects, and biomedical research would be impossible if the results could never be generalized to a broader range of conditions. But it may be important to know the extent to which an observed result is dependent on any of these variables. Thus, if the ambient temperature in the animal facility where the experiment was performed was maintained at 23°C, it would usually be reasonable to assume that similar results would have been obtained if the ambient temperature had been maintained at 21°C instead (plus or minus the same daily variation). On the other hand, in some circumstances, it really is important to find out whether males and females or strain X and strain Y animals differ in their response. In this circumstance, a factorial design can be used. In many cases, both sexes can be used without any increase in the total number of animals required. Be Simple Clearly, an experiment should not be so complex that mistakes are made in its execution. Detailed protocols should be prepared showing exactly what needs to be done at each stage of the experiment, and great care needs to be taken to ensure that there are no serious bottlenecks that make it impossible to carry out the assigned tasks in the time provided. A pilot experiment is recommended as a first step if there is any doubt about the logistics of the study. Be Capable of Being Analyzed Statistically Most experiments will need to be analyzed statistically to determine the probability that the observed results could simply have arisen by chance. The method of statistical analysis needs to be planned at the same time as the experiment is planned, and it is often a good idea to generate some simulated results and do a statistical analysis at the planning stage. A scientist should never start an animal experiment without knowing how the data are to be analyzed. Randomization Randomization throughout the whole experiment is fundamental to all controlled studies. Scientists should welcome the need to randomize because it protects them from making false claims about the results of the experiment. Picking animals from a box is not an adequate method of randomization because those caught first may respond differently than those caught later. There are several practical ways of doing the randomization: • Physical randomization can be done by writing numbers or treatment designations on pieces of paper and drawing them out of a bag. For example, if the aim is to randomize 20 rats to four different groups,
84550.indb 380
11/11/10 6:26:43 PM
Experimental Design and Statistical Analysis
381
A–D, then five bits of paper labeled “A,” five labeled “B,” etc. can be placed in a bag. The first rat is caught, and one of the pieces of paper is drawn from the bag to see the group to which it is assigned. This is repeated for each rat, with the pieces of paper not being returned to the bag. This can, of course, be done in the office, with the results being noted on paper, which is then taken to the animal house. • Many computer programs will generate random numbers or place a group of numbers in random order. Thus, in the previous example, the numbers 1–20 (or, alternatively, five ones, five twos, etc.) could be placed in a column of a software package such as MINITAB, and a command used to randomize their order. If numbers are used, then the first five numbers would represent the rats assigned to treatment A, the next five to B, etc. • Spreadsheets usually have a RAND() function. In the preceding example, five As, five Bs, five Cs, and five Ds can be entered in one column and 20 random numbers in the next column. Both columns can then be sorted on the column of random numbers to randomize the letters. • Tables of random numbers can also be used. Their use is described in most statistical texts. In general, these tables can be a bit tedious to use, though undoubtedly they produce good randomizations.
Randomized block designs (see later discussion) are divided into a number of “mini” experiments, and randomization is only done within a block. Typically, if there are four treatments, then block one would consist of four animals matched so as to be as similar as possible. Four bits of paper can be placed in the bag labeled A, B, C, and D. The first rat is caught, and one of the bits of paper is drawn from the bag to assign it to one of the four treatments, and so on. This is repeated for the remaining blocks. Sample Size Experiments should be just the right size to address the specified research questions. Very small experiments will lack power and may be incapable of detecting biologically important responses to the treatments. Very large experiments may waste time, laboratory reagents, and animals. This may be unethical. Unfortunately, there is no single way of determining the most appropriate sample size for a given study design. Though helpful, the two methods that are available—namely, power analysis and the resource equation methods—have inherent limitations. However, both are probably much better than relying entirely on past experience or intuition. All too often, investigators use groups of eight animals when, in many cases, they could use far fewer, provided they do the correct statistical analysis. Power Analysis This method is recommended in particular for large, relatively simple, expensive experiments such as clinical trials, although it can also be used for any other experiment where the requisite information is available. The method depends on a mathematical relationship between the following: • The sample size. Sample size is usually what is being calculated, given knowledge or estimates of the other parameters. However, in some cases, sample size may be fixed, and the aim may be to determine the power of the experiment or the effect size, which could be detected in the proposed experiment. • The effect size (signal) likely to be of biological or clinical interest. In most cases it is important to be able to detect a large difference between means, but a very small difference may be of little scientific interest. The effect size is the dividing line between these two. For example, in an experiment designed to study the effect of a test compound on blood pressure in rats, it might be decided that unless the compound reduces blood pressure by, say, 10 mm of mercury or more, the compound is unlikely to be of much interest. Thus, the effect size would be specified as a 10 mm change in the mean blood pressure between treatment groups. If it is larger, then it will also be detected, and if it is smaller, then it will be of little interest. As already noted, a good experiment will have a high signal/ noise ratio. The effect size is the size of the signal that it would be desirable to be able to detect. In
84550.indb 381
11/11/10 6:26:43 PM
382
•
•
•
•
Handbook of Laboratory Animal Science
experiments where there are multiple dependent variables, such as panels of hematology and blood biochemistry values, it is necessary to decide which of these is most important, and the experiment will need to be designed primarily to detect the change in that variable. Deciding which variable is most important in a study involving gene microarrays, where there may be several thousand dependent variables, is not likely to be easy! The standard deviation of the outcome (character) being measured. The standard deviation (for quantitative characters) must also be estimated as an estimate of the noise. This is a problem because the experiment has not yet been done, so the estimate must come from previous related experiments or from the literature. This presents difficulties for experiments where there is no estimate of the standard deviation. Pilot studies are sometimes suggested as a way of getting estimates of the standard deviation in advance of the main experiment, but such studies are usually very small, reducing the reliability of the estimate of the standard deviation. Unfortunately, small differences in the standard deviation can make a large difference to the estimate of the numbers of subjects required. The desired power (one minus the chance of a type II error) of the experiment. The desired power is usually set somewhat arbitrarily between about 80 and 90%. Higher power requires larger experiments. Power should be set in relation to the consequences of failing to detect an effect (i.e., false negative error). In a toxicological screening experiment, failure to detect an effect could have serious consequences, so power levels should be set high. In some biological experiments, it may be less important if some effects of the treatment are missed. The desired significance level (type I error rate) to be used in the analysis. The significance level is usually set at 5% (0.05) because this is generally accepted as the critical value for statistical significance. But there is nothing special about this, and a different significance level could be chosen. For example, setting the significance level at 10%, rather than 5%, will increase the chance of a false positive, but decrease the chance of a false negative result (i.e., increase the power). The alternative hypothesis (i.e., whether the response could only go in one direction or whether it could go either way). If the alternative hypothesis is that there is a difference between the means (or other parameter being studied), in which case a two-tailed test is used. This is the most usual situation. However, in some cases, the response can only go or is only of interest in one direction. In this case, the alternative hypothesis will be that the mean of the treated group is, say, greater than that of the control group (or vice versa) and a one-tailed test should be used.
Having specified five of the preceding variables, the sixth (e.g., sample size) is then estimated, using appropriate formulae. Fortunately, several modern statistical software packages include power calculation algorithms and several stand-alone packages also exist (Thomas 1997). A number of free Web-based resources also can be used for relatively simple calculations. A Web search for “statistical power calculations” should locate a number of such Web sites. Note that when two proportions are compared, the standard deviation is not required, since this is implicit in the proportion and sample size. Figure€13.1 shows sample size as a function of the signal/noise ratio (effect size divided by the standard deviation) for an 80 or 90% power level for a two-sample Student’s t-test. Note that, with a signal/noise ratio of 1.0, between 17 and 23 animals per treatment group would be required, depending on the desired power. To detect a signal/noise ratio of 0.4 of a standard deviation would require more than 100 animals per group; however, if the signal/noise ratio was as large as two standard deviations, then a sample size of only about eight animals in each group would be needed. The Windows program nQuery Advisor (Version 2.0, Statistical Solutions, Cork, Ireland) can be used for more complex situations. The Resource Equation Method In some situations, it is difficult to apply a power analysis, such as when there is no estimate of the standard deviation, when there are many dependent and/or independent variables, or when it is very difficult to specify an effect size of interest. The resource equation method (Mead 1988) is
84550.indb 382
11/11/10 6:26:43 PM
Experimental Design and Statistical Analysis
383
useful in these situations. It is based on the law of diminishing returns. Increasing the size of a small experiment gives good returns, but increasing sample size when the experiment is already large gives poor returns. Mead (1988) suggests that for experiments producing quantitative data that are to be analyzed by the ANOVA, the error degrees of freedom, E, should be somewhere between 10 and 20. The ANOVA is discussed later; however, for the very simplest design, the following formula can be used:
E = (total number of experimental units) – (total number of treatments)
Suppose the aim is to do a factorial experiment with four dose levels of some test chemical and four strains of rats. How many animals would be needed in each group? This is a 4 × 4 factorial design with 16 treatment means to be compared. It would be difficult to use a power analysis with such a design because this would require specification of an effect size across 16 groups. However, using the resource equation, the first step would be to take a guess at the sample size. Say there were to be five rats per group. This would mean a total of 5 × 16 = 80 rats. Because there are 16 groups, E = 80 – 16 = 64. This would suggest that fewer animals could be used. With two rats per group, E = 16, so this would probably be an acceptable number. Note that factorial designs in which there are many groups can have much smaller group sizes than most investigators are used to. Such designs are normally analyzed using a multiway ANOVA. A 2 × 2 × 2 × 2 (24) factorial design, for example, will have 16 groups; with two experimental units per group, E will be 16. As a hypothetical example, it might be used to study the effect of mouse strain (two strains), age (two ages), presence or absence of environmental enrichment (some or none), and the effect of some drug (some, none) on mouse behavior. The analysis of such an experiment would pose no problems. Each main effect (strain, age, etc.) will be based on eight experimental units, so it is estimated reasonably well. There will be six two-way interactions. These will show, for example, whether the response to the drug depends on environmental enrichment, whether the strains respond differently to the drug, etc. There will be four three-way interactions (e.g., strain × drug × age) and one four-way interaction. Experience shows that statistically significant three- and four-way interactions are usually rare. If they do occur, they are also difficult to interpret. In some cases, they will simply be assumed to be zero and the five degrees of freedom and the sums of squares associated with them will be added to the error term, making E = 21 (in this example). If even more factors are of interest, then special techniques can be used to avoid increasing the numbers of experimental units (Montgomery 1997). Such situations are rare, but if somebody wanted to optimize a PCR reaction, for example, it is easy to see that many variables would need to be taken into account simultaneously. These could best be explored using factorial designs, and the 2n series, in particular. The resource equation method is based on the need, in each experiment, to obtain a good estimate of the standard deviation. Too small an experiment (E < 10) gives a rather poor estimate and too large an experiment (E > 20) wastes resources that could be better employed by adding more treatment groups. However, the method should not be applied too rigidly and the cost of experimental units (both ethically and financially) should be taken into account. Thus, if the experimental unit is a petri dish of cells in an in vitro experiment, it is likely to be very cheap, and it would be acceptable for E to be much higher than 20. In contrast, if the experimental unit is a nonhuman primate or a whole room full of animals (say, in an experiment to compare room environments), then E could be less than 10, but, of course, this would reduce the power of the experiment. Scientists should clearly understand that cost is an important factor in experimental design. While it might make sense to design an experiment involving 50 rats in a toxicity study, it would usually not make sense to specify the use of 50 nonhuman primates in a similar study because it
84550.indb 383
11/11/10 6:26:43 PM
384
Handbook of Laboratory Animal Science
would cost too much. All other things being equal, the larger an experiment is, the more powerful it will be and the more it will cost both economically and ethically. Experimental design involves a balancing act, where the aim is to get as much good information as possible as economically as possible. Any resources saved as a result of getting the design right can be used to fund more experiments. Scientists on a fixed budget have every incentive to design their experiments in such a way as not to waste their financial resources. In general, the resource equation method is likely to be appropriate for complex biological experiments where relatively large treatment effects are expected. The power analysis method is preferred, but when it is difficult or impossible to use, the resource equation method is often useful. Formal Experimental Designs The Completely Randomized Design With this design, subjects are allocated to treatments strictly at random using one of the randomization methods described before. If the treatment is something like an injection, which can be given at the start of the experiment, the subjects can then be coded. The rest of the experiment can then be done “blind,” without the investigator knowing to which treatment an individual subject belongs until the data are decoded at the time of the statistical analysis. If the treatment is given in the diet, then color coding can be used so that staff do not know which animals are receiving the test diet. This design can have any number of treatment groups, and unequal numbers in each group usually present no problems, provided the treatments do not have a factorial structure, such as males and females with three dose levels. However, in most cases, the experiment has highest precision if there are equal numbers in each group, except in the situation where a control group is being compared with several different treatment groups. In this case, the control group might be increased in size to benefit study power. In this situation, ideally, if there are t treatment groups, the ratio of numbers in the control group to those in the other groups should be the square root of t. Thus, with five treated groups, including the control, the numbers in the control group should be approximately the square root of five (i.e., 2.2) times that in each of the treated groups (assuming these numbers are equal). This would normally be rounded down to twice the numbers in the treated groups. If the resulting data are quantitative, then they will often be analyzed by ANOVA methods or via nonparametric methods such as the Mann–Whitney rank-sum test or Kruskal–Wallis test described in many statistical textbooks (Snedecor and Cochran 1980; Maxwell and Delaney 1989). Table€13.2 shows the results of an experiment using a completely randomized design in which adult C57BL/6 mice were randomly allocated to one of four dose levels of a hormone compound, and the uterus weight was measured after an appropriate time interval. The method of analysis depends to some extent on the distribution of the data. These data are analyzed in the next section. Completely randomized designs may also produce categorical data. Table€13.3 comes from a study of the effect of the nonsteroidal antiestrogen agent Tamoxifen on mutations in BigBlue transgenic mice (Davies et al. 1996). Mice were assigned to either a control group or a group treated with Tamoxifen, and the number of genetic mutations occurring in a region of the transgene in the mice was recorded. Subsequently, approximately 50 such mutations from each group were chosen at random and their nucleotide sequence was determined. The table shows the number of mutations of three types. The question is whether the type of mutation differed between the two groups. The analysis of these data is discussed later. The main problem with the completely randomized design is that, with large experiments, it is often difficult to obtain animals of uniform weight and age. If the body weight of the animals is quite variable or if the experiment has to be split up and done at different times, it may be inefficient
84550.indb 384
11/11/10 6:26:43 PM
Experimental Design and Statistical Analysis
385
Table€13.2╅Uterus and Body Weight in C57BL/6 Mice Treated with Various Doses of Estrogen (Arbitrary Units) Dose
Body Weight (g)
0 0 0 0 1 1 1 1 2.5 2.5 2.5 2.5 7.5 7.5 7.5 7.5 50 50 50 50
Uterus Weight (g)
12.7 11.6 10.4 11.9 11.2 11.0 10.0 10.6 11.3 12.0 10.7 12.2 12.6 11.9 12.7 13.6 13.3 13.2 12.8 13.3
0.0118 0.0088 0.0069 0.0090 0.0295 0.0264 0.0189 0.0242 0.0515 0.0560 0.0449 0.0514 0.0833 0.0948 0.1017 0.0780 0.1130 0.0623 0.0802 0.0912
because the noise levels (interindividual variation) are too high. In such cases, a randomized block or Latin square design may be substantially more powerful. The Randomized Block Design This design can be used to take account of heterogeneous features of the study or subjects, such as animals differing in their initial body weight, time and space variables that result when the measÂ� urements cannot all be done at the same time, or when the animals have to be housed in more than one room or on different shelves in the same room. Essentially, a randomized block design involves splitting up the experimental material into a series of “mini-experiments”; each involves one or a few subjects on each treatment. For example, with an experiment involving four treatments and five animals per treatment, block 1 might consist of the four heaviest animals assigned at random to one of the four treatments. These animals would be started on the experiment at the same time, housed on the same shelf, and measurements made at Table€13.3â•…Type of Mutation in Transgenic Mice Treated with Tamoxifen or Kept as Controls Treatment Type of mutation C:C to T:A G:C to A:T Other Totals
84550.indb 385
Control
Tamoxifen
7 15 31 53
20 6 26 52
Totals 27 21 57 105
11/11/10 6:26:44 PM
386
Handbook of Laboratory Animal Science
Table€13.4╅ Cell Counts in Tissue Culture Dishes Treatment
Block 1
Block 2
Block 3
Block 4
Means
con TPA Gen
100 514 35
81 187 82
62 294 148
128 558 241
92.8 388.3 126.5
Gen + TPA
120
84
134
1011
337.3
Notes: All received media. The controls (con) had no additives; some dishes had TPA added, some Genistine (Gen), and some Genistine + TPA. For further details, see text.
the same time. Block 2 might have the next four heaviest animals, possibly started at a later date, but again housed together, possibly in a different room, and terminated at a different time from block 1, and so on with the next three blocks. Randomized block designs can be particularly useful when the experimental material (e.g., animals) has some sort of natural structure. For example, it may be difficult to collect enough animals of a transgenic strain of mice all at once to do an experiment with the desired sample size. However, it may be possible to do the experiment as a randomized block design using small numbers as they become available. The only limitation would be the need to collect enough animals to have one representative for each treatment group. Table€13.4 shows an experiment done as a randomized block. This was an in vitro experiment (emphasizing the point that the methods discussed here are applicable to all experiments, whether or not they involve animals) in which dishes of cells were treated with a vehicle (the control), TPA (a phorbol ester), Genistine, or TPA + Genistine (unpublished data). Data were the number of cells following further incubation under specified conditions. Notice that there are four treatments, but that they have a factorial structure in that half the dishes had Genistine and half did not, and half had TPA and half did not. The mini-experiment using just four dishes was repeated four times. Thus, this was a randomized block experiment with four treatments arranged as a 2 × 2 factorial layout. The statistical analysis of this experiment is explained later. Within-Subjects Designs Sometimes it is possible to do a “within animal” design, such that the animal receives the treatments in sequential order, usually with a rest period between them. This is generally a powerful design because the variation within an individual is usually less than that between individuals; however, it is only appropriate for certain types of relatively minor treatments that do not permanently alter the animal. As an example, Table€13.5 shows an experiment to study taste preferences in mice. Cages containing two mice were set up with two drinking bottles each, one containing distilled water and the other the test compound (alcohol, sucrose, etc.). The control was two bottles of distilled water; one was arbitrarily designated as the “treated” one. The position of the bottles on the cages was rotated each day, and the percentage of test fluid consumed as a percentage of total fluid consumption was recorded over 1-week periods, with a 1-week rest between treatments. Treatments were applied to cages in a random order. Notice that with this experimental design, the experimental unit is a cage of two mice for a period of time, and it is assumed that any one treatment does not permanently alter the behavior (taste preference) of the mice. The aim was to determine whether the mice preferred any of the test solutions over others. This design is very similar to a randomized block design, with the “block” in this case being the cage of mice and treatments (the different test fluids) being assigned to time periods at random. The analysis of these data is described later.
84550.indb 386
11/11/10 6:26:44 PM
Experimental Design and Statistical Analysis
387
Table€13.5╅ Percent of Test Fluid Consumed by C57BL Mice Offered Both Distilled Water and One of the Test Fluids in a Two-Bottle Choice Experiment Using a Repeated-Measures Design Cage
Treatment and Percent of Test Fluid Consumed
1
B 69.6
C 61.9
A 54.9
E 69.4
D 78.3
2
A 48.2
D 81.5
E 60.9
B 61.1
C 43.9
3
C 53.4
D 74.7
A 49.9
B 58.2
E 68.5
4
E 64.5
A 50.4
D 73.6
B 55.3
C 50.7
Notes: Treatments: A = control, distilled water in both bottles; B = 0.02% saccharin; C = 0.05 M sodium chloride; D = 0.04 M sucrose; E = 10% ethanol. Mice were housed two per cage. The control group had two water bottles, one of which was arbitrarily designated as the “treatment.”
Other Designs Several other designs can be used in appropriate circumstances, but cannot be discussed in detail here. A Latin square design can be used to control two sources of heterogeneity that cannot be included together in one block of a randomized block design. For example, a replicated 3 × 3 Latin square was used to compare the behavioral effects of bleeding from the orbital sinus, diethylether anesthesia, and sham anesthesia in rats (van Herck et al. 2000). Each of three rats received all three treatments, with a rest between them, but balanced in a way so as to cancel out any possible trend in behavior over the period of the experiment. An incomplete block design can be used when there is a natural block size due to the nature of the experimental units, but more treatments are to be compared than can fit into a block. These designs can be quite complex, so professional advice is usually necessary. Sequential designs can be used when the response of an individual subject to some treatment can be obtained quickly and a decision can be made on whether to proceed to the next individual or to terminate the experiment. This may follow because the accumulated information has provided sufficient evidence of a positive effect or the power of the experiment is now high enough to preclude an effect of a specified magnitude. The “up-and-down” method for determining the LD50 of a compound is an example of this type of design (Lipnick et al. 1995). Such designs are usually very efficient and use small numbers of animals, but they are only applicable in some circumstances. Split plot designs are ones where there are two types of experimental units. For example, several cages, each containing two mice, might be assigned to different dietary treatments. Within each cage, one of the mice might be given an injection of vitamins, while the other is given an injection of the vehicle as a control. In this case, the cage of mice is the experimental unit for comparing diets, but the mouse is the unit for comparing the effects of the vitamin injection. Although these are useful designs, their statistical analysis is quite complicated. Fractional factorial designs can be used to compare many different treatments simultaneously using relatively few experimental subjects (Montgomery 1997). They have been used, for example, in screening potentially active compounds in drug discovery. Animals may receive 10 or more treatments (presence or absence of compounds A, B, C, etc.) in such a way that if one or more of the compounds are biologically active for a particular end point, this can be demonstrated following the statistical analysis (Shaw et al. 2002). This is a useful way of reducing the use of animals and costs, but requires more advanced statistical methods than can be described here.
84550.indb 387
11/11/10 6:26:44 PM
388
Handbook of Laboratory Animal Science
Dose
50.0 7.5 2.5 1.0 0.0
0.01
0.06 Uterus Weight
0.11
Figure 13.2â•…Dot plot (MINITAB) of the raw uterus weight data in Table 13.1, by dose level. Note that uterus weight certainly increases as the dose level increases, but that the variability also increases.
Statistical Analysis Data Screening The first step in any statistical analysis is to screen the data for obvious entry, transcription, or measurement errors and to obtain a general feel for or impression of the results. Graphical methods available in all good modern statistical packages are invaluable for this. Figure€13.2 is a dot plot produced by the MINITAB statistical package of the data shown in Table€13.2. It shows individual observations, with clear evidence that the hormone compound treatments are increasing uterus weight in mice, but with the variation also increasing. Figure€13.3 is a similar dot plot for the data in Table€13.4. Treatment groups have been labeled one to four in this case, though the final statistical analysis will take into account the factorial nature of the treatments, and the blocking has been ignored. Figure€13.3 shows that one point in treatment group 4 appears to be an outlier. When such an outlier is found, it should be checked to ensure that it is not simply a typographical or other type of data entry error. Assuming that this is not the case (as is the situation here), the presence of the outlier will need to be taken into account when analyzing the data. This is discussed later. The Analysis of Variance
Treatment Code
Most well-designed experiments producing measurement data can be analyzed by the ANOVA. This is a highly versatile statistical method that can be used for experiments with several treatment groups and with all types of experimental designs. It is much more versatile than the t-test, which can only be used for comparing two groups. In fact, when there are only two groups, the t-test and the ANOVA give mathematically identical results. 4 3 2 1
200
400
600 Cell Count
800
1000
Figure 13.3â•…Dot plot for the cell count data in Table 13.3. Treatment groups have been coded 1–4; 1 is the control group (medium alone), 2 is TPA, 3 is Genistine, and 4 is TPA and Genistine. The blocking has been ignored. Note that there is some evidence, particularly in group 2, that variation increases as the mean increases. Note also the outlier in group 4.
84550.indb 388
11/11/10 6:26:45 PM
Experimental Design and Statistical Analysis
389
The ANOVA quantifies the total variation in the data and partitions it into components associated with treatments, error, and, in some cases, other components. With modern computer packages, it is extremely easy to do a one-way ANOVA. The data from Table€13.1 were analyzed using MINITAB. With this and most other software packages, the observations are placed in one column of a data sheet with codes representing the treatment groups in another column. A menu command is then used to do the analysis. The ANOVA assumes that the variation in each group is approximately the same the deviations from the group means (known as the residuals) have an approximately normal distribution there is independent replication of the observations
Serious departures from these three conditions will lead to results that are not reliable, although mild departures can be tolerated. The first two conditions can be assessed by doing a normal probability plot of the residuals, which should produce an approximately straight line, and by plotting the “fits” (group means) versus the residuals. In MINITAB and many other good packages, these graphs are produced as part of the ANOVA analysis. The two plots for the data in Table€13.1 are shown in Figures€13.4 and 13.5. Clearly, there is some evidence for non-normality of residuals because the points in Figure€13.4 do not lie on a very straight line, and evidence from Figure€13.5 indicates that there is more variation in some groups than in others. The practical importance of these two departures from the assumptions underlying the analysis is to reduce the ability of the study to discern true differences that might really exist between the treatment groups, increasing the chance of a type II error. The third assumption depends on the correct design of the experiment. Any two experimental units must be capable of being assigned to different treatments and there must be no systematic biases that could affect treatment groups differently. Unfortunately, there is no easy way of checking that these assumptions have been met. The output of an ANOVA is shown in Table€13.6, which is the analysis of the log (see later discussion) of the uterus weights (×100) from the data shown in Table€13.1. Note that coding the raw data by multiplying it by a constant such as 100 does not alter the statistical analysis and is advisable in cases such as this one in order to avoid rounding errors and excessive numbers of decimal places in the output.
Normal Score
2 1 0 –1 –2
–0.03
–0.02
–0.01
0.00 Residual
0.01
0.02
0.03
Figure 13.4â•…Normal probability plot of the residuals from the analysis of the uterus weight data from Table 13.1 (MINITAB). In judging whether the points fit a reasonably straight line, not too much attention should be given to a few apparent outliers. Therefore, in this case, it is debatable whether a transformation is necessary. However, the variation certainly increases with the mean, as shown in Figure 13.5, suggesting that a transformation would be advisable.
84550.indb 389
11/11/10 6:26:46 PM
390
Handbook of Laboratory Animal Science
0.03
Residual
0.02 0.01 0.00 –0.01 –0.02 –0.03 0.00
0.01
0.02
0.03
0.04 0.05 0.06 Fitted Value
0.07
0.08
0.09
Figure 13.5â•…Plot of fitted values versus residuals for the raw uterus weight of Table 13.1. Note the increase in variation as the fitted values (group means) increase.
The first column, headed “source,” lists the sources of variation present in the data—in this case “dose,” “error,” and “total.” The next column, headed “DF,” lists the degrees of freedom for each source of variation. These can be regarded as the number of units of variation associated with each source of variation. There are five dose levels, and the degrees of freedom for the dose are shown as four. This is because two observations will give one estimate of the variation among dose levels (the difference between them), three observations will give two estimates of variation, and n observations will give n – 1 estimates of variation. Thus, the DF are given by n – 1, where n is the number of objects. The next column, headed “SS,” is the sums of squares associated with each source of variation. This can be taken as a quantitative estimate of each source of variation. The column headed “MS” (for mean square) puts this on a per-unit basis by dividing the SS by the DF column. The column headed F shows the F-statistic associated with the relevant sources of variation (only dose in this case). This can be used to produce the value for “p” shown in the last column. Before computers were available, these F-values had to be looked up in statistical tables, but nowadays the computer Table€13.6â•…ANOVA Analysis of the Uterus Weight Data Given in Table€13.1, Using the Log of Uterus Weight (×100) One-way ANOVA: LogWt versus Dose: Analysis of Variance for LogWt Source
DF
SS
MS
F
Dose Error Total
4 15 19
2.81567 0.09552 2.91120
0.70392 0.00637
110.54
Level 0.0 1.0 2.5 7.5 50.0
N 4 4 4 4 4
Pooled StDev = 0.0798
84550.indb 390
Mean –0.0476 0.3879 0.7058 0.9492 0.9279
StDev 0.0953 0.0822 0.0397 0.0523 0.1081
P 0.000
Individual 95% CIs for mean based on pooled StDev – – – –+â•›– – – – – – – – –â•›+–â•›– – – – – – – – â•›+â•›– – – – – – – – –â•›+– – (--*-) (-*--) (-*--) (-*--) (--*-) – – – –+â•›– – – – – – – – –â•›+–â•›– – – – – – – – â•›+â•›– – – – – – – – –â•›+– – 0.00 0.35 0.70 1.05
11/11/10 6:26:47 PM
Experimental Design and Statistical Analysis
391
produces the p-value, which in this case is 0.000, shown to three decimal places. The true p-value is actually slightly larger than zero, and should normally be reported as p < 0.001. This p-value is the probability that differences as large as or larger than those obtained in this experiment could have arisen by chance sampling variation, in the absence of any true difference between treatment groups. If this value is small, then it is assumed that it is unlikely to have arisen by chance, so it must be a result of the treatment. Note that the error mean square is the pooled estimate of the variance. Its square root is the pooled standard deviation. This is used in calculating post hoc comparisons and can be used in presenting the results. Transformations When the data do not fit the assumptions necessary for an ANOVA, they can be transformed to another, more appropriate scale or a nonparametric test can be used. A log transformation is often useful when the standard deviation increases with the mean; an angular transformation (Snedecor and Cochran 1980) is often used for percentage data when many of the values are less than about 20% or more than about 80%, and a square root transformation can be used for data with a Poisson distribution, particularly when the mean count is quite low. A log transformation (to any base, such as base e or base 10) of the uterus weight data (×100 to avoid negative numbers) in Table€13.1 corrects the deviations from normality and makes the variation in each group about equal as judged by the two graphs. In each case, residual plots should be used to assess the success of any transformation. Note, however, that transformations of the raw data can sometimes complicate interpretation of findings, since they change the units of measurement and summary values. Nonetheless, if the null hypothesis is rejected following an appropriate scale transformation, this implies that the samples are unlikely to come from the same population, even if this is not always apparent on the untransformed scale. Another powerful and useful transformation when non-normality and/or outliers are present is to replace the observations by their ranks. An ANOVA and F-test applied to ranks is equivalent to the Kruskal–Wallis test. This transformation, with the resulting data analyzed by the ANOVA, can be used in many cases where a nonparametric method is not available (Montgomery 1997). Post hoc Comparisons and Orthogonal Contrasts When an experiment has more than two treatment groups, as in the preceding, the first step is to conduct an ANOVA to test the null hypothesis. However, if the ANOVA p-value is less than 0.05, the scientist may want to know which group means differ from one another or which differ from the control group. There are several methods for making such comparisons. Using t-tests to compare all possible combinations of paired means is not advisable, since the resulting type I error rate then becomes artificially inflated above the specified value. Most computer packages offer several post hoc comparison methods that differ in their most appropriate use and in their characteristics. Among the most common of these, Dunnett’s test is used when the aim is only to compare each group mean with a control group; however, when the treatments are dose levels, it may be more appropriate to study the dose–response relationship using regression analysis or orthogonal contrasts in an ANOVA. There are several different tests for comparing all means. These include Tukey’s test, Fisher’s least significant difference, Duncan’s multiple range test, and the Newman–Keuls test (Snedecor and Cochran 1980; Fisher and van Belle 1993; Montgomery 1997). Each of these has slightly different properties, and it is not possible to state which is most appropriate in any given situation. Other tests will be offered in some computer software packages. In most cases, it would be safe to use one of the post hoc methods available in the computer package being used.
84550.indb 391
11/11/10 6:26:47 PM
392
Handbook of Laboratory Animal Science
For the hormone compound data in Table€13.1, there was one control group and three treatment groups in a study that demonstrated a clear dose–response relationship on uterus weights, including one very high dose level. Presumably, the highest dose was chosen to try to gauge the maximum possible response under these test conditions. With this type of data, Dunnett’s test might be used to see which groups differed from the controls. Further comparisons would really depend on the purpose of the experiment. With the four lower dose levels, it may be of interest to explore the form of the dose–response relationship quantitatively. These methods will be discussed in the section on linear regression. When categorical data are produced in a “contingency table” (Table€13.3), an analysis appropriate to counts is needed, such as a chi-square (χ2) test. In this case, the hypothesis to be tested is that the proportions of each type of mutation are the same in the two groups. This is equivalent to testing the hypothesis that the rows and columns are independent. The formula for performing χ2 tests is given in most statistical textbooks and is available in most software packages. For the genetic mutation experiment of Table€13.3, the calculated χ2 statistic = 10.5 with two degrees of freedom—calculated as (r–1) × (c–1), where r and c are the numbers of rows and columns, respectively—and a corresponding p = 0.005; thus, the null hypothesis of no difference in mutation type by treatment group would be rejected. The results could be presented as percentages or proportions, in which case a confidence interval should be stated, as described in the section on presentation of data. Thus, in the controls, 13% (5–25%) of the mutations were C:C to T:A, 28% (17–42%) G:C to A:T, and 58% (44–72%) “other”; the numbers in parentheses were the 95% confidence interval for the percentage, as calculated using the MINITAB statistical package. The main assumptions required for χ2 tests are that the observations are independent and the expected values under the null hypothesis in each cell of the table are greater than or equal to five. Most good computer packages will list the expected values, but they can be calculated manually as the product of the row and column totals for a particular cell in the table, divided by the grand total number of observations. If there are small expected numbers in some of the cells of a contingency table, Fisher’s exact test can be used instead. However, that test is tedious to do by hand, even for a 2 × 2 table, so published tables (Fisher and van Belle 1993) or software packages such as StatXact (version 3 or later for Windows, Cytel Software Corp., Cambridge, MA) can be used for this purpose. Several Web sites will also do the calculations. A Web search on Fisher’s exact test should find them. Another way of comparing groups, often favored by statisticians, is to use a set of “orthogonal contrasts” (Snedecor and Cochran 1980; Fisher and Van Belle 1993; Montgomery 1997). Orthogonal contrasts can be used to test whether there is a linear or nonlinear trend when the treatments represent different levels of a variable, such as dose of a test compound, and these are equally spaced on some scale. However, a discussion of the use of these methods is beyond the scope of this chapter. Two- or Three-Way ANOVA The analysis of the data in Table€13.4 is a bit more complicated, though still easy using the appropriate computer software package. The experiment used a randomized block design, and the block effect needs to be removed in the analysis. The treatments had a factorial structure, so this needs to be built into the analysis, and there is one outlier value in the data to think about (Figure€13.3). In the MINITAB statistical package, the observations are put in one column of the data sheet, the block number in the next, the TPA treatment (coded 1 or 2) in a third column, and the coded Genistine treatment in the fourth column. An ANOVA is then done taking account of the block, the TPA treatment, the Genistine treatment, and the interaction between TPA and Genistine. Details of exactly how this is done are given in the software manuals. The usual plots of the type shown in Figures€13.4 and 13.5 are produced. From these it is clear that the data need to be transformed prior to the final analysis. A log transformation was made and the data were reanalyzed. This time, the
84550.indb 392
11/11/10 6:26:47 PM
Experimental Design and Statistical Analysis
393
Table€13.7a╅Analysis of Variance of the Log of the Data in Table€13.4 Analysis of Variance for LogScore Source
DF
Replicat TPA Gen TPA*Gen Error Total
3 1 1 1 9 15
SS
MS
F
P
0.74028 0.77213 0.04627 0.09994 0.56775 2.22636
0.24676 0.77213 0.04627 0.09994 0.06308
3.91 12.24 0.73 1.58
0.049 0.007 0.414 0.240
two graphs showed no evidence of non-normality or heterogeneous variances. Moreover, on this scale, the outlier disappeared. This analysis should now be the final one necessary to address the hypothesis (Table€13.7). However, what if the outlier did not disappear after the transformation? One approach would be to do the analysis with and without the outlier to see what difference it makes in the analysis. In this experiment, elimination of the data point representing the outlier on the log scale (where, as noted, it no longer appeared to be an outlying value) did not make any meaningful difference to the interpretation of results (data not shown). The main change was that the TPA*Genistine interaction now approached significance at p = 0.06, so it might be concluded that if the experiment had been larger, it might have shown that TPA and Genistine applied together reduced the number of cells more than their combined individual effects. Looking at the ANOVA results in Table€13.7a (with the outlier), revealed that the “main effect” (i.e., averaged over both levels of Genistine) of TPA was to increase the counts significantly from a mean of 110 to 363 cells (Table€13.7b, on the untransformed scale). The main effect of Genistine was not significant and there was no interaction between them. Analysis of the drinking taste preference data in Table€13.5 presents no problems. It is analyzed as a two-way ANOVA without interaction. This eliminates the effect of differences between cages so that the effects of the treatments can be assessed with high precision. The graphs to explore normality of the residuals and heterogeneity of variances give no evidence that would preclude the use of the raw data in the analysis. However, the data in this case are the percentages of test fluid consumed. Had these percentages been higher (say, many above 80%) or much lower (below 20%), then an angular transformation may have been necessary. This is again an example where Dunnett’s test seems most appropriate in order to compare each solution with the control. A potential criticism of the design might be that, because four treatments are being compared with the control, the latter should probably have been assessed more accurately using another set of control measurements on each cage. The results of the analysis are presented in Table€ 13.8. Dunnett’s test was used for the post hoc comparisons, and this indicated that the controls did not differ significantly from group 3 (sodium chloride), but did differ significantly (p < 0.001) from each of the other three groups. Table€13.7bâ•… Means (on the Untransformed Scale) for the Main Effects of TPA and Gen and the TPA*Gen Interactions
84550.indb 393
Main Effect, TPA
Mean
Main Effect, Gen
Mean
None TPA
109.63 362.75
None Gen
240.50 231.88
Interaction, TPA*Gen
Mean
None, none None, Gen TPA, none TPA, Gen
92.75 126.50 388.25 337.25
11/11/10 6:26:47 PM
394
Handbook of Laboratory Animal Science
Table€13.8â•…Analysis of Variance with Dunnett’s Post Hoc Comparisons of the Data from Table€13.5 ANOVA: Percent versus Cage and Treats Factor
Type
Levels
Values
Cage Treats
Random Fixed
4 5
1 1
2 2
3 3
4 4
5
Analysis of Variance for Percent Source
DF
SS
MS
F
P
Cage Treats Error Total
3 4 12 19
205.14 1819.17 184.78 2209.09
68.38 454.79 15.40
4.44 29.53
0.026 0.000
Means Treats 1 2 3 4 5
N 4 4 4 4 4
Percent 50.850 61.050 52.475 77.025 65.825
Dunnett 95.0% simultaneous confidence intervals Response variable percent Comparisons with control level Treats = 1 subtracted from: Treats 2 3 4 5
Lower 2.411 –6.164 18.386 7.186
Center
Upper
---------+----------------+-----------------+-----------------+
10.200 1.625 26.175 14.975
17.989 9.414 33.964 22.764
(----------*-----------) (-----------*------------) (--------------*---------------) (-----------*------------) ---------+----------------+-----------------+-----------------+ 0 12 24 36
Dunnett simultaneous tests Response variable percent Comparisons with control level Treats = 1 subtracted from: Level Treats 2 3 4 5
Difference of Means
SE of Difference
T-value
Adjusted P-value
10.200 1.625 26.175 14.975
2.775 2.775 2.775 2.775
3.6760 0.5856 9.4333 5.3969
0.0106 0.9360 0.0000 0.0006
Note that the cages did differ significantly, implying that the blocking was worthwhile. With the analysis shown in Table€13.8, the pooled within-group variance (given by the error mean square) was 15.4, yielding a standard deviation of 3.9. When the data were reanalyzed ignoring the blocking (not shown), the error mean square was 25.99, giving a standard deviation of 5.1. Thus, by decreasing the standard deviation from 5.1 to 3.9, the blocking increased the power of the experiment to a useful extent.
84550.indb 394
11/11/10 6:26:48 PM
Experimental Design and Statistical Analysis
395
Multiple Dependent Variables In many experiments, more than one dependent variable (character) is being measured. Commonly, each variable will be analyzed separately. However, if the variables are correlated, then the tests will not be independent of each other. Thus, if a false positive or negative result occurs with one of the variables, it may also occur with the other. Various “multivariate” methods, such as principal components analysis, can sometimes be used in such cases. In extreme cases, hundreds or even thousands of measurements may be done on each individual, such as in experiments involving gene microarrays. Specialist advice is needed in such situations. Another common situation is when an experimental subject is measured several times over a period of minutes, hours, or days. Growth curves are a typical example. These are sometimes analyzed as a “repeated measures” design. However, there are objections to such an analysis because it treats each observation over time as an independent variable, when in fact they are correlated. It is usually better to analyze some function of the measurements, such as their average, the difference between the first and last, the slope of the best fitting line (estimated separately for each individual using regression, discussed later), or the area under the curve. If the repeated measures-type response variable is of the categorical type, other sophisticated approaches (e.g., the general mixed model of SAS) may be required. Linear Regression and Correlation In some experiments, the aim is to see whether there is any linear association or causal relationship between two or more variables. For example, Table€13.2 shows both body and uterus weight in the mice. The “product–moment correlation,” also sometimes called the Pearson correlation, can be used to quantify the linear association between these on a scale of –1, which implies a perfect inverse relationship, to +1, which implies a perfect positive relationship. The correlation between body and uterus weight in this case is 0.685, 0.592 between body weight and log uterus weight, and 0.945 between uterus weight and log uterus weight. Note that the last of these is less than 1.0 because the relationship between the two is not linear. The value of a linear correlation may depend on the scale of measurement. Correlation coefficients are often tested to see whether they differ significantly from zero. This can be done using tables available in some textbooks. Alternatively, p-values are often given in computer output. All these correlations differ from 0 at p < 0.01. A number of other types of correlation might be appropriate if one or more of the variables is categorical. These are discussed in most statistical textbooks (Snedecor and Cochran 1980; Fisher and van Belle 1993; Montgomery 1997). A linear association between an independent or predictor variable, usually designated “X,” and a dependent or outcome variable designated “Y” can be quantified using linear regression analysis. This gives the best fitting straight line, Y = a + bX, where Y is the estimated Y-value; a and b are the intercept and slope constants estimated from the data, respectively; and X is the value of the X-variable. “Best fitting” in this case means the line that minimizes the sum of squared deviations from the line. As an example, the data in Table€13.2, excluding the top dose, can be used to explore the relationship between the dose of an estrogen and uterus weight. The output of a regression analysis of these data is given in Table€13.9. The output includes an ANOVA table, in addition to the parameters for the best-fitting regression equation, which indicates whether the slope of the line differs significantly from zero. Note that this analysis gives the best fitting straight line, though the true relationship may not be linear. Many software packages will also produce a graph like the one shown in Figure€ 13.6. This shows the best fitting straight line, which is clearly not an ideal fit since all of the 2.5-dose group values lie clearly above the line. This plot also shows an inner set of confidence bands, which are the
84550.indb 395
11/11/10 6:26:48 PM
396
Handbook of Laboratory Animal Science
Table€13.9╅Regression Analysis of Uterus Weight versus Dose, Omitting the Highest Dose Level Regression analysis: uterus wt versus dose The regression equation is Uterus wt = 0.0151388 + 0.0103382 dose S = 0.0087338 R-Sq = 93.0% R-Sq (adj) = 92.5% Analysis of Variance Source
DF
SS
MS
F
P
Regression Error Total
1 14 15
0.0142147 0.0010679 0.0152826
0.0142147 0.0000763
186.352
0.000
Note: It would probably have been better to multiply the uterus weights by, say, 100 in order to avoid the large number of decimal places in the analysis. This can lead to serious arithmetical errors, particularly when using spreadsheets, which do not always use double precision arithmetic.
95% confidence interval for the mean value of Y, given any value of X (i.e., for any value of X, we can be 95% confident that the true value of Y falls within the inner bands shown). The outer set of bands represents the 95% prediction interval for individual values of Y, given any value of X (i.e., there is a 95% confidence that individual points will fall within these outer bands). The plot also includes the formula for estimating any value of Y given any value of X and shows the R2 value—that is, the proportion of the variation in Y that is accounted for by variation in X. In this case, approximately 93% of the variation in Y is accounted for by variation in X. However, the plot also shows that the relationship between dose and uterus weight is apparently not linear, but instead somewhat curved, as noted. A plot of log (dose + 1) versus uterus weight provides a slightly better fit (R2 = 94.4%). Some software, such as MINITAB, also makes it relatively easy to fit a second degree polynomial curve to the data. This gives the best fitting curve for the data. However, discussion of curve fitting is beyond the scope of this chapter. Regression 95% Cl 95% PI
Uterus Weight
0.10
0.05
0.00 0
1
2
3
4 Dose
5
6
7
8
Figure 13.6â•…Regression plot for the uterus weight data of Table 13.1 versus dose of estrogen (omitting the highest dose level because it was well outside the range of the other doses). The best fitting line is shown (solid). The inner dashed lines show the 95% confidence interval for the mean (the solid line). The outer dashed lines show the 95% prediction interval for individual points (i.e., 95% of individual values should fall within these lines). Note that the straight line fit is not ideal in this case; other transformations may result in a better fit.
84550.indb 396
11/11/10 6:26:49 PM
Experimental Design and Statistical Analysis
397
Presentation of Data Once a powerful, unbiased experiment has been designed and analyzed, it is important to ensure that it is well presented: • Studies should be presented in a way that allows others to understand their purpose, including the hypothesis being tested, the basis for selecting the animal model, and the clinical or biological conclusions. • Details of the animals, including genetic nomenclature when available and husbandry conditions, should be described. • The sampling technique, methods of randomization to treatment group, rationale for blocking and/ or pairing—along with any criteria for exclusion from treatment (e.g., weight range)—should be included as part of the methods section. • The method of determining sample size should be specified. If this is done using a power analysis, then this should be fully described, including the specified power, significance level, and nature of the statistical tests. • The brand and version number of statistical software used for analysis should be stated.
A thorough set of annotated guidelines for reporting statistics in biomedical research (Lang and Secis 1997) includes appendices showing style preferences for presenting numbers and checklists for reporting on clinical trials in the peer-reviewed literature. Guidelines for the design and reporting of animal experiments are also available (Festing and Altman 2002). Similar guidelines and checklists have been published by others (Altman et al. 1989). These authors emphasize that numbers should be rounded when presented, but not when analyzed (due to potential loss of information) and that two or three significant digits are generally sufficient. Percentages should normally be reported to one decimal place at maximum, unless the sample size is less than 100, in which case whole numbers should be used. Numerators and denominators should always be shown. Continuous data should be summarized with the mean, number of observations, and standard deviation if the data are approximately normally distributed; otherwise, the median and interquartile range (i.e., the values representing the 25th to 75th percentiles of the distribution) are preferred. Standard errors of the mean (SEM) should only be used to provide a measure of precision for estimates of a mean drawn from a (possibly hypothetical) population. The use of unlabeled error bars on charts can likewise present confusion. It must be made clear whether they represent the sample SD, SEM, or confidence interval (CI). When possible, scatter plots should be used to show the individual observations. In any event, the sample size used to calculate SEM values must be included for them to be of any value, since they are calculated as
SEM = (standard deviation) / (sample size)
Scientists should use confidence intervals to describe the precision of estimates, including the upper and lower limits that form the bounds on the value of interest. Many statistical programs use confidence intervals in presenting the results of an analysis. For example, the results of Dunnett’s test in Table€13.8 are given in the form of the difference of each group mean from the control and 95% confidence interval for that difference. A 95% confidence interval of the differences will not include zero if the results are significant at the 5% level. Confidence intervals add value beyond that of p-values, since they span the range of estimates for the true (typically unknowable) population difference or effect, as well as showing whether the difference is statistically significant. Therefore, they can be used to aid in judgments about the importance of the findings, particularly when biological importance may differ markedly from statistical significance.
84550.indb 397
11/11/10 6:26:50 PM
398
Handbook of Laboratory Animal Science
Tables or figures should be used to present the main study findings whenever possible and should be constructed in a way that emphasizes important trends or comparisons. When it is of interest to compare several mean values by treatment group, they should be listed as columns in tables instead of rows to facilitate visual comparisons. Means and standard deviations should be reported to no more than three significant digits. Categorical data can often be presented in the text to save space, unless the numbers of categories warrant a table or chart. Reports should indicate the methods used to provide assurance that statistical assumptions for the chosen hypothesis tests were satisfied. This can include plots of residuals, as shown in Figures€13.4 and 13.5, and design considerations to demonstrate independence of the observations. Test statistics (such as the “F” value or “χ2” statistic) should be reported along with the corresponding degrees of freedom, allowing readers to verify the p-value for each explanatory variable. Actual p-values should be shown to two significant digits, and the smallest value that needs to be reported is p < 0.001. ANOVA results should be reported as a table when possible, including the “source,” “SS,” and “MS” values for exploring between-group differences. Any efforts to test for interaction and their subsequent treatment in the analysis should be described, along with how any outliers were handled in the analysis. Outliers should not be deleted unless independent evidence suggests that the data are incorrect, and the reasons for any deletions should be stated. If data have been transformed to fulfill any statistical assumptions, the results of the analysis should be converted back to the original units of measurement in the report. Note that the standard deviation cannot be converted back to the original scale, but the 95% confidence limits can be. Finally, the clinical or biological implications of the conclusions should be thoughtfully developed to aid in any judgments or decisions that might emerge from the work. In conclusion, a well-designed experiment with the correct statistical analysis should provide the best possible test of the experimental hypothesis, while at the same time minimizing the use of scientific resources and animals. These resources can then be used for further experiments to speed the rate of scientific progress. References Altman, D. G. 1982a. Statistics in medical journals. Statistics in Medicine 1:59. ———. 1982b. Misuse of statistics is unethical. In Statistics in practice, ed. S. M. Gore, 1–2. London: British Medical Association. Altman, D. G., S. M. Gore, M. J. Gardner, and S. J. Pocock. 1989. Statistical guidelines for contributors to medical journals. In Statistics with confidence, ed. M. J. Gardner, 83–100. London: British Medical Journal. Bebarta, V., D. Luyten, and K. Heard. 2003. Emergency medicine animal research: Does use of randomization and blinding affect the results? Academic Emergency Medicine 10:684. Crawley, M. J. 2005. Statistics. An introduction using R. Chichester, England: John Wiley & Sons, Ltd. Dalgaard, P. 2003. Introductory statistics with R. New York: Springer. Davies, R., V. I. Oreffo, S. Bayliss, P. A. Dinh, K. S. Lilley, I. N. White, L. L. Smith, and J. A. Styles. 1996. Mutational spectra of tamoxifen-induced mutations in the livers of lacI transgenic rats. Environmental and Molecular Mutagenesis 28:430. Festing, M. F. W. 1992. The scope for improving the design of laboratory animal experiments. Laboratory Animals 26:256. ———. 1994. Reduction of animal use: Experimental design and quality of experiments. Laboratory Animals 28:212. ———. 1996. Are animal experiments in toxicological research the “right” size? In Statistics in toxicology, ed. B. J. T. Morgan, 3–11. Oxford, England: Clarendon Press. ———. 1999. Warning: The use of genetically heterogeneous mice may seriously damage your research. Neurobiology of Aging 20:237.
84550.indb 398
11/11/10 6:26:50 PM
Experimental Design and Statistical Analysis
399
Festing, M. F. W., and D. G. Altman. 2002. Guidelines for the design and statistical analysis of experiments using laboratory animals. ILAR Journal 43:233. Festing, M. F. W., P. Diamanti, and J. A. Turton. 2001. Strain differences in hematological response to chloramphenicol succinate in mice: Implications for toxicological research. Food and Chemical Toxicology 39:375. Festing, M. F. W., P. Overend, R. Gaines Das, M. Cortina Borja, and M. Berdoy. 2002. The design of animal experiments. London: Laboratory Animals Ltd. Fisher, L. D., and G. van Belle. 1993. Biostatistics. A methodology for the health sciences. New York: John Wiley & Sons, Inc. Fisher, R. A. 1960. The design of experiments. New York: Hafner Publishing Company, Inc. Kacew, S., and M. F. W. Festing. 1996. Role of rat strain in the differential sensitivity to pharmaceutical agents and naturally occurring substances. Journal of Toxicology and Environmental Health 47:1. Kilkenny, C., N. Parsons, E. Kadyszewski, M. F. Festing, I. C. Cuthill, D. Fry, J. Hutton, and D. G. Altman. 2009. Survey of the quality of experimental design, statistical analysis and reporting of research using animals. PLoS One 4:e7824. Lang, T. A., and M. Secis. 1997. How to report statistics in medicine. Philadelphia, PA: American College of Physicians. Latham, N. 2010. Brief introduction to welfare assessment: A “toolbox” of techniques. In The UFAW handbook on the care and management of laboratory and other research animals, 8th ed., ed. R. Hubrecht and J. Kirkwood, 76–91. Oxford, England: Wiley-Blackwell. Lipnick, R. L., J. A. Cotruvo, R. N. Hill, R. D. Bruce, K. A. Stitzel, A. P. Walker, I. Chu, M. Goddard, L. Segal, J. A. Springer, and R. C. Myers. 1995. Comparison of the up-and-down, conventional LD50, and fixeddose acute toxicity procedures. Food and Chemical Toxicology 33:223. Maxwell, S. E., and H. D. Delaney. 1989. Designing experiments and analyzing data. Belmont, CA: Wadsworth Publishing Company. McCance, I. 1995. Assessment of statistical procedures used in papers in the Australian Veterinary Journal. Australian Veterinary Journal 72:322. Mead, R. 1988. The design of experiments. New York: Cambridge University Press. Montgomery, D. C. 1997. Design and analysis of experiments. New York: John Wiley & Sons. Perel, P., I. Roberts, E. Sena, P. Wheble, C. Briscoe, P. Sandercock, M. Macleod, L. E. Mignini, P. Jayaram, and K. S. Khan. 2007. Comparison of treatment effects between animal experiments and clinical trials: Systematic review. British Medical Journal 334:197. Roberts, I., I. Kwan, P. Evans, and S. Haig. 2002. Does animal experimentation inform human healthcare? Observations from a systematic review of international animal experiments on fluid resuscitation. British Medical Journal 324:474. Russell, W., and R. Burch. 1959. The principles of humane experimental technique, 2nd ed. London: Methuen. Ruxton, G. D., and N. Colgrave. 2006. Experimental design for the life sciences. Oxford, England: Oxford University Press. Shaw, R., M. F. W. Festing, I. Peers, and L. Furlong. 2002. The use of factorial designs to optimize animal experiments and reduce animal use. ILAR Journal 43:223. Snedecor, G. W., and W. G. Cochran. 1980. Statistical methods. Ames: Iowa State University Press. Thomas, L. 1997. A review of statistical power analysis software. Bulletin of the Ecological Society of America 78:126. van Herck, H., V. Baumans, H. A. G. Boere, A. M. P. Hesp, and H. A. van Lith. 2000. Orbital sinus blood sampling in rats: Effects upon selected behavioral variables. Laboratory Animals 34:10. Weigler, B. J. 2001. A primer in epidemiologic methodology. Comparative Medicine 51:208.
84550.indb 399
11/11/10 6:26:50 PM
84550.indb 400
11/11/10 6:26:50 PM
Chapter 14
Common Nonsurgical Techniques and Procedures
Vera Baumans and Cynthia A. Pekow Contents Introduction.....................................................................................................................................402 Handling and Physical Restraint.....................................................................................................402 Rats.............................................................................................................................................403 Mice...........................................................................................................................................407 Hamsters....................................................................................................................................407 Guinea Pigs................................................................................................................................407 Rabbits........................................................................................................................................408 Pigs.............................................................................................................................................409 Methods of Identification................................................................................................................ 413 Administration of Substances......................................................................................................... 416 Application to Skin or Mucous Membranes.............................................................................. 417 Enteral Administration............................................................................................................... 417 Rats........................................................................................................................................ 419 Mice....................................................................................................................................... 419 Hamsters................................................................................................................................ 419 Guinea Pigs........................................................................................................................... 420 Rabbits................................................................................................................................... 420 Pigs........................................................................................................................................ 421 Parenteral Administration.......................................................................................................... 421 Rats........................................................................................................................................ 424 Mice....................................................................................................................................... 426 Hamsters................................................................................................................................ 428 Guinea Pigs........................................................................................................................... 428 Rabbits................................................................................................................................... 428 Pigs........................................................................................................................................ 431 Intratracheal and Intrabronchial Instillation................................................................................... 431 Blood Sampling.............................................................................................................................. 431 Rats............................................................................................................................................. 432 Mice........................................................................................................................................... 434 Hamsters.................................................................................................................................... 435 401
84550.indb 401
11/11/10 6:26:50 PM
402
Handbook of Laboratory Animal Science
Guinea Pigs................................................................................................................................ 436 Rabbits........................................................................................................................................ 436 Pigs............................................................................................................................................. 436 Milk Collection............................................................................................................................... 439 Bone Marrow Collection................................................................................................................440 Urine and Feces Collection.............................................................................................................440 Cerebrospinal Fluid........................................................................................................................ 441 Rats............................................................................................................................................. 442 Mice........................................................................................................................................... 442 Hamster...................................................................................................................................... 442 Guinea Pigs................................................................................................................................ 442 Rabbits........................................................................................................................................ 442 Blood Pressure Measurement......................................................................................................... 442 Imaging Techniques........................................................................................................................ 443 Conclusion....................................................................................................................................... 443 References....................................................................................................................................... 443 Introduction Humane experimental technique minimizes distress to research animals and research personnel alike. This chapter outlines procedures for restraint, sampling, and dosing, with suggested humane refinements. Primary species covered are the mouse, rat, hamster, guinea pig, rabbit, and pig. Successful, efficient, humane technique requires practice and skill development. Adequate time must be budgeted to train and to work alongside experienced instructors in order to develop comfort and proficiency with specific procedures. Appropriate use of sedatives, analgesics, and anesthetics in research animals can minimize stress. Use of such medication is specified or suggested with a number of the techniques described in this chapter, and in the chapter on anesthetics (Olfert, Cross, and McWilliam 1993; Baumans et al. 2001; MacDonald, Chang, and Mitzner 2009). In addition, stress may be minimized if animals are first adapted to the research environment and are familiar with caretakers and research staff. Acclimatization and quarantine periods vary with the species and their places of origin. Many animals can be conditioned to accept common handling and restraint techniques and procedures readily (Lawson 1999; MacDonald et al. 2009). The techniques described here are certainly not the only methods available. In selecting a particular method, individuals should consider their proficiency, the stress to the animal of the restraint as well as of the procedure, and the history of success. All people privileged to work with research animals must have the integrity to know when a technique or method is not right in their hands and to seek assistance or employ another method. The use of experimental techniques requires specific skills that can be obtained only by intensive and careful training under the supervision of an experienced animal technician or laboratory animal scientist. Handling and Physical Restraint Safe, firm, gentle handling and restraint are integral to humane technique. A soft voice and demeanor, plus a calm, consistent laboratory atmosphere, assist in minimizing tension for animals and researchers. Animals communicate with smells and sounds that are often undetected by humans, and they should be protected from exposure to what they may perceive as distressful situations in other animals. Ideally, potentially distressful procedures are not performed in the animal housing area (Olfert et al. 1993; Baumans et al. 2001). Group housing of rats with familiar conspecifics can attenuate the stress response induced by handling and common procedures. A companion rat not
84550.indb 402
11/11/10 6:26:50 PM
Common Nonsurgical Techniques and Procedures
403
Figure 14.1â•…To remove a rat from the cage, the handler gently grasps the tail near the base and supports the body with the other hand.
subjected to research procedures has the most marked effect at reducing the length and magnitude of the stress response in its cage mates (Sharp et al. 2002a, 2002b, 2003a, 2003b). Rats Rats respond positively to gentle handling. Their inclination to hide and enter small spaces can be used to assist with restraint. Use of leather or wire mesh gloves is not necessary. Heavy gloves decrease dexterity of the handler, and rats may tear toenails on wire mesh gloves. A nervous or aggressive animal can be grasped with the aid of a towel or cloth drape. Rats may be handled by the tail for initiation of restraint, but care must be taken to grasp the tail at the base, near the body. The skin of the tail may be torn if handled near the tail tip. In general, rats are uncomfortable restrained by the tail alone, and this technique is used only for quick transfer of animals—for example, from cage to cage. Preferably, the tail is held only to keep the animal in place, and the weight of the rat’s body is supported from underneath (Figure€14.1). Rats unaccustomed to being handled may be placed in a towel or drape and allowed to hide in folds of material or gently secured in a stockinet tube. Three basic grip techniques are useful for restraining rats for procedures such as injection. In general, rats placed on a solid surface such as a countertop or handler’s forearm can be readily grasped for restraint. Rats should not be placed onto wire bar cage lids because they will grasp the bars and may damage their toenails as their grip is pried loose. Selection of restraint method will
84550.indb 403
11/11/10 6:26:51 PM
404
Handbook of Laboratory Animal Science
Figure 14.2â•…A secure two-handed rat restraint technique. Care is taken not to compress the chest.
vary with the size and docility of the rat, as well as the size of hands, comfort level, and procedure to be performed by the handler. For a secure two-handed grasp, one hand is placed over the rat’s shoulders, with thumb and index finger under the forepaws, gently pushing the forepaws toward the rat’s head. The forepaws may cross under the rat’s chin. The other hand encircles the body at mid-abdomen, to keep the rear legs restrained. Care is taken not to compress the chest or restrict the rat’s breathing (Figure€14.2). In a similar method using a single hand, the rat’s head is restrained by the index and middle fingers pushing up from behind on each side of the rat’s lower jaw. The thumb and fourth finger encircle the thorax behind the rat’s forepaws (Figure€14.3). When procedures are performed that require direct control of the head, rats may be grasped by the scruff. The skin over the neck, including the skin at the base of the ears, is firmly held by the index finger and thumb to control the head; the skin of the back is held by the other three fingers. Care is taken that the grasp does not restrict the rat’s ability to breathe (Figure€14.4). Use of terry towels, drapes, or stockinet provides secure and rapid restraint. A small rat can be gently pressed against the cage floor to perform an intramuscular injection (Figure€14.5). Cotton tube stockinet material is an adjunct to other restraint methods and provides a hiding place for the rat, yet easy access for the handler. The rat is gently guided into the bunched material, much like putting on a sweater or jumper. A band of tape at the neck area prevents the rat from exiting the cloth tube (Figure€14.6). A variety of commercially available restraint devices come in a range of sizes and are useful for permitting access to tail vessels. Most rats will voluntarily enter the openings of these rigid restrainers (Figure€14.7). Rats can be trained to sit quietly in the restrainer. Cone-shaped plastic bags (“decapicones”) are available for restraint of animals for decapitation by guillotine. These
84550.indb 404
11/11/10 6:26:52 PM
Common Nonsurgical Techniques and Procedures
405
Figure 14.3â•…Restraint of the rat using one hand.
Figure 14.4â•…A color version of this figure follows page 336. Restraint of the rat by the scruff. (Photo courtesy of T. P. Rooymans.)
84550.indb 405
11/11/10 6:26:52 PM
406
Handbook of Laboratory Animal Science
Figure 14.5â•…A color version of this figure follows page 336. Intramuscular injection in the rat. (Photo courtesy of T. P. Rooymans.)
Figure 14.6â•…Restraint of the rat in stockinet. Injections may be made through the material, or the material may be cut or peeled back to allow access.
Figure 14.7â•…A rigid restrainer is used to permit access to the tail vessels in rats and mice.
84550.indb 406
11/11/10 6:26:53 PM
Common Nonsurgical Techniques and Procedures
407
Figure 14.8â•…A plastic bag restrainer allows access to tail vessels in rats and mice and can be cut to permit access to rear legs. Animals become hot in these bags and must not be left in them for more than a few minutes. Note the breathing hole at the front end of the bag.
bags can also be used to briefly restrain rats or mice for access to blood vessels in the tail or rear legs (Figure€14.8). Mice Mice can be removed from the cage by grasping the tail near the base. For restraint, the animal can be placed on a surface on which it can take a firm grip, such as a wire bar cage lid. A confident, gentle grasp that keeps control of the head will counter the mouse’s common bite-defense response to restraint. The skin, including the base of the ears, must be held in the grip. The mouse should not be grasped so tightly as to restrict its respiration. The mouse can then be lifted and the tail held between the palm and the ring finger or little finger (Figure€14.9). Rigid plastic and metal restraint devices are available to permit access to tail vessels. A plastic cone-shaped bag (“decapicone”) may be used to restrain mice for access to the tail or a rear leg (Figure€14.8). Hamsters Hamsters should not be picked up before they are fully awake. Docile hamsters can be moved by simply scooping them into cupped palms or a small container or by grasping the loose skin over the back and lifting the animal. Hamsters have copious quantities of loose skin, which must be included in the grip to secure the animal for procedures such as injections. With the handler’s palm held flat against the hamster’s back, the animal should be firmly, yet gently, pressed to the cage floor to keep it in place. The loose skin is then gathered between fingers and palm. The skin by the ears must be included in the grip to allow control of the animal’s head. Care must be taken to ensure that the grip is not so tight as to compromise the animal’s breathing (Figure€14.10). Guinea Pigs Guinea pigs must be approached calmly and surely to decrease the chance of a panic response, in which frightened animals race about the cage. Guinea pigs are handled with one hand encircling the thorax from in front or behind and the other hand supporting the hindquarters (Figure€14.11). Holding the thorax too tightly may induce shock. Guinea pigs tend to hide in terry towels or drapes,
84550.indb 407
11/11/10 6:26:54 PM
408
Handbook of Laboratory Animal Science
(a) Figure 14.9aâ•…A color version of this figure follows page 336. Restraint of the mouse. The skin, including the base of the ears, must be held in the grasp to keep control of the mouse’s head. (Photo courtesy of T. P. Rooymans.)
which may be used to assist restraint. A firmly wrapped towel may be used as a primary method of restraint. Rabbits As an adaptation for rapid escape from predators, rabbits have powerful hind leg muscles, coupled with a light skeleton. This combination predisposes rabbits to fracture or subluxation of the lumbar spinal vertebrae (broken back), likely to occur if an animal attempts to leap or kicks while being restrained. Restraint methods therefore emphasize secure control of the front and hind ends. To remove a rabbit from a cage, the scruff or loose skin over the shoulders is grasped. The rabbit is turned to face the handler. The other hand is placed over or under the rabbit’s hindquarters, and the animal is lifted out (Figure€14.12). Rabbits can be securely carried close to the handler’s body with a two-handed restraint, with the animal’s head tucked under the handler’s arm (Figure€14.13). Rabbits should never be handled or lifted by the ears or by the scruff alone or allowed to struggle while restrained. If a rabbit struggles while being carried, the animal should be placed on the floor and temporarily released until a more secure grip can be achieved.
84550.indb 408
11/11/10 6:26:55 PM
Common Nonsurgical Techniques and Procedures
409
(b) Figure 14.9bâ•…A color version of this figure follows page 336. Restraint of the mouse. The tail is tucked between the palm and last fingers. (Photo courtesy of T. P. Rooymans.)
Use of tranquilizing medications such as phenothiazine drugs (e.g., acepromazine maleate at 0.8 mg/kg, s.c.) or opioid combinations (such as fentanyl/fluanisone [Hypnorm] at 0.2 mL/kg i.m.) can make restraint less stressful. About 15 min prior to handling, the medication can be given to an animal held by the scruff, without removing it from its cage. Aggressive or markedly frightened animals may first be covered with a drape while in the cage, then grasped for lifting or given tranquilizing medication. A variety of restraint devices can be used to permit access to blood vessels, for oral dosing, or other procedures. Ideally, rabbits are acclimated to restraint in these devices prior to experimental use. Rigid metal or plastic units (Figure€14.14), cloth bags (Figure€14.15), and towels (Figure€14.16) can be used. Once secured, rabbits may be less likely to struggle if their eyes are covered by a lightweight cloth. Struggling while restrained in a device can easily lead to lumbar spinal damage. Cloth bags and towels may be safer in this respect. Pigs Pigs can be trained to cooperate in a variety of procedures, with the use of positive reinforcement, such as food treats (Coria-Avila et al. 2007; Framstad, Oystein, and Aass 1988). Untrained
84550.indb 409
11/11/10 6:26:56 PM
410
Handbook of Laboratory Animal Science
Figure 14.10â•…Restraint of the hamster by grasping the loose skin of head and neck.
Figure 14.11â•…Restraint of the guinea pig. The body weight is supported with one hand, and the thorax is encircled from the front or rear.
84550.indb 410
11/11/10 6:26:57 PM
Common Nonsurgical Techniques and Procedures
411
Figure 14.12â•…Restraint of a rabbit for removal from a cage. The scruff and hindquarters are securely held to decrease the chance of injury.
or frightened animals can be extremely noisy and may injure themselves or handlers in attempts to flee restraint. Pig boards or similar large, flat objects may be used to herd a pig into a corner for a rapid procedure such as a quick injection. Soft snares or loop twitches may also be used around the upper jaw for such short procedures, although this method may be stressful (Figure€14.17). Longer procedures may require use of sedation. V-shaped troughs may be used to restrain small pigs in dorsal recumbency.
Figure 14.13â•…A secure restraint for carrying a rabbit. The rabbit’s head is tucked under the handler’s arm.
84550.indb 411
11/11/10 6:26:58 PM
412
Handbook of Laboratory Animal Science
Figure 14.14â•…A plastic rabbit restrainer.
Figure 14.15â•…A rabbit held in a cloth restraint bag designed for cats. Zippered openings in the bag allow access to the body.
Figure 14.16â•…A terry towel restraint permits access to the ears. The rabbit’s weight secures the towel. With any method of restraint, covering the eyes may calm the animal.
84550.indb 412
11/11/10 6:26:59 PM
Common Nonsurgical Techniques and Procedures
413
Figure 14.17â•…Restraint of a large pig with a soft loop twitch.
Use of slings is a humane way to restrain pigs for many procedures. Pigs are trained with treats to enter a sling readily and to remain calm while in the sling (Framstad et al. 1988). For heavier animals, slings are available that use cranks to raise the pig to table height after the pig has walked into the sling at floor level (Figure€14.18). Methods of Identification Temporary marking methods are those that are typically groomed off, rubbed off, or shed by an animal. These methods include use of colored pens on albino animals or light-colored parts of darker animals, livestock marking crayons, and some dyes. In rats, colored ink rings marked on the
Figure 14.18â•…A pig resting in a sling. (Photo courtesy of Panepinto and Associates, Colorado.)
84550.indb 413
11/11/10 6:27:00 PM
414
Handbook of Laboratory Animal Science
Figure 14.19â•…Tattoo clamp, numbers, and tattoo ink.
tail provide ready identification that lasts for several days, and the color can be reapplied as it fades. Shaving a patch of hair can also serve as a temporary identification. Color photographs may be used for identification of animals with distinctive patterns of markings on their coats. Tattoos, made with clamp tongs (Figure€14.19), tattoo pencils (Figure€14.20), or by piercing the skin with an ink-coated needle, permanently place dye into the skin. Tattooing is commonly used
(a) Figure 14.20aâ•…Tattoo pencil used on a rat tail. (Photo courtesy of AIMS, Inc., [email protected].)
84550.indb 414
11/11/10 6:27:01 PM
Common Nonsurgical Techniques and Procedures
415
(b) Figure 14.20bâ•…Adult rat with tail tattoo. (Photo courtesy of AIMS, Inc., [email protected].)
in rodents (ears, tails, or toes), rabbits (ears), and pigs (ears). Local anesthetic such as prilocainelidocaine cream, possibly supplemented with general sedation, or general anesthesia, is a humane consideration during tattoo placement. Tattooing is useful for marking neonatal animals because the technique is permanent, yet minimally invasive. A tattoo placed with correct technique using a tattoo pencil on a neonatal rodent tail can be clearly read throughout the animal’s life. Considerable skill is required to place the ink correctly into the dermal layer of the skin. Another possibility is to apply tattoo ink on the caudal surface of the toes of neonates, utilizing a small needle (Figure€ 14.20D). Removal of toes from neonates is a method of identification that is difficult to justify; tattooing is a strongly preferred method.
(c) Figure 14.20câ•…Neonatal rat with tail tattoo. (Photo courtesy of AIMS, Inc., [email protected].)
84550.indb 415
11/11/10 6:27:04 PM
416
Handbook of Laboratory Animal Science
(d) Figure 14.20dâ•…A color version of this figure follows page 336. Young mouse with toe tattoo. (Photo courtesy of M. Carlos Joao Castilhana.)
Ear tags, used in rodents, rabbits, and pigs, are generally placed at weaning. Correct placement decreases chances that the tag will cause pain or irritation to the animal. Tags placed too close to the ear margins are readily torn out, while those placed too deeply crimp the ear. Tags used in rodents should be placed hanging on the lower aspect of the ear, so that the ear is not bent from the weight of the tag when the animal’s head is held in a naturally erect position. In rabbits, tags are placed in the fold at the lower medial aspect of the pinna. Ear punches and ear snips may be used to identify pigs, mice, rats, and other rodents; in mice, the tissue may be used for genotyping instead of snipping off the tail tip. Ear snips are made in neonatal piglets. In rodents, punches or snips may be made at weaning or later. A numbering system is based upon the number and location of the holes or snips (top, middle, or bottom, and right or left ear) (Lawson 1999). Implantable microchips are approximately 1 cm in size. In rodents and rabbits, these chips are inserted subcutaneously using a trocar in the dorsal neck. Surgical glue may be used to seal the hole left by the trocar in small animals, such as mice, but in most cases this is hardly necessary. An electronic scanner is used to read the identification number unique to each chip. Though more expensive than other methods of identification, microchip systems can be integrated with data recording and census programs.
Administration of Substances Three methods for the administration of substances can be distinguished: via the skin, enteral, and parenteral.
84550.indb 416
11/11/10 6:27:05 PM
Common Nonsurgical Techniques and Procedures
417
Application to Skin or Mucous Membranes The substance is applied in solution or as an ointment on shaved skin or on mucous membranes. However, this method may be inaccurate and cause discomfort to the animal, especially when the substance causes irritation. Enteral Administration Substances administered enterally are delivered into the gastrointestinal tract orally. Although it is an animal-friendly method, including a substance in food or drinking water is not an accurate method and is difficult when the substance is unpalatable, insoluble, or chemically unstable. Medicines or other substances can be incorporated into commercial diets or treats, or they can be mixed with palatable substances to stimulate their consumption. Substance administration via a stomach/esophageal tube is more accurate and can be performed in mice and rats by a curved needle with a blunt end (external diameter 0.8 mm), such as an infant lacrimal sac cannula for mice. Mice and rats should be held firmly by the scruff while passing the needle along the palate into the esophagus. The animal should be allowed to “swallow” the needle. As the head is directed upward, the risk of entering the trachea is minimal. Fasting an animal prior to oral gavage may be appropriate, depending on the feeding pattern of the species and the time of day or light cycle. Rodents and rabbits practice cecotrophy and may not have an empty stomach even after fasting (Jeffery, Burrows, and Bye 1987). Gavage tubes come in flexible and rigid forms. Metal feeding needles with bulbed tips for rodents are available in straight, curved, and malleable forms and do not require use of a mouth gag (Figure€14.21). Rubber or vinyl tubes often require the use of a mouth gag to prevent the animal from chewing on the tube. The length of the needle or tube is determined by measuring the distance from the animal’s lips to its first rib (Figure€14.23). This distance should be marked on a long tube to guide correct placement depth. Once the tube or needle is in place, confirmation that the tube has not entered the respiratory tract is essential prior to administering the dose. With proper tube placement, the animal is observed to breathe, but there is an absence of air passage through the tube as the animal respires. Administration of large volumes can overload the stomach and may reflux into the esophagus or trachea, causing acute pneumonia, or may pass immediately into the small bowel. Table€14.3 gives
Figure 14.21â•…Oral gavage needles for rodents come in a range of gauges and lengths. All feature a probeended tip to reduce trauma and decrease likelihood of insertion into the respiratory tract.
84550.indb 417
11/11/10 6:27:06 PM
418
Handbook of Laboratory Animal Science
(a) Figure 14.22aâ•…Oral gavage in the rat using a metal gavage needle. Tipping the head back.
suggested volumes for gavage. In planning a dosing protocol, the frequency of dosing, volume, pH, tonicity, and vehicle (aqueous or oil) must be considered. Larger volumes in an aqueous vehicle are better tolerated than in an oil vehicle. Depending on the length of the needle or tube inserted, substances may be administered into the esophagus instead of the stomach. Animals are not anesthetized prior to dosing because maintenance of gag and swallow reflexes is important to detect proper tube placement and to decrease chances of reflux and aspiration after dosing. However, animals resisting the procedure may require light sedation. Gentle technique is essential because severe trauma may occur to the soft tissues of the mouth, pharynx, trachea, or esophagus if the technique is forced.
(b) Figure 14.22bâ•…A color version of this figure follows page 336. Oral gavage in the rat using a metal gavage needle. The correct needle length is the distance from lips to the first rib. (Photo courtesy of T. P. Rooymans.)
84550.indb 418
11/11/10 6:27:08 PM
Common Nonsurgical Techniques and Procedures
419
(c) Figure 14.22câ•…A color version of this figure follows page 336. Oral gavage in the rat using a metal gavage needle. The needle should slip its entire length without meeting any resistance. The needle must never be forced. (Photo courtesy of T. P. Rooymans.)
Rats Rats can be trained to accept routine dosing using equipment such as a probe-ended gavage needle or a rubber tube. The rat is manually restrained and held so that the rear legs cannot kick. Its head is tipped back with gentle pressure by the probe end of the needle on the roof of its mouth. When the animal is restrained properly by the scruff, the esophagus will be in the perfect position for the needle to pass without the risk of entering the trachea. The needle should easily slip its entire (premeasured) length into the esophagus. Ideally, the handler should observe the animal swallowing the tube or needle (Figure€14.22). If any resistance is met, the tube may be in the respiratory tract and should be retracted to the level of the mouth and reinserted. A technique for dosing rats using rubber tubing held in a ring stand has been described (Svendsen and Hau 1994). Capsule dosing of rats can be used for materials that are insoluble or form poor suspensions. A mini gel capsule (size 9, approximately 8.4 mm length, 2.6 mm diameter, with a capacity of 0.025 mL) is used for rats weighing ≥150 g. A capsule-dosing syringe is used to eject the capsule into the distal esophagus. Insertion of the dosing syringe follows the same technique described for dosing with liquids (De Brandt and Remon 1991). Mice Straight or curved probe-ended metal feeding needles are used for oral gavage in mice (Figure€14.23). The technique follows that described for rats. However, the swallowing reflex will not be as clear as in the rat. Hamsters The technique is similar to that described for the rat and the mouse but is considered more difficult in the hamster due to the position of the molars. A restraint grip that includes control of the head is necessary to perform oral gavage in hamsters.
84550.indb 419
11/11/10 6:27:09 PM
420
Handbook of Laboratory Animal Science
Figure 14.23â•…Oral gavage in the mouse using a metal gavage needle. Correctly placed, the needle is advanced without meeting resistance.
Guinea Pigs Guinea pigs can be dosed using metal probe-ended feeding needles or with a mouth gag and rubber tube, size 5 or 6 French (F). If a guinea pig becomes agitated by the gavage procedure, it may require light sedation (but not general anesthesia). The technique is similar to that described for the rat, but is considered more difficult in the guinea pig due to the position of the molars. A technique for capsule dosing of guinea pigs has been described. A mini gel capsule (size 9, approximately 8.4 mm length, 2.6 mm diameter, with a capacity of 0.025 mL) is used for guinea pigs weighing ≥300 g. A capsule-dosing syringe is used to place the capsule at the esophagus entrance. The capsule-dosing syringe is placed into the mouth using the same technique described for dosing with liquids, but the animal does not swallow the end of the syringe. The syringe is removed from the mouth, and the animal’s neck is gently stroked to encourage swallowing (Svendsen and Hau 1994). A technique for obtaining gastric juice samples has been described in which the guinea pig is held in a vertical position by an assistant, who uses a strip of gauze looped under the top incisor teeth to pull the animal’s head into a position of dorsoflexion. A second strip of gauze is used to control the lower jaw. A flexible feeding tube is used (Shomer et al. 1999). Rabbits Rabbits may be dosed using a rubber tube and mouth gag. The animal is restrained in a towel, cloth bag, or restraint box. Tipping the rabbit back may facilitate the procedure. A mouth gag can be fashioned from a 3 cc syringe barrel with small holes drilled directly across the middle of the barrel. The gag is positioned behind the front teeth, and the tube inserted through the drilled holes. Measuring the distance from mouth to stomach, which is about the last rib, provides information
84550.indb 420
11/11/10 6:27:10 PM
Common Nonsurgical Techniques and Procedures
421
Table€14.1╅Needle Sizes in Relation to Standard Wire Gauge External Diameter Metric Gauge (mm) 0.25 0.35 0.40 0.45 0.50 0.55 0.65 0.70 0.80 0.90 1.10 1.25 1.45 1.65 1.80 2.10 2.40 2.80 3.00 3.25 3.65
Standard Wire Gauge 30 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9
on how far the tube should be inserted without meeting any resistance as it is passed. The external diameter should be 8–10 F. Pigs Pigs can often be induced to take fluids from a syringe inserted into the side of the mouth. The pig can be held in a sling or restrained with a soft snare. Well-socialized pigs under 25 kg can be restrained in a “bear hug”: held with the pig’s back against the handler’s chest. Pigs can be trained to accept passage of a rubber stomach tube while restrained in a sling or held in a hug. A 30 French size tube is appropriate for pigs weighing 15 kg (Table€14.2). The soft tube is inserted until the pig swallows the end. Listening for stomach noises at the free end of the tube can help confirm proper placement. A mouth gag is not used unless the pig is resistant to the procedure. Alternatively, a long metal feeding needle designed for large rats can be used for small pigs. Parenteral Administration Substances administered parenterally are generally given by means of an injection, and several factors are considered when selecting needle diameter and length. A smaller needle causes less damage and pain when inserted into tissue. Larger diameter needles are advantageous in permitting large volumes to be given rapidly, and they facilitate passage of viscous substances. When administering blood or cells, cell damage is less likely if a large-bore needle is used. However, diameter is also limited by the size of the animal and, in particular for intravenous injections, by the size of the selected vein (Table€14.1). Recommended needle sizes for specific species and routes of injection are given in Table€14.3.
84550.indb 421
11/11/10 6:27:10 PM
422
Handbook of Laboratory Animal Science
Table€14.2╅ Catheter Dimensions in Relation to French Size External Diameter Metric Gauge (mm) 0.33 0.67 1.00 1.33 1.67 2.00 2.33 2.67 3.00 3.33 3.67 4.00 4.33 4.67 5.00 5.33 5.67 6.00 6.33 6.67 7.00 7.33 7.67 8.00 8.33 8.67 9.00 9.33 9.67 10.00 10.33 10.67 11.00 11.33
French Gauge/Charrièr No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Other injection considerations include the pH, tonicity, and temperature of the substance being administered (Merck Veterinary Manual 2005). Extremes in pH cause local tissue damage; ideally, solutions should be at pH = 7.4. Tonicity refers to the concentration of solutes that affects the movement of water in or out of cells. Isotonic solutions such as 0.9% saline or 5% dextrose are often used for injection. Hypertonic solutions are most commonly administered by the intravenous route in a large vessel to promote rapid dilution and avoid irritation to the vessel wall. Solutions for rehydration or any large volume to be administered should be warmed to prevent inducing hypothermia, particularly in small animals. Diehl et al. (2001) have reviewed best practices for substance administration. Warming an animal or the extremity to be injected induces vasodilation, making the blood vessels easier to see and to access. Rate of intravenous infusion can vary from a bolus to slowly over
84550.indb 422
11/11/10 6:27:10 PM
84550.indb 423
0.25 1 0.5 1
4 25
26 25 26 25
23 22
Volume (mL)
23 20
25 25 25 25 20 25
0.5 2.0 1 2
Volume (mL)
Intraperitoneal Needle Gauge
25 23
27 25 26 25 1 6.25
0.05/site 0.1/site 0.05/site 0.1/site
Volume (mL)
Intramuscular Needle Gauge
23 23
26 25 27 26 5 62.5
0.2 1 0.3 1
Bolus Volume (mL)
Intravenous Needle Gauge
40 (10 mL/kg) 125 (5 mL/kg)
0.3 (12 mL/kg) 4 (20 mL/kg) 0.5 4
Slow Infusion Volume (mL)
Oral
13 or 8.Fr 22
20 18 20 18
Needle Gauge or French Tube Size
7.55 150
0.25 1 0.5 1
Volume (mL)
Note: Volume refers to good practice for single or multiple doses. Prior to redosing, consider time for absorption of prior dose, particularly for substances in nonaqueous base or irritating substances. Multiple sites may be used. Bolus IV injection rate for rodents and rabbits is about 3 mL/min. Slow IV infusion rate is over the course of 5–10 min or longer. Tolerance of IV injection is highly dependent on the vehicle used. Oral dosing volume is dependent on whether the stomach is empty. Oilbased substances are less well tolerated in large volume than are aqueous-based substances. Intracutaneous injection volume: 0.05–0.1 mL per site for all mentioned species using a 26 G needle.
Mouse, 25 g Rat, 200 g Hamster, 30 g Guinea pig, 200 g Rabbit, 4 kg Pig, 25 kg
Species
Needle Gauge
Subcutaneous
Table€14.3╅Recommended Needle Gauge and Dosing Volume for Different Routes of Administration
Common Nonsurgical Techniques and Procedures 423
11/11/10 6:27:11 PM
424
Handbook of Laboratory Animal Science
Figure 14.24â•…Intravenous injection in the rat tail.
several minutes to continuous over several hours or longer. Solutions of high or low pH and nonisotonic solutions are best tolerated as slow or continuous infusions. Adequate restraint may need to be augmented with local or general anesthesia to ensure accurate needle placement with minimal tissue trauma and pain for the animal. For example, prilocaine + lidocaine topical cream may be applied to rabbit ears over the vessels 15 min prior to the cannulation, to numb the site. Butterfly needles (needles with attached flexible tubing) are advantageous in many situations. Once a butterfly needle has been seated in a blood vessel and taped or glued in place, the flexible tubing permits the animal some freedom from complete restraint during the injection period. Rats Subcutaneous injections are made in the scruff or flank. Rats may be restrained manually or wrapped in a towel. The skin is gently raised or “tented.” The needle is inserted at a shallow angle, parallel to the animal, and is directed away from the fingers of the handler. Gently lifting the inserted needle to raise the skin confirms correct placement in the subcutis. Intravenous injection sites include the lateral tail veins, lateral saphenous veins, dorsal metatarsal veins, tongue veins, and penis vein. For access to the tail and leg veins, restraint is achieved with the use of a rigid restraint tube, towel, cloth, stockinet, or plastic cone (Figure€14.24). Topical anesthetic cream has not been found to be effective in decreasing pain response to tail stick in the rat (Flecknell, Liles, and Williamson 1990). General anesthesia is necessary for access to the veins of the tongue or penis. Gently warming the rat or its tail with a heating pad, heat lamp, or hot-water bag will dilate the vessels, but may induce physiological changes in the animal. Warming only the tail using warm water will produce a short, local vasodilation. Some rats have a buildup of reddishbrown scale stain on the tail, which may obscure the blood vessels. The stain can be cleaned off by gently rubbing the tail with a gauze sponge moistened with rubbing alcohol. The needle is inserted at a very shallow angle. Correct placement can be confirmed by aspiration of blood. For repeat intravenous injections over a short period, catheters can be placed into the tail vein. For repeated access over several days, surgical implantation of a jugular catheter is recommended. Injection via the retro-orbital sinuses is controversial because the retro-orbital tissue may be readily damaged, leading to formation of a hematoma. Retro-orbital injection is made in the anesthetized rat. The needle is directed under the lid and behind the eye at the dorsal aspect or at the medial canthus, into the retrobulbar sinus. Aspiration of a small volume of blood confirms correct placement. After the needle is removed, the lids are held shut with a dry gauze sponge for several seconds, to achieve hemostasis. There is an attendant risk of hematoma or ocular damage with this technique.
84550.indb 424
11/11/10 6:27:11 PM
Common Nonsurgical Techniques and Procedures
425
Figure 14.25â•…Use of a towel for restraint of the rat for intraperitoneal injection in the lower half of the abdomen.
Intramuscular injections are made in the quadriceps muscles (anterior aspect) of the thigh. The posterior muscles of the thigh may be used, with caution, because the sciatic nerve passes through this muscle group. Firm restraint or anesthesia is necessary to preclude the leg from kicking and to prevent the needle from tearing and bruising the muscles during the injection. Care is taken to visualize the depth at which the needle is placed, ideally centering the tip near the middle of the muscle group, with a short needle providing better control. Restraint may be achieved with the use of a cloth or towel (Figure€14.25). Intraperitoneal injections are made in the lower half of the abdomen, off the midline. One study suggests that injection into the right side of the abdomen decreases chances of damage to the cecum (Coria-Avila et al. 2007). Animals can be restrained manually, beneath or in a towel, or with the upper half of the body in a rigid tube (Figure€14.25). The needle is directed at a shallow angle. Once the needle is in the abdomen, but prior to injection, aspiration will help confirm correct needle placement. The plunger is gently pulled back to check for the influx of urine, blood, or intestinal contents into the syringe. Should aspiration confirm that the needle has entered the bladder, a blood vessel, or the gut, the needle is removed from the animal and the entire procedure is begun again with a fresh needle and syringe. For intradermal (intracutaneous) injections, the site of the injection must first be shaved. The skin is held between two fingers, and the needle delicately advanced into the cutis. A bleb (swelling) will be formed on the skin as the injectate enters. Local or general anesthesia is generally required for proper intradermal injection technique and humane considerations. Intracardiac injections are performed when the rat is in a deep plane of anesthesia. This technique is most commonly performed as a terminal (nonrecovery) procedure. The heartbeat can be palpated by placing fingers on either side of the chest, at the level of the rat’s elbows. The needle is directed into the heart from one of several possible sites. With the animal positioned in dorsal recumbency, a longer needle may be directed on the midline from below the sternum, through the diaphragm, at about a 45° angle toward the back. Alternatively, the animal can be placed on its side and the needle directed at a 90° angle from the side of the chest just behind the elbow, between the ribs (Figure€ 14.26). With any approach, a single well-controlled, quick thrust of the needle is needed to pierce the heart. Rapid flow of blood into the syringe confirms correct placement. Repeated attempts to place the needle increase the likelihood of causing severe damage to the lungs, heart, or blood vessels within the chest.
84550.indb 425
11/11/10 6:27:12 PM
426
Handbook of Laboratory Animal Science
Figure 14.26â•…One technique for intracardiac injection in the rat.
Mice Subcutaneous injections are made along the flank, scruff, or behind the elbow. The mouse is restrained by the scruff and tail, and the needle is inserted at a shallow angle (Figure€ 14.27). If the scruff is used, the needle is directed parallel to the animal to avoid the cervical vertebrae and away from the handler’s fingers. Gently lifting the inserted needle to raise the skin confirms correct placement in the subcutis. Intravenous injections can be given via the lateral tail veins. Injection via the retro-orbital sinuses is controversial because the retro-orbital tissue may be easily damaged, leading to formation of a hematoma (see preceding section). Gently warming the mouse with a heat lamp or pad will cause vasodilation of the tail vessels (see preceding section). Use of lidocaine-prilocaine cream does not decrease reaction to tail stick in the mouse (Flecknell et al. 1990). For tail vein injections, mice are
Figure 14.27â•…Subcutaneous injection in the mouse. (Photo courtesy of T. P. Rooymans.)
84550.indb 426
11/11/10 6:27:15 PM
Common Nonsurgical Techniques and Procedures
427
Figure 14.28â•…A technique for restraint of the mouse for tail vein injection.
held in a rigid restraint device (Figure€14.28). The needle is inserted at a very shallow angle, almost parallel to the tail, just far enough into the tissue to seat the needle past the bevel. Aspiration of blood to confirm needle placement is not productive in mice. Pushing a small volume of injectate can confirm correct placement. If the needle is correctly seated in the vein, the vessel will appear to clear as the injectate flows through. If the needle is outside the vein, a bleb (swelling) will appear. Accessing the vessel is easiest about two-thirds of the distance down the tail (one-third of the way up from the tail tip). Intramuscular injection, a delicate procedure in the mouse, is made in the quadriceps muscle. Restraint technique should include holding the leg to be injected to prevent it from kicking during the injection. The foot may be secured alongside the tail beneath the fingers holding the tail. Care is taken to visualize the depth at which the needle is placed, ideally centering the tip near the middle of the muscle group and preventing damage to the knee joint. This route is painful, somewhat risky, and only a small volume can be injected; therefore, other routes are preferable. Intraperitoneal injection is made in the lower half of the abdomen, off the midline. The mouse is restrained manually (Figure€14.29). The technique is as described for the rat, except that the head
Figure 14.29â•…Intraperitoneal injection in the mouse.
84550.indb 427
11/11/10 6:27:16 PM
428
Handbook of Laboratory Animal Science
Figure 14.30â•…Subcutaneous injection in the guinea pig.
of the mouse should be tilted downward in order to avoid damaging the liver. Intradermal injections follow the technique described for the rat. Intracardiac injections are performed as in the rat, with the mouse in a deep plane of anesthesia. Hamsters Subcutaneous, intradermal, intramuscular, intraperitoneal, intracardiac, and retro-orbital injections are performed as in rats and mice. Intravenous injections can be difficult in the hamster because superficial veins are not easily accessible. The lateral saphenous veins can be accessed in the same way as in rats and mice. Guinea Pigs The scruff or flank is generally used as the site for subcutaneous injections. The guinea pig is held in a towel or on its abdomen, gently pressed on a firm surface. A tent of skin is made, and the needle is directed away from the handler’s fingers (Figure€14.30). Intravenous access can be challenging in the guinea pig. Sites include the vessels of the ear (although they are very small; see Figure€14.31), the lateral pedal veins, dorsal metatarsal veins, and penis vein. General anesthesia is necessary for access to the penis vein. Warming the animal gently with a heating pad or warm towel will dilate blood vessels. The quadriceps muscles are the preferred site for intramuscular injections. The guinea pig can be restrained in a towel with the rear leg extended, or the animal can be held between the handler’s arm and chest. The leg must be held firmly to prevent kicking during the injection. Care is taken to visualize the depth at which the needle is placed, ideally centering the tip near the middle of the muscle group. Intraperitoneal injection is accomplished as described for the rat. The guinea pig is restrained manually or wrapped in a towel. Intradermal and intracardiac injections are as described for the rat. Rabbits The scruff is the most common site for subcutaneous injection in the rabbit, although any area of loose skin over the back or flanks may be used. If the substance to be injected is an irritant—for
84550.indb 428
11/11/10 6:27:17 PM
Common Nonsurgical Techniques and Procedures
429
Figure 14.31â•…The ear vessels of the guinea pig.
example, antigen plus adjuvant for antibody production—the scruff should not be used as an injection site, since the scruff is routinely grasped to restrain the animal. The skin is raised or tented, and the needle inserted pointing away from the handler’s fingers. The lateral ear veins are the most readily accessible site for intravenous injection in the rabbit. Warming the rabbit or the ears and gently rubbing the ear will cause vasodilation. A subcutaneous injection of phenothiazine tranquilizer (acepromazine maleate at 0.8 mg/kg) or an opioid combination (fentanyl/fluanisone [Hypnorm] at 0.2 ml/kg given i.m.) 15 min prior to injection will calm the rabbit and cause dilation of the blood vessels. Topical anesthetic (lidocaine-prilocaine cream), generously applied 15 min prior to injection, will effectively numb the ear vessels (Flecknell et al. 1990). Rabbits may be restrained for injection in any number of rigid or cloth restraint devices or wrapped in a towel (Figures€14.13–14.15). The ear veins are very superficial to the skin. Using a butterfly needle assists in needle insertion at a very shallow angle, almost parallel to the ear surface (Figure€14.32). Small catheters may be inserted and taped or glued in place for repeat injections or infusions. Intramuscular injection sites include the quadriceps muscles of the anterior thigh and lumbar muscles (Figure€14.33). The rabbit is restrained on a tabletop with its head tucked under the elbow of the handler. The quadriceps is readily palpated above the femur, and the leg is held steady with a hand cupped around the rabbit’s stifle. The lumbar muscle sites are on either side of the spinal column. The needle is placed parallel to the spinal column, in the muscles between the caudal-most rib and the anterior aspect of the ilium (point of the hip). Intraperitoneal injections are made in the lower half of the abdomen, off the midline. The rabbit is held by the scruff and raised to an upright position on the handler’s lap. If the rabbit is agitated, an assistant may give the injection while the handler grasps the rabbit’s feet and scruff. Alternatively, a cloth bag or towel may be used to restrain the rabbit. Aspiration is recommended to assist in confirming correct needle placement. Intradermal injections follow the techniques described for the rat. Intracardiac injections are made when the rabbit is in a deep plane of anesthesia. This is most commonly a terminal (nonsurvival) procedure for the rabbit. The rabbit is placed in dorsal recumbency. The needle is directed from the lateral aspect of the chest at the level of the rabbit’s elbow, between the ribs. The needle is advanced with a smooth thrust to ensure that the heart is pierced. Blood will pulsate back into the needle once it enters the heart. Use of a butterfly needle facilitates the procedure.
84550.indb 429
11/11/10 6:27:18 PM
430
Handbook of Laboratory Animal Science
(a)
(b) Figure 14.32â•…The ear vessels of the rabbit. The central ear artery can be seen as it runs from base to tip down the center of the ear. (a) Injection into the lateral ear vein. A 25-gauge butterfly needle is positioned flat along the ear. (b) Blood flow into the tubing confirms proper needle placement.
Figure 14.33â•…Intramuscular injection in the lumbar muscles of the rabbit.
84550.indb 430
11/11/10 6:27:21 PM
Common Nonsurgical Techniques and Procedures
431
Pigs The moderately loose skin over the dorsal neck area caudal to the ears is the primary site for subcutaneous injections in the pig. Depending on the volume and “sting” of the injectate, the animal can be completely restrained or the injection made via a small butterfly needle while the pig is confined in a relatively small space. The vessels of the ear are the primary site for intravenous injections in the pig. Use of a small butterfly needle is recommended, as is the use of topical anesthetic cream (lidocaine-prilocaine) applied 15 min prior to injection to numb the site. Gently warming or rubbing the ear can dilate the vessels. Butterfly needles or catheters can be taped in place for repeated injections or longer infusions (Framstad et al. 1988). Intramuscular injection sites in pigs include the quadriceps muscles of the hind leg and the muscles on the dorsal lateral part of the neck. The pig can be restrained in a sling or with the use of a soft snare or twitch. If a butterfly needle is used, the pig may be allowed to move about in a small area, while the person giving the injection seats the needle, then walks beside the pig while giving the injection (Swindle 2007). Intraperitoneal injections are made in the lower half of the abdomen. The pig can be restrained in a sling or with a snare. Aspiration is used to assess needle placement prior to the injection. Intratracheal and Intrabronchial Instillation Intratracheal and intrabronchial instillation are accomplished in anesthetized animals. Small volumes are injected directly into the trachea via the pharynx. Once anesthetized, a small rodent can be positioned on its back, held against a steeply slanted support surface. A string or rubber band is attached at the top of the support on one end. The animal is suspended head-upward by the string or rubber band, which is positioned behind the upper incisors on the other end. The tongue and lower jaw are gently pulled down to expose the back of the mouth. A bright light source is directed at the throat to transilluminate the trachea. An otoscope may be used to assist with visualization to allow tubing to be placed into the trachea or the trachea may be pierced from the ventral aspect of the neck, although this engenders greater risk for the animal. A lit fiber optic cable may be used as a stylet to permit visualization for tubing as it is placed into the trachea (MacDonald et al. 2009). Intrabronchial administration is accomplished after passing a small-gauge flexible plastic tube through an endotracheal tube into the lung (Brown et al. 1999). A sterile fiber optic bronchoscope passed through the endotracheal tube may be used to guide catheter placement. Blood Sampling Blood samples can be obtained from various sites of the body, using a variety of methods: from the veins, from the arteries, by puncturing the orbital vessels, by cardiac puncture, or by decapitation. The choice of method depends on several factors, including purpose of the blood collection, need for arterial or venous sample, duration and frequency of sampling, and whether the experiment is terminal for the animal. In small species, blood is often collected with the animal under anesthesia, to assist in immobilization and to decrease distress to the animal. When selecting a method for blood sampling in the conscious animal, consideration must be given to the potential for stressinduced effects on physiological and behavioral parameters. Blood volume is estimated based on body weight. Several papers have reviewed recommendations on how much blood can be removed over a given time interval without causing abnormal
84550.indb 431
11/11/10 6:27:22 PM
432
Handbook of Laboratory Animal Science
physiologic response and distress in animals (Joint Working Group on Refinement 1993; McGuill and Rowan 1989). A typical rule is not to exceed removal of 10% of blood volume (roughly 1% of body weight or 8 mL/kg) in any given 2-week period. For example, this would equate to removal of 3 mL from a 300 g rat, 0.2 mL from a 20 g mouse, or 35 mL from a 3.5 kg rabbit. Removal of larger volumes necessitates longer recovery periods for the animal and may cause distress or even death. In rabbits, plasmapheresis (after centrifugation injecting the red blood cells again) may allow for the collection of larger total volumes of blood. Exsanguination is only performed on deeply anesthetized animals. One can expect to obtain approximately half the animal’s total blood volume at exsanguination via cardiac puncture. For example, one can obtain approximately 9 mL from a 300 g rat, 0.6 mL from a 20 g mouse, or 115 mL from a 3.5 kg rabbit. Exsanguination is also possible in small rodents by removing the eyeball under anesthesia and collecting blood from the eye artery. Blood can also be collected after decapitation using a guillotine or scissors, but the handler must be competent and well trained at restraint to complete the procedure safely and humanely. Sharp blades on scissors and guillotines are required to perform these techniques properly. More precise techniques, such as a cut-down and cannulation of the carotid artery in the neck with infusion of fluids via the contralateral jugular vein or an ear vein, or aorta puncture, can maximize the blood volume collected at exsanguination. Infusion of fluids may influence blood parameters, however. Local anesthesia, sedation, and general anesthesia are refinements that improve success and decrease stress for many blood-draw techniques. For repeated sampling over long periods, surgical placement of indwelling catheters is recommended. To attain multiple samples at short intervals, catheters or butterfly needles can be taped or glued in place. Once the needle has been removed from the blood vessel, light pressure with a dry gauze sponge is applied until hemostasis is confirmed. Rats For blood withdrawal from the lateral tail vein, the rat is held in a restraint device (rigid tube, towel, or bag). If the tail has a buildup of reddish brown scale stain, a gauze sponge moistened with rubbing alcohol may be used gently to rub away the stain and to permit visualization of the blood vessels. The tail may be dipped in 45°C water for about 1 min to induce vasodilation. The vein is most readily accessed about one-third of the length up from the tip. Subsequent venipunctures are made ascending the tail. A 25-gauge needle or butterfly needle is inserted at a very shallow angle. Blood flow may be augmented by very gently milking with the fingers from the base of the tail toward the tip. This technique is typically used for small volumes of blood, up to about 1 mL. The ventral tail artery accompanied by a small vein is accessed with the rat placed in dorsal recumbency. Using a finger, pressure is applied about 5 cm from the tail tip to raise the vessel. The artery is then punctured near the base of the tail, and blood is collected. The technique may be performed in animals that are awake if an assistant is available to hold the rat firmly wrapped in a towel. However, anesthesia of the rat assists with this technique; this eliminates struggling against restraint and eases access to the blood vessel and hemostasis. The dorsal metatarsal or lateral saphenous veins on the lateral aspect of the rear leg may be punctured for blood collection. The rat is held wrapped in a towel or plastic cone or stockinet with the leg extended. Bland ointment (petrolatum) applied to the leg will slick back the hair to permit visualization of the vessel and also will prevent blood from soaking into the hair. Alternatively, the site may be shaved. The loose skin of the thigh is gently gathered to raise the blood vessel (Figure€14.34). The vessel is punctured, and blood is collected in a capillary tube or test tube (Hem, Smith, and Solberg 1998). A technique for blood draw from the jugular vein in the unanesthetized rat requires the participation of two people. A handler places the animal in dorsal recumbency on a board designed
84550.indb 432
11/11/10 6:27:22 PM
Common Nonsurgical Techniques and Procedures
433
(a)
(b) Figure 14.34â•…Dorsal metatarsal (a) and lateral saphenous (b) venipuncture in the rat.
with a flat level surface that ends in a 45° slant so that the rat’s head can be tipped below the body. The forepaws are secured straight out to the sides with soft rope ties. The person taking the blood sample manipulates the rat’s head downward and to the side, using a plastic cup held over the head. A needle with syringe attached is inserted 1 cm lateral to the midline, to about a 1 cm depth, and the sample is collected. Digital pressure is applied immediately after the needle is removed to aid in hemostasis. Restraint in this manner is stressful to the animal, and the team performing the technique must be well practiced and coordinated to complete the procedure rapidly.
84550.indb 433
11/11/10 6:27:24 PM
434
Handbook of Laboratory Animal Science
In the anesthetized rat, the jugular vein is accessed with the rat placed in dorsal recumbency on a flat surface. The syringe is kept parallel to the surface and inserted about 1 cm lateral to the midline, to a depth of about 1 cm. Gentle suction is maintained in the syringe, which is slowly withdrawn from the animal until blood is seen to flow into the syringe. Alternatively, a small incision in the neck permits direct visualization of the jugular vein. Tail tip removal has been justified in some protocols where multiple short-interval, small-volume samples are required. The last 0.5–1 mm of the tail is snipped off with iris scissors. Blood is gently milked from the cut, and then pressure is applied with gauze to achieve hemostasis. Repeat samples are obtained by removing the clot. Volumes of 0.1–0.2 mL are obtained. Blood may be obtained from the retro-orbital plexus in the anesthetized rat. The rat is placed in lateral or ventral recumbency. A Pasteur pipette or microhematocrit capillary tube is passed beneath the upper lid at the medial canthus (inside corner) of the eye. The tube is gently pushed and twisted until it penetrates the conjunctiva. Without removing the tube from beneath the lid, the tube is gently retracted until blood flows into the tube. Once the flow is started, tilting a Pasteur pipette can allow gravity to assist. When the sample has been obtained, the tube is removed from the eye and the lids are immediately held closed with a dry gauze sponge for several seconds, to achieve hemostasis. This technique is used to obtain amounts of blood from 0.5 to 3 mL. Contamination with tissue fluids and porphyrins from the Harderian gland can occur. It is not possible to take sterile blood samples using this method. Also, complications such as hemorrhage, inflammation, and blindness may occur, especially when the same eye is used repeatedly. Moreover, this technique may be aesthetically unpleasant for the operator to perform. For these reasons, this method is not considered acceptable in some countries (Baumans 2001). However, the extent of trauma is proportional to the skill of the technician (Van Herck 2001). Cardiac blood withdrawal is performed when the rat is in a deep plane of anesthesia, and it is generally a terminal (nonrecovery) procedure. The technique is as described for intracardiac injection in the rat. Alternatively, the chest may be opened and the heart directly visualized for accurate direction of the needle. Arterial blood can be obtained from the abdominal aorta, brachial artery, or carotid artery under anesthesia. Mice Blood may be collected from the lateral saphenous vein of the unanesthetized mouse. The mouse is held head-first in a restraint device so that only the rear legs and tail are free. A rolled paper towel may be used for restraint, or a 50 cc plastic centrifuge tube may be used if holes are cut to permit air flow into the tube. The skin on the upper thigh is gently but firmly squeezed by the handler, using the same hand that is holding the tube. This serves to secure the mouse in the tube and raises the vein. Bland ointment (petrolatum) applied to the leg will slick back the hair to permit visualization of the vessel and also will prevent blood from soaking into the hair. Alternatively, the site may be shaved. Using a 25-gauge needle, the vessel is punctured at the most proximal visible aspect, and blood is collected as it wells up. A dry gauze sponge is used to apply pressure to the puncture site, and the pressure on the upper thigh is released. For repeat samples, the scab may be brushed off with a dry gauze sponge. A volume of 0.2 mL may be readily collected with this technique (Hem et al. 1998). Blood may be collected from the retro-orbital sinus of the mouse. This technique is performed with the mouse under anesthesia. The technique is as described for retro-orbital plexus sampling in the rat. Volumes of up to 1% of body weight are readily and rapidly obtained with this method (Figure€14.35). There is greater danger of corneal or ocular damage with this technique than in the rat.
84550.indb 434
11/11/10 6:27:24 PM
Common Nonsurgical Techniques and Procedures
435
(a)
(b) Figure 14.35â•…Retro-orbital sinus puncture in the mouse. (a) The microhematocrit tube is placed under the lid and gently twisted until it penetrates the conjunctiva. (b) Blood is collected by capillary action.
Cardiac blood withdrawal is performed when the mouse is in a deep plane of anesthesia, and it is generally a terminal (nonrecovery) procedure. The technique is as described for intracardiac injection in the rat. Alternatively, the chest may be opened and the heart directly visualized for accurate direction of the needle. Arterial blood can be obtained from the abdominal aorta, brachial artery, or carotid artery under anesthesia. Approximately 0.2 mL of blood may be collected after puncture of the facial vein of the mouse. The mouse is restrained by the scruff. A 23-gauge needle or a lancet of 0.3–0.5 mm length is used to quickly puncture the vessel, which is located just caudal and dorsal to a prominent whisker follicle on the lateral aspect of the mouse’s face. Blood flow begins immediately upon vessel puncture and is collected in a small tube. Blood flow must be stopped with gentle pressure with a gauze sponge (Golde, Gollobin, and Rodriguez 2005). Hemorrhage or damage of the facial nerve and artery may occur. Hamsters Blood may be collected from the lateral saphenous vein of the unanesthetized hamster as described for the mouse. In the anesthetized hamster, blood collection from the aorta or via cardiac puncture or from the retro-orbital sinus is possible as described for the mouse.
84550.indb 435
11/11/10 6:27:27 PM
436
Handbook of Laboratory Animal Science
Guinea Pigs Small amounts of blood can be obtained from the unanesthetized guinea pig. With the animal restrained in a towel, the lateral pedal vein on a rear foot can be punctured with a small (25-gauge) needle, and blood can be collected into a microhematocrit tube or small test tube (Figure€14.36). The saphenous vein can also be used. In larger guinea pigs, small samples can be obtained from the lateral ear vessels (Figure€14.27). These vessels are quite small, and warming the ear may stimulate vasodilatation. A technique has been described for collection of blood from the jugular vein of mesmerized guinea pigs (Shomer et al. 1999). In the anesthetized guinea pig, blood collection from the jugular vein or aorta or via cardiac puncture is as described for the rat. Rabbits The rabbit ear provides an easily accessible site for blood collection. Blood may be removed from the lateral ear vein or central ear artery. The artery is easy to cannulate and is the suggested site for volumes of more than 1 mL. Warming the rabbit or the ears will cause vasodilatation. A subcutaneous injection of phenothiazine tranquilizer (acepromazine maleate at 0.8 mg/kg) or an opioid combination (fentanyl/fluanisone [Hypnorm] at 0.2 mL/kg i.m.) 15 min prior to injection will calm the rabbit and cause dilation of the blood vessels. Topical anesthetic (lidocaine-prilocaine cream) generously applied 15 min prior to blood withdrawal will effectively numb the ear vessels (Diehl et al. 2001). Rabbits may be restrained in any number of rigid or cloth restraint devices or wrapped in a towel. The ear vessels are superficial to the skin. Use of butterfly needles assists in needle insertion at a very shallow angle, almost parallel to the ear surface (Figure€14.37). Correct placement of the needle into the artery is confirmed by a pulsing blood flow back into the tubing. Small catheters may be inserted and taped in place for repeat sampling. Hemostasis after an arterial stick must be carefully observed and may require pressure with a dry gauze sponge for 60 sec or longer. Cardiac puncture, typically as a terminal (nonsurvival) procedure for exsanguination, is performed with the rabbit in a surgical plane of anesthesia. A large needle attached to flexible tubing is used to pierce the heart. With the rabbit in dorsal recumbency, the needle is directed from the level of the elbow, between the ribs, to the center of the chest. The needle is inserted with a firm thrust, so as to pierce the heart. A pulsing flow of blood into the attached tubing confirms correct placement. If blood does not immediately flow, slight redirection of the needle may assist. However, repeated probing in the chest increases the likelihood of killing the animal prior to obtaining the sample. If blood does not appear, the needle should be removed completely and a new attempt made with a fresh needle. An alternative method involves opening the skin of the ventral neck to visualize the carotid artery, which is then cannulated for blood removal. A high-rate flow of saline into the contralateral jugular vein or an ear vein during the exsanguination procedure maintains blood pressure for a longer time and will allow more blood to be collected from the animal. Fluid administration may affect blood parameters. When blood flow ceases, the animal should be euthanatized with a barbiturate injection or by another approved method. Pigs Common sites for blood removal from the pig include the ear veins, jugular veins, and anterior vena cava. Occasionally, the cephalic vein is used, most often in smaller pigs. Often, untrained pigs must be sedated or anesthetized to permit ear vessel cannulation. Sedation has the advantage of promoting vasodilatation of the vessels, depending on the medication used. However, pigs trained to rest in a sling may permit vessel cannulation, particularly if the site is first numbed with topical
84550.indb 436
11/11/10 6:27:27 PM
Common Nonsurgical Techniques and Procedures
437
(a)
(b) Figure 14.36â•…Blood collection from the lateral pedal vein of the guinea pig. The guinea pig is restrained in a towel. (a) A 25-gauge needle is threaded into the vein. (b) Blood is collected into a hematocrit tube placed into the needle hub.
84550.indb 437
11/11/10 6:27:31 PM
438
Handbook of Laboratory Animal Science
(a)
(b)
Figure 14.37â•…Blood collection from the central ear artery of the rabbit, using a butterfly needle. (a) The needle is inserted at a very shallow angle. (b) Pulsing blood into the tubing confirms correct placement.
84550.indb 438
11/11/10 6:27:35 PM
Common Nonsurgical Techniques and Procedures
439
anesthetic (lidocaine-prilocaine cream). Small catheters or butterfly needles may be taped or glued into place for long-term or repeat sampling in the restrained pig. Puncture of the jugular vein may be done in the anesthetized pig or in an unanesthetized pig restrained in a sling or held by a soft rope snare around the upper jaw (Framstad et al. 1988). The pig’s head is pulled up and back. Most pigs will pull backward from a snare, so the assistant holding the snare should stand in front of the animal. The needle is directed caudodorsally in the deepest point of the jugular groove formed between the medial sternocephalic and brachiocephalic muscles. Use of a vacuum blood collection tube facilitates the technique. The collection tube is advanced onto the needle once the needle has been placed into the neck. Good restraint is essential to prevent laceration of the vessel (Framstad et al. 1988; Swindle 2007). Milk Collection A milk collection apparatus can be made for use in rodents (Figure€14.38). Size of the equipment will vary with the size of the animal. A large glass tube is fitted with a two-hole rubber stopper. A T-shaped glass tube and a straight glass tube are placed through the stopper. The straight tube is placed into a collection tube within the larger glass tube, and the other end of the straight tube is placed over a teat. One end of the T-shaped tube is connected to a vacuum, with between 250 and 300 mm Hg suction pressure. The other end of the T-shaped tube is intermittently occluded with the operator’s finger to obtain an interrupted vacuum on the teat, to mimic sucking action. Gently massaging the mammary glands toward the teat during milking can assist flow. In rats, up to 7 cc of milk can be obtained per day.
Figure 14.38â•…Milk collection equipment for rats. The T-shaped glass tube is connected to a vacuum pump via the rubber tubing. The open limb is operated with a fingertip to create intermittent vacuum, simulating sucking. The open, straight, glass pipe through the stopper ends in a small collecting tube inside the large glass tube, and its outer end is applied to the teat.
84550.indb 439
11/11/10 6:27:36 PM
440
Handbook of Laboratory Animal Science
A discussion of practical aspects of milk collection in rats has been published (Rodgers 1995). Milk collection can be accomplished in rabbits during days 5 through 25 postpartum, at peak milk production. Does typically nurse their young once per day, most often in the morning, so milk should be collected prior to allowing the doe to nurse. A description of a rabbit milk collection unit has been published (Marcus, Shum, and Goldman 1990).
Bone Marrow Collection Bone marrow collection is accomplished while the animal is under anesthesia. For survival procedures, the site is shaved and prepped, as for surgery. In rodents and rabbits, the tibia is the most common collection site. The knee joint is flexed, and a needle is inserted through the trochlea into the marrow cavity, parallel to the long axis of the bone. For mice, a 25-gauge needle is gently yet firmly pushed and twisted into the marrow cavity. Somewhat larger needles may be used in other species, depending on the size of the bone. A low-speed dental drill may be used to create a hole for needle insertion. Up to 1 mL of marrow may be obtained from the tibia of a rabbit. Preemptive and postprocedural analgesia is recommended. For pigs, the long bones, pelvis, or sternum are the sites for marrow collection. Consult a standard veterinary text for technique in this species (Swindle 2007).
Urine and Feces Collection Rodents, particularly mice, tend to void samples of urine and feces when they are picked up and restrained. Gentle manual compression of the abdomen or placing the animal in an empty cage or box may provoke urination (Van Loo et al. 2002). To collect uncontaminated urine samples, percutaneous cystocentesis (aspiration of urine via needle through the abdominal wall) may be used. The anesthetized or sedated animal is restrained in an upright position or in dorsal recumbency. A needle is directed into the abdomen toward the midline, just anterior to the pubis, and gentle aspiration is used to collect the urine. In some species, such as rabbits, the full bladder may be palpated to confirm location prior to centesis. The bladder will decrease in size as urine is aspirated and may move away from the needle. However, the bladder should not be squeezed during the procedure. If urine does not flow into the syringe, the needle should be withdrawn and another attempt made with a clean needle. Probing in the abdomen may damage gut or other tissues and contaminate the bladder. If the bladder cannot be palpated with the animal in dorsal recumbency, a blind attempt may be made, directing the needle on the midline, midway between the umbilicus and brim of the pelvis. Blind technique carries increased risk of internal damage or contamination from the gut. For longer term collection, metabolism cages are available to separate and collect urine and feces (Figure€14.39), but the animals will be singly housed on grid floors, which may cause stress. Catheterization of the urethra is possible in some species but requires experience. Care is taken to avoid contamination of the catheter, which can introduce pathogenic bacteria into the bladder. Once the catheter has entered the bladder, urine should flow readily with gentle compression of the abdomen or light aspiration with a syringe on the catheter. In the majority of female mammals, the urethra opens into the vagina and is not visible. In the mouse, rat, guinea pig, and hamster, however, the urethra and the vaginal orifice are completely separate. If catheterization is performed in females, detailed knowledge of the anatomy is required together with well-developed technical skills.
84550.indb 440
11/11/10 6:27:36 PM
Common Nonsurgical Techniques and Procedures
441
Figure 14.39â•…Sampling urine and feces from a rat using a metabolism cage. (Techniplast Gazzada S.A.R.L., Italy.)
In female rats, a 22-gauge flexible, over-the-needle catheter is used, with the needle removed. The anesthetized rat is held in dorsal recumbency, with the head toward the handler. The rat’s tail is held by the handler’s index finger, and the thumb applies gentle traction cranially on the abdomen to open the genital papilla. The catheter is advanced first in a caudal direction over the pelvis, then cranially into the bladder. In unanesthetized rabbits, a flexible size 9 F catheter may be passed through the urethra. Sedation with subcutaneous acepromazine maleate at 0.8 mg/kg will calm a struggling animal and will encourage male rabbits to protrude the penis, which may facilitate catheterization. The male is restrained in dorsal recumbency, and the catheter is passed into the urethra and directed caudally. The catheter will follow the urethra caudally first, and then curve up and ventrally into the bladder. A rabbit doe is placed in ventral recumbency. The catheter must be directed vertically into the caudal part of the vagina first and then brought into a horizontal position and into the urethral opening located in the ventral vagina. Urine collection may be facilitated by placing the rabbit in a box some hours before collection, to discourage premature voiding. Cerebrospinal Fluid For all species, collection of cerebrospinal fluid (CSF) is conducted with the animal in a surgical plane of anesthesia. The area over the dorsal atlanto-occipital area is shaved and cleaned with a presurgical scrub. The animal is placed in lateral recumbency, and the head is held firmly flexed forward, taking care not to obstruct respiration. A needle is introduced through the
84550.indb 441
11/11/10 6:27:37 PM
442
Handbook of Laboratory Animal Science
atlanto-occipital membrane, on the midline. The needle is kept parallel to the table surface and is advanced slowly and carefully, keeping in mind that the target area is usually quite small and shallow. A decrease in resistance to advancing the needle usually indicates that the needle has entered the CSF space. The sample may be contaminated with blood, particularly if repeated samples are taken (Vogelweid and Kier 1988). Catheters may be placed for repeated sampling in larger rodents, rabbits, and larger species. Catheters may be placed in the lumbar subarachnoid space or through the skull into the third intracerebroventricular space. Clinical textbooks on veterinary medicine should be consulted prior to CSF collection in the pig (Swindle 2007). Rats Between 0.1 and 0.5 mL may be obtained from the cisterna magna. A 22- or 23-gauge needle is used. A technique for collection from the lumbar region has also been described (Strake et al. 1996; Petty 1982). Mice To reveal the puncture site, the skin is incised from a point 4 mm cranial to the external occipital protuberance to a point 1 cm cranial to the shoulder. About 0.025 mL of CSF can be collected with a 22-gauge needle (Vogelweid and Kier 1988). Hamster This technique can be performed as in the rat. Guinea Pigs A 23-gauge needle is introduced through the atlanto-occipital membrane, at a 20–0° angle between the needle and the axis of the head. A volume of up to 0.33 mL can be obtained. Rabbits A 22- or 23-gauge 1.5 in. spinal needle is used. The needle is inserted about 2 mm caudal to the occipital protuberance and advanced slowly toward the midline. A volume of 1.5–2.0 mL can be obtained.
Blood Pressure Measurement The most precise blood pressure measurements are obtained from instruments implanted in central vessels, typically the femoral artery or abdominal aorta. Several commercially available, implantable telemetric blood pressure monitors are available for use in animals as small as mice. A surgical procedure is required for their placement. Use of telemetry allows collection of data that are free from the physiological effects of restraint on the animal (Kramer and Remie 2005; Poole 1999). Indirect blood pressure measurements can be made with adapted versions of the sphygmomanometer used in humans. In rats and mice, a pneumatic cuff can be fitted over the animal’s tail and inflated to occlude blood flow. A detector and pressure transducer in the cuff provide a signal that is converted into analog voltage for recording. Because mice and rats alter blood circulation in their
84550.indb 442
11/11/10 6:27:37 PM
Common Nonsurgical Techniques and Procedures
443
tails depending on ambient temperature, room and restraint conditions must be held constant from one reading to the next. In addition, restrained animals that are not conditioned to handling will not provide reliable readings. Tail cuff blood pressure monitors are available commercially. In large species, noninvasive methods (Doppler ultrasonography and oscillometry) are available for detecting blood pressure but are not considered to be as accurate as direct arterial puncture measures. The Doppler method only measures systolic pressure, while the oscillometric system measures systolic and diastolic pressures. A pressure cuff can be placed around a forepaw or hind paw or tail. Allowing an animal at least 15 min to become used to the pressure cuff each time it is applied may improve consistency of the readings. Imaging Techniques Radiological studies are frequently performed in laboratory animals. However, this requires adequate anesthesia and cannulation techniques. When contrast fluid has been introduced through the cannula, blood flow through organs can be studied. Scanning techniques are used more and more frequently to study tumor growth, organ transplants, etc. Labeled substances that attach to the transplanted tissue or tumor are injected. The volume of the transplant or tumor can be estimated by a scanner. In a CT (computer tomography) scan, the absorption of photons is used. With MRI (magnetic resonance imaging), differences in magnetic quality of compounds within different tissues become visible and result in an image of the tissue. Conclusion In this chapter, some basic experimental procedures have been described in order to provide a general background. This chapter is not intended to be used as a direct practical guide. One should never use a technique that has not been fully explained and demonstrated by an experienced person. Also, practicing on a dummy or a nonliving or unconscious animal should always precede the use of these techniques in an experimental setting.
References Baumans, V., R. Remie, H. J. Hackbarth, and A. Timmerman. 2001. Experimental procedures. In Principles of laboratory animal science, rev. ed., ed. L. F. M. van Zutphen, V. Baumans, and A. C. Beynen, 313–333. Amsterdam: Elsevier. Brown, R. H., D. M. Walters, R. S. Greenberg, and W. Mitzner. 1999. A method of endotracheal intubation and pulmonary functional assessment for repeated studies in mice. Journal of Applied Physiology 87:2362 (http://www.jap.org). Coria-Avila, G. A., A. M. Gavrila, S. Menard, N. Ismail, and J. G. Pfaus. 2007. Cecum location in Ratus and the implications for intraperitoneal injections. Lab Animal 36:7. De Brandt, V., and J. P. Remon. 1991. A simple method for the intragastric administration of drugs to fully conscious guinea pigs. Laboratory Animals 25:308–309. Diehl, K.-H., R. Hull, D. Morton, et al. 2001. A good practice guide to the administration of substances and removal of blood, including routes and volumes. Journal of Applied Toxicology 21:15 (http://www3. interscience.wiley.com/cgi-bin/abstract/76510682/START). Flecknell, P. A., J. H. Liles, and H. A. Williamson. 1990. The use of lignocaine-prilocaine local anesthetic cream for pain-free venipuncture in laboratory animals. Lab Anim 24:142. Framstad, T., S. Oystein, and R. Aass. 1988. Bleeding and intravenous techniques in pigs. Norwegian School of Veterinary Science (http://www.oslovet.veths.no/teaching/pig/pigbleed).
84550.indb 443
11/11/10 6:27:37 PM
444
Handbook of Laboratory Animal Science
Golde, W. T., P. Gollobin, and L. L. Rodriguez. 2005. A rapid, simple and humane method for submandibular bleeding of mice using a lancet. Lab Animal Europe 5 (9): 29–34. Hem, A., A. J. Smith, and P. Solberg. 1998. Saphenous vein puncture for blood sampling of the mouse, rat, hamster, gerbil, guinea pig, ferret and mink. Lab Anim 32:364 (http://www.uib.no/vivariet/mou_blood/ Blood_coll_mice.html). Jeffery, P., M. Burrows, and A. Bye. 1987. Does the rat have an empty stomach after an overnight fast? Laboratory Animals 21:330–334. Joint Working Group on Refinement: British Veterinary Association, Fund for the Replacement of Animals in Medical Experiments, Royal Society for the Prevention of Cruelty to Animals, Universities Federation for Animal Welfare. 1993. Removal of blood from laboratory animals. Lab Anim 27:1 (http://www.lal.org.uk/pdffiles/blood.pdf). Kramer, K., and R. Remie. 2005. Measuring blood pressure in small laboratory animals. In Methods in molecular medicine, vol. 108: Hypertension: Methods and protocols, ed. J. P. Fennell and A. H. Baker, 51–62. Totowa, NJ: Humana Press Inc. Lawson, P. T., ed. 1999. Assistant laboratory animal technician, American Association for Laboratory Animal Science. Chelsea, MI: Sheridan Books, 45. MacDonald, K. D., H.-Y. S. Chang, and W. Mitzner. 2009. An improved simple method of mouse lung intubation. Journal of Applied Physiology 106:984–987. Marcus, G. E., F. T. F. Shum, and S. L. Goldman. 1990. A device for collecting milk from rabbits. Laboratory Animal Science 40:219–220. McGuill, M. W., and A. N. Rowan. 1989. Biological effects of blood loss: Implications for sampling volumes and techniques. ILAR News 31:4–5 (http://www4.nas.edu/cls/ijhome.nsf/web/McGuill3104). Merck Veterinary Manual. 2005. Fluid compartment dynamics (http://www.merckvetmanual.com/mvm/index. jsp?cfile=htm/bc/160402.htm). Olfert, E. D., B. M. Cross, and A. A. McWilliam, eds. 1993. Guide to the care and use of experimental animals, vol. 1. Ottawa, Ontario, Canada: Canadian Council on Animal Care (http://www.ccac.ca/guides/english/ toc_v1.htm). Petty, C. 1982. Research techniques in the rat. Springfield, IL: Charles C Thomas. Poole, T., ed. 1999. UFAW handbook on the care and management of laboratory animals, vol. 1, 7th ed. Oxford, England: Blackwell Science Ltd. Rodgers, C. T. 1995. Practical aspects of milk collection in the rat. Lab Anim 29 (4): 450–455. Sharp, J. L., T. Zammit, T. Azar, and D. Lawson. 2002a. Does witnessing experimental procedures produce stress in male rats? Contemporary Topics in Laboratory Animal Science 41 (5): 8–12. ———. 2002b. Stress-like responses to common procedures in male rats housed alone or with other rats. Contemporary Topics in Laboratory Animal Science 41 (4): 8–14. ———. 2003a. Are bystander female Sprague Dawley rats affected by experimental procedures? Contemporary Topics in Laboratory Animal Science 42 (1): 19–27. ———. 2003b. Stress-like response to common procedures in individually and group housed female rats. Contemporary Topics in Laboratory Animal Science 42 (1): 9–18. Shomer, N. H., K. M. Aasrofsky, C. A. Dangler, et al. 1999. Biomethod for obtaining gastric juice and serum from the unanesthetized guinea pig (porcellus). Contemporary Topics in Laboratory Animal Science 38 (5): 32. Strake, J. G., M. J. Mitten, P. J., Ewing, et al. 1996. Model of Streptococcus pneumoniae meningitis in adult rats. Laboratory Animal Science 45:5. Svendsen, P., and J. Hau, eds. 1994. Handbook of laboratory animal science, vol. I. Selection and handling of animals in biomedical research, 254–255. Boca Raton, FL: CRC Press. Swindle, M. M. 2007. Swine in the laboratory: Surgery, anesthesia, imaging and experimental technique, 2nd ed. Boca Raton, FL: CRC Press. Van Herck, H., V. Baumans, C. J. W. M. Brandt, et al. 2001. Blood sampling from the retro-orbital plexus, the saphenous vein and the tail vein in rats: Comparative effects on selected behavioral and blood variables. Lab Anim 35 (2): 131.
84550.indb 444
11/11/10 6:27:37 PM
Common Nonsurgical Techniques and Procedures
445
Van Loo P. L. P., C. L. J. J. Kruitwagen, J. M. Koolhaas, H. A. Van de Weerd, L. F. M. Van Zutphen, and V. Baumans. 2002. Influence of cage enrichment on aggressive behavior and physiological parameters in male mice. Applied Animal Behavior Science 76:65–81. Vogelweid, C. M., and A. B. Kier. 1988. A technique for the collection of cerebrospinal fluid from mice. Laboratory Animal Science 38 (1): 9102.
84550.indb 445
11/11/10 6:27:38 PM
84550.indb 446
11/11/10 6:27:38 PM
Chapter 15
Applications of Radiotelemetry in Small Laboratory Animals
Klaas Kramer and Steve Hachtman Contents Introduction.....................................................................................................................................448 Applications and Evaluation........................................................................................................... 450 Transmitter Implantation........................................................................................................... 450 Arterial Blood Pressure............................................................................................................. 451 Core Body Temperature/Movement/Activity............................................................................. 452 Heart Rate.................................................................................................................................. 452 Electrocardiogram...................................................................................................................... 453 Respiratory Rate........................................................................................................................ 454 Electroencephalogram............................................................................................................... 454 Telemetry Application in Transgenic Models................................................................................. 455 Telemetry and Laboratory Animal Welfare................................................................................... 457 Telemetry and Reduction Alternatives............................................................................................ 458 Telemetry and Refinement Alternatives.......................................................................................... 459 New and Future Developments.......................................................................................................460 Conclusions..................................................................................................................................... 461 References....................................................................................................................................... 462 Radiotelemetry provides an alternative method for obtaining physiological measurements from awake and freely moving laboratory animals, without introducing stress artifacts. For researchers, especially those in the fields of pharmacology and toxicology, the technique may provide a valuable tool for predicting the effectiveness and safety of new compounds in humans. Also, since radiotelemetry with an implantable transmitter provides a way to obtain accurate and reliable physiological measurements from awake and freely moving animals in their own environment, it is a valuable tool to investigate animal welfare without the overlay of stress-producing factors associated with conventional techniques. Ample current evidence can sufficiently validate the use of radiotelemetry for measuring physiological parameters. Today, this technology is an important tool for the stress-free collection of physiological parameters, such as blood pressure, electrocardiogram, heart rate, respiratory rate, body temperature and locomotor activity in laboratory animals, including mice and even fish. 447
84550.indb 447
11/11/10 6:27:38 PM
448
Handbook of Laboratory Animal Science
Receiver Receiver
Receiver
Ambient Pressure Reference
Data Acquisition and Analysis System Matrix
Figure 15.1â•…Schematic drawing of a telemetry system for measuring physiological parameters in small laboratory animals.
Introduction The desire to experience life from the perspective of animals dates from antiquity (Kramer 2000). Scientists have adapted available technologies to study animals in their quests to unravel and understand biological functions and processes for several centuries; in the most recent of these endeavors, scientists have applied radiotelemetry technology. Radiotelemetry combines miniature sensors and transmitters to detect and broadcast biological signals in animals to a remote receiver. The receiver converts an analog frequency signal into a digital signal for input into a computerized data acquisition system. The acquisition system can store, manipulate, format, tabulate, and output the data in accord with the instructions of the user. Currently, radiotelemetry systems can collect blood pressure (BP), heart rate (HR), blood flow (BF), electrocardiogram (ECG) and other biopotentials (electroencephalogram [EEG], electromyogram [EMG]), respiratory rate (RR), pH, body temperature (BT), and activity indices (Figure€15.1). As with any new technology, scientists should be skeptical and demand validations, to the extent possible, of the new technology compared to conventional measurement techniques. However, new technologies make it possible to perform measurements under conditions that have not previously been possible, making direct comparisons with conventional measurement techniques sometimes impractical. With the exception of studies of anesthetic agents and certain other types of experiments involving acutely painful or stressful procedures, it is generally acknowledged that the quality of physiological measurements collected from conscious, unstressed animals is superior to data collected from unconscious, stressed animals. Unstressed conditions best represent the normal state of the animal, are least influenced by chemical and psychological factors, and (when appropriate) are most predictive of results that might be achieved in human beings. When physiological parameters in conscious animals are monitored, it is better to use noninvasive methods such as surface electrodes for monitoring an ECG or a tail-cuff manometry for monitoring BP.
84550.indb 448
11/11/10 6:27:40 PM
Applications of Radiotelemetry in Small Laboratory Animals
449
However, use of noninvasive techniques can introduce important artifacts because these techniques often require physical restraint and high levels of technician interaction to ensure that surface connections stay in place and do not accidentally damage, or get damaged by, the animal. The literature extensively documents the effects of restraint on laboratory animals, including increases in BT, HR, and BP; plasma levels of epinephrine and norepinephrine; changes in responses to pharmaceuticals; and decreased food intake and body weight (Kramer et al. 2001; Kramer and Kinter 2003; Hawkins et al. 2004). Temporary use of invasive methods to implant sensors under the skin and/or within body cavities can eliminate the experimental artifacts associated with noninvasive procedures. The development of modern microsurgical techniques and surgical recovery procedures has greatly facilitated the use of indwelling catheters (for monitoring fluid pressure and removing fluid samples), electrodes (for monitoring biopotentials and temperatures), and devices (e.g., flow probes, vascular cuffs) (Van Dongen et al. 1990). When indwelling catheters, sensors, and/or electrodes are used, several methods (including wireless radiotelemetry) for accessing the information from the sensor and forwarding it to a recording system are available; the advantages and disadvantages of each of these methods have been extensively summarized in previous reviews (Kramer et al. 2001; Hawkins et al. 2004). Although wireless radiotelemetry technology for monitoring laboratory animals has existed for at least 50 years (Brockway and Hassler 1993; Kramer 2000), it has only been in the last 10 years that affordable, reliable, and easily used commercial products have been readily available for monitoring physiological signals in laboratory animals (Kramer 2000). Understandably, greater availability has resulted in a significant increase in the use of implantable radiotelemetry in biomedical research. Several authors have highlighted its advantages: • Reduction of distress when compared to conventional measurement techniques; telemetry represents the most humane method for monitoring physiological parameters in conscious, freely moving laboratory animals (Brockway and Hassler 1993) • Elimination of the use of restraints, thereby alleviating a potential source of experimental artifact and interanimal variability (Schnell and Gerber 1997) • Reduction of animal use by 60–70% when animals are used in a single study (Van Acker et al. 1996) and by more than 90% when animals are used in multiple studies (Kinter 1996) • Virtually unrestricted continuous data collection (days, weeks, months, or more) without the need of any special animal care (Brockway and Hassler 1993). • Available for use in all laboratory species, from mice to monkeys (Kramer 2000) and even fish (Snelderwaard et al. 2006)
Considerations include: • The high costs associated with the acquisition of the requisite equipment (usually completely recovered through reduced animal use) • Specialized training or certifications required to prepare and study transmitter-implanted animals surgically • Large amounts of data generated by continuous or regularly scheduled sampling, which can lead to analysis difficulties (e.g., sampling every 5 minutes over an 8-week period will generate more than 16,000 data points per parameter) • Requirement of dedicated space within the animal facility in which to conduct studies
While the use of telemetry technology is well established in many diverse areas of animal research involving small and large laboratory animals, we have limited the scope of this chapter to the most common telemetry applications in small laboratory animals.
84550.indb 449
11/11/10 6:27:40 PM
450
Handbook of Laboratory Animal Science
Figure 15.2â•…X-ray of a mouse with a radio-telemetry transmitter with the ECG leads sutured subcutaneously in the lead II position and with the body of the transmitter implanted in the peritoneal cavity to measure ECG, HR, and BT.
Applications and Evaluation Transmitter Implantation Transmitter implantation is carried out by placing the body of the transmitter subcutaneously or into the peritoneal cavity. To obtain ECG measurements, the positive and negative electrodes are placed subcutaneously (Figure€15.2), while the fluid-filled catheter can be placed into the carotid artery (rat and mouse), thoracic aorta (rat), femoral artery (rat) and/or abdominal aorta (rat and mouse) to obtain blood pressure measurements. Even though the radiotelemetry technique is an invasive method and thus the weight and volume of the implanted transmitter may cause discomfort (Hawkins et al. 2004; Arras et al. 2007), animals seem to tolerate the implantation of the transmitter without evident problems (Moran et al. 1998; Baumans et al. 2001; Leon et al. 2004; Chin 2005; Greene, Clapp, and Alper 2007). Baumans et al. (2001) assessed changes in body weight and behavior of BALB/c and 129/Sv mice after the implantation of an intra-abdominal transmitter. During the first days after surgery, body weight and certain behaviors (climbing, locomotion, and eating) decreased in both strains, whereas certain other behaviors (grooming and immobility) increased. These changes were more pronounced in the animals with transmitters than in the sham-operated controls, indicative of a temporary impairment in well-being. However, data collected at later time points showed that all animals were fully recovered by 2 weeks after surgery (Baumans et al. 2001). Also, Chin (2005) reported that at least an 8- to 10-day recovery period is needed for stabilization of the circadian rhythm of HR and even longer
84550.indb 450
11/11/10 6:27:41 PM
Applications of Radiotelemetry in Small Laboratory Animals
Mouse and Rat Blood Pressure
mmHg
Mouse 1/9/2004 2:00:13 PM 120.0 100.0 80.0
0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
mmHg
QAM8705.Pressure 11/22/2003 12:00:01 AM Rat 140.0 130.0 120.0 110.0 100.0 0 1.0
451
QAX24594.Pressure
Figure 15.3â•…Representative blood pressure waveforms measured in a freely moving mouse and rat with the catheter of the transmitter inserted upstream into the abdominal aorta and with the body of the transmitter positioned in the peritoneal cavity.
periods are required for stabilization of BT and activity in mice. Moreover, implanted mice required 12–14 days for their body weights to reach presurgical levels. Finally, Greene et al. (2007) have concluded that male Sprague–Dawley rats recover from surgery and can be used approximately 1 week after intraperitoneal implantation of a radiotelemetry transmitter. Arterial Blood Pressure The typical implant procedure for monitoring blood pressure (BP) in small laboratory animals such as (spontaneously hypertensive) rats and mice involves a medial laparotomy and placement of the pressure-sensing catheter into the lower abdominal aorta between the bifurcation of the iliac and the renal arteries (Kramer and Remie 2005; Figure€15.3). Other investigators (Balakrishnan and McNeill 1996) have inserted the catheter into the femoral artery while placing the transmitter body under the skin or in the abdominal cavity of the rat. Femoral access reduces the risk of bleeding (especially in hypertensive animals), reduces hind limb paralysis, and is less invasive. Circadian variations of hemodynamic variables are of interest to many investigators in connection with the circadian patterns of incidences of cardiovascular diseases and the effects of drugs on circadian rhythms. Mattes and Lemmer (1991) reported the effect of amplodipine, a calcium channel blocker, on circadian rhythms in BP and HR obtained from the abdominal aorta of rats using the radiotelemetry system. The data from this chronopharmacologic study support the usefulness of a radiotelemetry system to investigate the dose–response relationship of cardiovascularly active drugs. A study by Vleeming and colleagues (1997) clearly showed the advantage and necessity of measuring BP and HR around the clock by radiotelemetry to gain greater insight into the cardiovascular changes after nitrite administration. The authors found that nitrite administration in drinking water decreased BP in rats only at night, a finding that could not be made by another research group that was using the tail-cuff method to measure BP only during the day (Til, Kuper, and Falke 1997). Validation of telemetric pressure measurement accuracy has been described in several publications (e.g., Kramer and Remie 2005). In addition, Mills et al. (2000) first described the ability to record systolic, diastolic, and mean BP, HR, and locomotor activity in freely moving mice in their home cage over a period of several months. For this purpose, the catheter of the transmitter was inserted into the abdominal aorta just caudal to the left renal artery, while the body of the transmitter was positioned in the peritoneal cavity. Because the device was developed for abdominal aorta implantation at the renal artery level, its use has not been feasible in studies where infrarenal blood flow is critical (i.e., in pregnant mice).
84550.indb 451
11/11/10 6:27:42 PM
452
Handbook of Laboratory Animal Science
Butz and Davisson (2001) developed an alternative approach for implanting radiotelemetry transmitters for measuring BP in mice whereby thoracic aorta implantation of the pressure-sensing catheter is combined with subcutaneous placement of the transmitter body along the right flank. They used female C57BL/6 or BPH/5 mice strains derived from the cross of inbred hypertensive and hypotensive mouse strains. It was shown that their approach was a reliable method to measure (under stress-free conditions) mean arterial pressure and HR recordings for 50–60 days in mice weighing an average of 22 g, but as small as 17 g, before, during, and after pregnancy. They also mentioned possible application in cardiovascular functional genomic research. Although placing the BP catheter in the carotid rather than in the abdominal aorta of the mouse may permit a less invasive surgical procedure, one must be aware that the circle of Willis is not completely developed in some strains of mice; thus, the carotid approach is not advised for these strains (Barone et al. 1993). Finally, Kaidi et al. (2007) compared two surgical methods for mouse instrumentation with telemetry devices—the left carotid (LC) and the abdominal aorta (AA)—to determine the best method for measuring cardiovascular (CV) parameters by radio telemetry in freely moving mice. Surgery success rate, postsurgical recovery rate, clinical parameters, CV data (baseline and response to nicotine), and circadian rhythm measurements were compared. It was concluded that LC implantation in mice is superior to the AA technique and is more appropriate for long-term telemetry studies, especially for smaller (transgenic) animals. Core Body Temperature/Movement/Activity Clement, Mills, and Brockway (1989) initially demonstrated that a telemetry system (transmitter implanted in the peritoneal cavity of a mouse) for the recording of body temperature (BT) and activity can provide an accurate assessment of the temporal effects of various drugs with increased efficiency and less labor than the use of a rectal temperature probe. Kluger et al. (1990) investigated the effect of the gastrointestinal flora on BT of rats and mice and concluded that gut flora has a tonic stimulatory effect on both the daytime and nighttime BT of rodents. Today, more and more researchers are using both parameters as research tools in their pharmacological, toxicological, physiological, biological, and behavioral studies (Meerlo et al. 1996; Billing 2005; Chin 2005; Capdevila et al. 2007). Recently, Conley and Hutson (2007) used radio telemetry in rats and mice to record BT, locomotor activity, and ethologically relevant stressors to examine the effects of acute and chronic administration of fluoxetine on a physiological measure of anxiety: stress-induced hyperthermia. They concluded that their observations support the view that chronic administration of fluoxetine is anxiolytic; however, the stress-induced hyperthermia assay did not reveal anxiogenic effects of acute administration of fluoxetine in rats or mice. Heart Rate Heart rate (HR) can be used as an indicator in the assessment of animal welfare (Broom and Johnson 1993). Because an increase in HR is one of the major cardiovascular responses to stress (Herd 1991), HR is an important parameter for measuring stress responses in biomedical research. We found that regular handling of male BALB/c mice during a long period of time (<6 months) remained a very stressful experience for these mice; HR increased to maximum values of between 750 and 800 beats per minute (bpm), while minimum values between 400 and 450 bpm were measÂ� ured when the animals were sleeping in their home cages (Kramer 2000; Kramer et al. 2004). Investigations on conditioned response in C57BL/6 mice revealed that when a single acoustic stimulus was paired with foot shock, the stimulus could provoke defensive behavior (freezing) in a memory test (Paylor et al. 1994). Moreover, the results from Stiedl and Spiess (1997) on HR effects during a tone-dependent retention test, after one-trial classical fear conditioning, indicated that HR
84550.indb 452
11/11/10 6:27:42 PM
Applications of Radiotelemetry in Small Laboratory Animals
453
reflects associative learning in C57BL/6N mice. It was shown that in mice subjected to a conditioning stimulus (tone) paired with foot shock, the tone onset triggered an immediate HR increase. Both studies indicate that mice can be used in learning and memory tests and that the learning abilities of different strains and substrains of mice may differ (Gutekunst et al. 1993; Stiedl et al. 1999). It seems that C57Bl/6N mice are fast learners, whereas BALB/c mice have been reported to be slow learners (Stiedl and Spiess 1997). We have used radiotelemetry to investigate the influence of conditioning on the increase of HR as provoked by handling of C57Bl/6N mice. It was found that after a conditioning period of 12 days, the increase of HR due to handling was significantly reduced (Kramer et al. 2004). Also, Billing (2005) demonstrated a range of physiological effects in mice that occurred in response to repeated restraint stress. A significant increase in HR was found that habituated over the 14-day experimental protocol. Recently, HR was measured before and after several hours of van transport in rats and guinea pigs; surprisingly, a decrease of HR was reported in rats (Capdevila et al. 2007), as well as in guinea pigs (Stemkens-Sevens et al. 2009). Electrocardiogram The use of radiotelemetry implies that the animals may be moving actively, and this possibility means that proper positioning of electrocardiogram (ECG) electrodes is very important for collecting reliable ECG data (Figure€15.4). Improper placement of the electrodes will cause the ECG waveform to be noisy and/or distorted by electrical potentials from local muscle activity. The bipolar configuration of the radiotelemetry system is the standard limb-lead system commonly used for lead I, II, and III configurations (Einthoven 1957). The typical implantation procedure for monitoring ECGs involves positioning the body of the transmitter in the peritoneal cavity of the animal, followed by placement of the ECG electrodes subcutaneously in a lead II position (Kramer et al. 1993)—that is, the negative electrode at the right shoulder and the positive electrode at the lower left chest. However, Sgoifo et al. (1996) described several lead positions for optimizing the location of telemetry ECG electrodes in rats. According to this publication, the optimal location of telemetry ECG electrodes could be achieved by fixing one electrode to the dorsal surface of the xiphoid process. The other electrode was subcutaneously tunneled on the thorax toward the upper insertion of the sternohyoid muscle. Here, this electrode was pushed under the muscle and then along the trachea into the anterior mediastinum close to the atrium. This location ensured a low level of noise and minimized the effects of intense physical activity on the amplitude of the signals. Mitchell, Jeron, and Koren (1998) showed ECG traces from mice obtained with a lead placement in which the negative lead was sutured subcutaneously at the right scapula, whereas the positive lead was sutured to the chest wall overlying the apex of the heart. The preferred lead placement found in Mouse and Rat ECG
mV
0.8 0
mV
0.50
0 0.5 Mouse.ECG
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0 Rat.ECG
0.5
Figure 15.4â•…Representative ECG waveforms measured in a freely moving mouse and rat with the ECG leads sutured subcutaneously in the lead II position and with the body of the transmitter positioned in the peritoneal cavity.
84550.indb 453
11/11/10 6:27:43 PM
454
Handbook of Laboratory Animal Science
the rat studies of Brockway (1998a) was archived by placing the positive lead near the lower thorax and the negative electrode at the right shoulder. This placement minimizes muscle artifacts from the abdominal muscles. This positioning of leads resembles the lead II configuration described for several laboratory animals (Kramer 2000). Considerable effort has been put into optimizing the location of telemetry ECG electrodes in small laboratory animals. It seems that the two sensing leads of the radiotelemetry system yield the best results when positioned in the lead II configuration. Minor artifacts mainly related to physical activity probably could be overcome by placing one or two electrodes directly on the heart, but this solution would require open thorax surgery. Van Acker et al. (1996) showed that after implanting an ECG transmitter with the electrodes in the lead II position, the most important advantage was the opportunity for direct and accurate measurement of chronic effects of cardioactive compounds, without the stress resulting from conventional methods. These findings resulted in a model that allowed for the testing of protectors against cardiotoxicity caused by the antitumor agent doxorubicin. Respiratory Rate A method for measuring the respiratory rate (RR) in rats (Brockway and Huetteman 1999) and in mice (Brockway and Huetteman 2001; Data Sciences International, personal communication) has been developed that has the potential to provide unique data in long-term biomedical studies. The technique allows measurements of the RR from freely moving, conscious laboratory rats and mice by using an implantable radiotelemetry transmitter for measuring BP. This technique uses digital signal processing methods to deduct RR automatically from beat-to-beat information obtained from the BP signal. The small telemetric transmitters are implanted into the peritoneal cavity and the sensor catheter is inserted into the abdominal aorta and/or into the carotid artery. Electroencephalogram Cotugno et al. (1996) described a method of implanting a radiotelemetric transmitter to measure electroencephalogram (EEG) signals in the rat and recommended radiotelemetry EEG recording for the following reasons: (a) subcutaneous placement of the transmitter, (b) EEG recordings around the clock for weeks, (c) quite simple surgery, and (d) a commercially available system (Figure€15.5). A new commercially available transmitter for mice (Data Sciences International, St. Paul, Minnesota), that allows simultaneous monitoring of two biopotential channels, such as EEG and EMG or temperature and locomotor activity, was validated by comparing data from Swiss Webster mice using a commercially available sleep analysis program (Somnologica, Medcare, Reykjavik,
Figure 15.5â•…X-ray of a rat with a radio-telemetry transmitter with the EEG and EMG leads sutured subcutaneously to the head and to the neck of the animal and with the body of the transmitter implanted subcutaneously to measure EEG and EMG.
84550.indb 454
11/11/10 6:27:43 PM
Applications of Radiotelemetry in Small Laboratory Animals
455
10/19/2007 6:59:50 PM
0.4 mV
0.2 0 –0.2 –0.4 0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
SLP2ETM49102.EEG 10/19/2007 6:59:50 PM
0.4 mV
1.0
0.2 0 –0.2 –0.4 0
1.0 SLP2ETM49102.EEG
Figure 15.6â•…Representative EEG and EMG waveforms measured in a freely moving mouse with EEG and EMG leads sutured subcutaneously to the head and to the neck of the animal and with the body of the transmitter implanted subcutaneously.
Iceland). Data assessment for 24-hour periods of cortical EEG and neck EMG recordings indicated agreement of manually scored sleep data with the automated scoring program (Figure€15.6). Tang and Sanford (2002) used this new mouse transmitter for the first time in a study in which they assessed differences in spontaneous sleep and locomotor activity in inbred and hybrid mouse strains. EEG and activity were recorded by telemetry, and behavioral states were visually scored based on EEG and activity records. Home cage activity was determined utilizing photobeam interruptions. It was concluded that differences in spontaneous sleep and activity exist among inbred and hybrid mouse strains and that accurate determination of sleep states in mice can be achieved with telemetrically recorded EEG and activity. Weiergräber et al. (2005) published a paper in which they described in detail a successful and quick technique for intraperitoneal implantation or subcutaneous pouch implantation of a radio frequency transmitter in mice and subsequent lead placement in both epidural and deep intracerebral positions. In this paper, they showed nice examples of electrocorticograms and deep intracerebral recordings, and special attention was given to major pitfalls and possible artifacts occurring in telemetric EEG recording in mice. To quantify the anesthetic effect of desflurane, isoflurane, and sevoflurane in rats, Ihmsen et al. (2008) investigated the suitability of different EEG parameters by using implanted electrodes and a wireless telemetry system. Recently, a new portable and reusable telemetric system to record the EEG from adult freely moving rats under various experimental conditions was presented (Lapray et al. 2008). This new system is optimized for recording electrical activity from the animal’s brain; simultaneous recording and storage of a video signal for direct comparison of the animal’s EEG with its behavior is also possible because it can be easily modified to record other physiological parameters. Today, more and more papers demonstrate that implantable radiotelemetry is a reliable and sensitive method for monitoring surface and deep EEG signals in rats and mice (Paterson et al. 2009; Mouri et al. 2008; Yang et al. 2009). Telemetry Application in Transgenic Models During the last decade, use of transgenic models to investigate the cause of or potential treatments for cardiovascular diseases has increased significantly. By using transgenic and gene targeting technologies, it is possible to explore many aspects of molecular cardiovascular biology (Fewell et al. 1997; James, Hewett, and Robbins 1998). With recent advances in molecular biology,
84550.indb 455
11/11/10 6:27:44 PM
456
Handbook of Laboratory Animal Science
transgenic mice have become important experimental models for cardiovascular diseases (Saba, Wang, and Estes 2000). Expression of abnormal quantities or altered forms of gene products in an intact animal provides a powerful tool to gain insight into the role of specific genes in cardiac growth, development, and function (Mitchell et al. 1998). Examples of transgenic mouse models relevant to cardiovascular physiology and diseases have been extensively described and discussed in a broad review paper (Kramer and Kinter 2003). It can be concluded from these reviews that implantable radiotelemetry measurement of the possible differences in physiological parameters, such as ECG, HR, and/or BP, between the transgenic mice and the nontransgenic mice is becoming an important research technique for use in assessment of genetically engineered mice. Indeed, recent measurement of physiological parameters, like ECG, HR, or BP, in transgenic mouse models has resulted in significantly different findings in the shape of ECGs, in baseline values of HR and BP, and even in responses to autonomic agents compared to nontransgenic control mice (Kramer and Kinter 2003; Jackson et al. 2007; Sällström et al. 2008). Radiotelemetry has also been used in combination with a treadmill test in transgenic mice overexpressing ventricular myosin regulatory light chain. The capacity for exercise at higher belt speed (27 m/min) was significantly decreased in transgenic mice compared to nontransgenic mice (Fewell et al. 1997). Numerous researchers have advanced their understanding of transgenic cardiovascular alterations and changes by obtaining conscious animal ECGs via improved high-fidelity physiological data offered by telemetric monitoring. Mitchell et al. (1998) described a system for long-term monitoring and evaluation of electrocardiographic RR and QT intervals in transgenic mice with ion-channel defects. These intervals were measured by using radiotelemetry in conscious as well as anesthetized mice. Surface ECG findings were consistent with QRST morphology observed in the telemetric studies. Ketamine anesthesia caused a markedly increased duration and variability in RR and QT intervals. This difference led Mitchell et al. (1998) to the conclusion that telemetric studies in the conscious animals allow for more accurate characterization of the electrocardiographic phenotype in transgenic mice without the effects of acute anesthesia and experimental manipulation. The ability to obtain freely moving chronic measurements with telemetric implants has also enabled measuring and assessing changes in blood pressure and heart rate with a large variety of new transgenic and knockout mouse models. Stauss et al. (1999) compared BP variability of eNOS knockout mice (lacking endothelial nitric oxide synthase) with their respective wild-type controls. One day after carotid artery cannulation, BP was recorded in these conscious mice. During resting conditions, BP variability was markedly enhanced in knockout mice compared with wild-type mice. Gehrmann et al. (2000) developed a technology for HR variability (HRV) analysis in the mouse for characterization of HR dynamics modulated by vagal and sympathetic activity. The presented telemetry technology and protocol allow for assessment of automatic regulation of the murine HR. It was concluded that phenotypic screening for HR regulation in mice further enhances the value of the mouse as a model of inheritable electrophysiological human disease. Another paper illustrated the power of radiotelemetry in monitoring long-term cardiovascular changes and circadian BP and HR rhythms in genetically engineered mice. This paper showed that baseline BP values were higher in AT(2) receptor knockout than in wild-type mice (Gross et al. 2000). Shusterman et al. (2002) hypothesized that a significant portion of the differences between various wild-type and transgenic mouse models might be attributable to intrinsic variability in autonomic activity in different mouse strains. To test this hypothesis, they measured HRV and pharmacological responses in three different experimental settings: (a) wild-type FVB mice, a common strain used in transgenic overexpressors; (b) C57Black6/SV129 mice, the most common strain found in gene-targeted mice; and (c) TNF-α mice (on an FVB background) that develop a dilated cardiomyopathy by 12 weeks of age. From this study, it was concluded that interstrain differences in
84550.indb 456
11/11/10 6:27:44 PM
Applications of Radiotelemetry in Small Laboratory Animals
457
autonomic nervous system activity are important when studying transgenic models; the homeostatic states of autonomic nervous system activity are strain specific, leading to differences in HR, responses to pharmacological agents, and pathophysiological changes in disease models; and that strain-specific differences should be considered in selecting the strains of mice used for transgenic and gene targeting experiments. Davisson et al. (2002) reported an inbred mouse strain (BPH/5) with mildly elevated BP that spontaneously develops a syndrome that bears close resemblance to preeclampsia, the most prevalent hypertensive disorder of pregnancy. Given that preeclampsia likely has an important genetic component, it was concluded that this inbred genetic strain may provide an opportunity for studying its molecular genetic pathophysiology. This new technology-enabled ability to look consistently at smaller changes with much better precision in these different models has offered new insights and understanding of those variables that affect cardiovascular dynamics and, most particularly, systemic blood pressure.
Telemetry and Laboratory Animal Welfare The humane use of laboratory animals requires implementation of the “three Rs” (replacement, reduction, and refinement) of Russell and Burch (1959). The replacement alternative is defined by Russell and Burch as any scientific method employing nonsentient material that may replace methods that use conscious vertebrates. The reduction alternative directs the use of experimental designs and statistical methods to support the minimum number of animals required to test the hypotheses at an appropriate power. Finally, the refinement alternative focuses upon methods to minimize or eliminate pain and distress and to enhance laboratory animal well-being. Years of research and hundreds of publications have continually explained the responses of laboratory subjects to experimental changes, but not to general laboratory handling and procedures. The isolation of different laboratory conditions and the measurement of resultant changes of the behavior and physical state of our animal subjects have improved rapidly with new automated technology. Newer wireless monitoring methods and improved laboratory tests have allowed us to better assess the well-being and physiological status of each of our research subjects. Indications from these newer, more sensitive tests and measures have pointed toward the realization that many routine laboratory conditions and treatments present stress that can significantly alter physiology and behavior. Through the use of improved methodology and better principles, informed laboratory scientists can understand these effects and improve not only laboratory animal health and welfare, but also the validity of their research results. In their book, Stress and Animal Welfare, Broom and Johnson (1993) have defined welfare as “the state of an individual as regards its attempts to cope with its environment” (p. 74)—in other words, the ability of an animal to cope with or adapt to internal and external stressors. To investigate animal welfare, the impact of stress on physiological parameters, such as body weight, hormonal levels in plasma and/or urine, HR, BP, and BT, can be measured (Kramer and Kinter 2003; Kramer et al. 2004; Sharp, Azar, and Lawson 2005; Meijer et al. 2006; Van Loo et al. 2007; Capdevila et al. 2007; Stemkens-Sevens et al. 2008). When measurements of the physiological parameters are performed via conventional measurement techniques, the results must be interpreted with caution, since these conventional techniques have by themselves an effect on the animals (Kramer 2000). Since radiotelemetry with an implantable transmitter provides a way to obtain accurate and reliable physiological measurements from awake and freely moving animals in their own environment (Brockway et al. 1998a), it would seem to be a valuable tool to investigate animal welfare without the influence of stressful handling, restraints, or tethers. Duke, Zammit, and Lawson (2001)
84550.indb 457
11/11/10 6:27:45 PM
458
Handbook of Laboratory Animal Science
described the effects of routine cage changing on cardiovascular and behavioral parameters in male Sprague–Dawley rats and concluded that common animal husbandry procedures, such as moving rats to a clean cage, can induce transient, but significant, cardiovascular and behavioral changes. We have used radiotelemetry to investigate the influence of conditioning on the increase of HR as provoked by handling of C57Bl/6N mice. It was found that after a conditioning period of 12 days, the increase of HR due to handling was significantly reduced (Kramer et al. 2004). Also, Meijer et al. (2006) investigated the integrated effects of cage enrichment, social housing, and handling on the acute stress response of animals subjected to routine experimental procedures. Female mice of two inbred strains (BALB/c and C57BL/6) were housed under either minimal husbandry conditions (no cage enrichment, infrequent handling, and a period of individual housing) or enriched husbandry conditions (with cage enrichment, frequent handling, and social housing at all times). The animals were subjected to intraperitoneal injections or short periods of restraint. It was found that individual housing under minimal husbandry conditions, as compared with social housing under enriched husbandry conditions, elevated both basal HR and BT and significantly elevated the relative recovery time following routine experimental procedures. According to the authors, these results also emphasize that husbandry conditions should be taken into account when evaluating physiological measures after routine procedures. Van Loo et al. (2007) investigated whether housing two female mice in a cage, separated by a grid partition (“living apart together”: LAT), counters the adverse effects of individual housing on postoperative recovery. Ten individually (IND) housed mice, nine socially (SOC) housed mice, and nine LAT housed mice were surgically implanted with a telemetry transmitter. Results indicated that SOC mice appear to be less affected by abdominal surgery than IND mice, as indicated by HR and behavior. LAT, however, did not appear to be beneficial to the mice. Increased HR levels and differences in behavior as compared with both SOC and IND animals indicate that LAT may even be the most stressful of the three housing conditions. From this paper it was concluded that mice benefit most from social housing after surgery. However, if social housing is not possible, individual housing appears to be a better option than separating mice by a grid partition. Capdevila et al. (2007) published a paper in which the time needed by rats (who had not been previously transported) to acclimate to a new environment after 5 hours of van transport was assessed. Their results suggested that rats take 3 days to acclimatize to a new environment, as measured (using implanted radiotelemetry transmitters) by the physiological parameters of body weight, HR, and activity. In addition, Stemkens-Sevens et al. (2009) recently showed in a study similar to that of Capdevila and colleagues (2007) that guinea pigs had to recover from the transport for at least 12 days, as reflected by the return of HR, BT, and activity to pretransport levels. Finally, in the field of laboratory animal research, an increasing number of tools are now available for the reliable measurement and recording of either behavioral or physiological parameters. In order to provide simultaneous collection of these parameters, we have successfully integrated two behavior systems (Metris B.V., Hoofddorp, the Netherlands [LABORAS] and Noldus Information Technology Inc. [Pheno Typer], Wageningen, the Netherlands) with a telemetry system (Data Sciences International, St. Paul, Minnesota) (Sommer et al. 2005; De Heer et al. 2010). We believe that the combination of monitoring behavioral and physiological parameters should be of great value in laboratory “animal welfare” research using small animal models. Telemetry and Reduction Alternatives Radiotelemetry can substantially reduce animal use and associated research costs (e.g., in singledose, dose-ranging, or repeat-dose toxicology studies). Cardiovascular and general toxicity data are gained from one set of animals. Changes in blood pressure, in heart rate, or electrocardiographic
84550.indb 458
11/11/10 6:27:45 PM
Applications of Radiotelemetry in Small Laboratory Animals
459
variables detected by radiotelemetry may provide direct evidence of a dose-limiting effect (Kinter et al. 1997), eliminating the need for more animals to evaluate higher doses for toxicity. Statistics play an important role in the reduction of the number of animals used in an experiment, such as in a completely randomized study design where there are several groups of animals with each animal in each group receiving one treatment. Blocking works to filter out the interanimal variation and reduces the numbers of animals needed to obtain the same level of statistical power for evaluating dose and treatment responses (Festing, Chapter 13, this volume). Because radiotelemetry permits continuous or periodic collection of cardiovascular data over prolonged periods, it is compatible with the use of randomized block designs, replacing conventional, completely randomized study designs. In a randomized block design, there is one group of animals, with each animal receiving each treatment in a randomized fashion. The reason for using a randomized block design is that a sufficient washout period can be incorporated between treatments to minimize or prevent carryover. The reduction in animal use achieved by using randomized block designs, instead of completely randomized designs, can be 75%, without loss of statistical power (Kinter and Johnsen 1999). A variant of the randomized block design is one in which all animals receive all treatments, but in ascending (vehicle, then low, mid, and high dose) or descending (high, then mid, low dose, and vehicle) order. This design has the same improved statistical power as the randomized block design, but may require an additional time control to separate treatment effects from time effects (Kinter and Johnsen 1999). Because of the growing length of usefulness for radiotelemetry implants (up to a year and longer), investigators may consider reusing experimental animals to study additional doses, different treatments, or alternate routes of administration, or to verify response differences in individual animals. According to Kinter and Johnsen (1999), the total reduction in animal use achieved by eliminating a study, by using randomized blocking, and by reusing animals in multiple studies is potentially greater than 90%. Van Acker et al. (1996) have shown that a reduction in animal use up to 70% can be achieved when the data obtained in their mouse cardiotoxicity studies are compared with literature data obtained via morphometry or histology studies. Finally, Fraser et al. (2001) published a validation study comparing a telemetric BP monitoring system with a tail-cuff BP measuring system in both adrenocorticotropic hormone (ACTH)- and sham-treated rats. It was concluded that the results measured with telemetry in this study were very similar to the results measured with the tail-cuff method. These researchers also concluded that telemetry allows for a longer period of measurement, giving greater power to the study so that fewer animals are needed.
Telemetry and Refinement Alternatives Radiotelemetry also offers refinement alternatives. The technique reduces animal stress and discomfort, improves the quality of experimental data, and reduces costs. In freely moving mice and rats equipped with cardiovascular radiotelemetry transmitters and monitored in their home cages, HR and BP values were found to be much lower when compared to published values (Kramer 2000). Animals implanted with a radiotelemetry transmitter are best monitored while in their home cages, so no additional dedicated laboratory space other than that used to house the animals is necessary. No isolation from other animals of the same species is required, and no continuous monitoring by trained laboratory staff is necessary during the conduct of a study, as is required for almost all other cardiovascular preparations. Since radiotelemetry is useful for investigating drug effects in freely moving, unrestrained animals at any time in their circadian rhythm, radiotelemetry can be considered as a contribution to the concept of refinement alternatives (Lemmer et al. 1997).
84550.indb 459
11/11/10 6:27:45 PM
460
Handbook of Laboratory Animal Science
New and Future Developments Historical and current applications with implanted telemetric monitoring have increased many researchers’ desire for new and innovative approaches with the methodology. Scientists have presented the technology providers with a wide range of transmitter capability requests: for smaller implants to use in the smaller transgenic murine models, for additional sensors to monitor additional parameters like blood gas and glucose levels, for longer battery life for longer studies, for greater transmission range to accommodate larger cages, for group housing to enable social behavior studies, for additional transmission channels to enable more complete profiling of animal physiology, and for Internet-enabled data collection and archive ability. While the recent advances and subsequent growth of small animal telemetry in numerous previously described areas of research science have been stimulated by advances in computer science, biomaterials, and microelectronics, there is one controlling theme: business practicality. New, stateof-the-art technology is very expensive to develop. Implantable telemetry for laboratory animal research applications is the net result of innovative advances in biomaterials, leads, catheters, sensors, integrated circuits, microelectronics, battery science, computer hardware, computer operating systems, and computer analysis software. Each of these technology frontiers contributes to creating components of systems that support implantable telemetry applications in the life sciences. All are very costly to develop and each receives initial funding support from sources other than its potential sales markets within the life sciences research market. Basic science funding grants from government agencies drive a large portion of the funding for new technology research. Computer capabilities of new operating system and analysis software industry advances are funded by large, growing consumer and industrial markets. Costly implantable biomaterials and sophisticated sensor technology development are funded by billiondollar clinical and medical device markets. Current electronic and battery technology advances are funded by large consumer markets for smaller and more power-hungry mobile devices like cell phones, hearing aids, and other portable electronics. As these new technology elements come to life, they can then be integrated into life sciences animal telemetry by vendors utilizing them as critical elements in their small, market-driven life science applications, where hundreds of units, rather than millions, are sold. Implant size is directly related to the combined volume of sensor, electronic circuitry, and battery or power supply. Battery technology still widely dominates as the common, proven, reliable, and functional implant power supply. For the most part, contemporary telemetry implants are limited to the power and volume capacity of current designs of silver and lithium cells. This limitation of power for size challenges the current engineering of implant designs made with stock components and technology. Advances in low-draw, custom-design integrated circuits fight a cost battle for utilization in the smallvolume life science telemetry markets. As this technology lowers in cost and volumes of implants marketed increase, the use of custom integrated chip circuitry will become increasingly commonplace. Transmitter electronic life is directly related to a quotient of battery technology, electronic efficiency, transmission frequency, antenna size, and range required. Larger transmission coils and higher power give greater range, but dramatically shorten battery and transmitter life. This balance of range, life, and frequency is demonstrated in the multiple months of long life in temperature transmitters as compared to the relatively shorter life of mouse BP implants. Manufacturers are unfortunately bound by laws of physics in attempting to meet the future needs of researchers. Inductively recharged implantable medical devices like cochlear implants (Nucleus Freedom Cochlear Implant System: www.cochlearamericas.co) have offered newer technology that is being adapted to power rechargeable telemetry implants. While this approach to implanted inductively recharged battery power still faces size limitations and has requirements for frequent recharging, it offers promise that it may become a practical implant power technology of the future.
84550.indb 460
11/11/10 6:27:45 PM
Applications of Radiotelemetry in Small Laboratory Animals
461
Sensor size, sensor stability, and new sensor capability advancements have been driven by developments for sensor improvements in aerospace, in aviation, and in industrial markets. Similarly, other current advances in sensors for enhanced performance, as well as new capabilities, will come to telemetry from other sciences. Research in microelectromechanical systems (MEMS) sensor science promises exciting outcomes as these are integrated into telemetry. The poster child for MEMS development in the real world is the tire pressure monitoring systems being developed for the auto industry by a number of companies (www.bartecusa.com and others). These common new devices contain a variety of robust advanced components including MEMS pressure sensors, accelerometers, temperature sensors, and custom-designed chips performing battery power management, signal processing, and signal transmission. The host vehicle’s onboard electronic system includes the receiver and advanced software that integrates the solution continuously and in real time for the driver operator. Similarly, new thin-film microelectrode MEMS arrays are being created to serve as electroanalytical systems for detection of parts per billion in reactive chemical environments (www.sensors.stanford.edu). Advances like these may offer tomorrow’s researcher telemetry implants with sensitivity for blood and tissue parameters not yet imagined. Glucose sensor development is exploding as clinical market product demands offer hundreds of millions of dollars in research spending. Sensors from companies like Minimed, Biotex, Animas, Spect RX, Sensors Med, and Science Inc. offer advances and a future opportunity to bring this capability into the life science telemetry markets. Clinical products like “glucowatch” sensors for diabetic online monitoring (Tierney et al. 2001; Tsalikian 2004) offer telemetry an access point to new sensors for blood glucose monitoring. Implantable monitoring devices coming to market for current human applications offer a very near entry point for cross fertilization of new cutting-edge technology advances into life science animal monitoring. Implantable ECG monitoring (Medtronic Reveal ECG loop recorder, www. medtronic.com, implantable pH monitoring (Medtronic Bravo esophageal GERD diagnosis device, www.medtronic.com), and an orally accepted video camera (Given Imaging M2A Camera used in diagnosing internal bleeding, www.given.com) all offer examples of technology that can be brought into the life sciences implantable animal telemetry world. Imagine, all this improved technology integrated into a small implant in your laboratory. You will be working in the future life science laboratory with more implants, new implant capabilities, preimplanted animals, implantable flow monitoring, group housing, animal ID, automatic calibration, new sensor capabilities, increased range, longer or limitless battery life, remote data analysis, Web-enabled networked data storage and access…and much more. We all stand anxiously waiting at the threshold of technology and discovery for bringing new advanced microelectronics into our laboratories.
Conclusions Implantable radiotelemetry is now a viable and commercially available alternative to conventional methods for measurements of physiological data in biomedical research. Currently, well over 800 peer-reviewed publications include data measured via fully implanted telemetry. Several of these publications describe validation studies concerning the accuracy, utility, or safety of the devices (Kramer 2000). The scientific literature provides several compelling arguments for the use of implantable radiotelemetry as an acceptable and, often, more appropriate alternative to traditional methods of monitoring laboratory animals. The most common conventional measurement techniques for monitoring physiological parameters, such as connecting sensors, lead wires, catheters, and tail-cuffs to a restrained
84550.indb 461
11/11/10 6:27:45 PM
462
Handbook of Laboratory Animal Science
or tethered awake animal, are stressful and limited methods and may provide artifactual data (Kramer and Kinter 2003). Special equipment is needed for these techniques, such as isolation chambers and other restraint devices, or the use of protective jackets and collars (including battery-powered monitoring equipment carried in external jackets) and the use of protective tethers and swivels. All this equipment is based upon requirements to protect transcutaneous catheters and electrodes from accidental damage. Failures of these preparations are largely related to damage to these transcutaneous systems and the risk of chronic infection. Also, significant resources must be dedicated to animal training, care of skin penetration sites, and flushing catheters. While these systems have contributed considerably to our knowledge of comparative physiology, pharmacology, and toxicology in the past half century, it is well documented in the literature that these systems are expensive and labor intensive to maintain and inherently unreliable over significant periods of time; they may contribute to animal discomfort and stress, which directly affect the animals’ physiological functions and result in high variability of experimental results or even erroneous measurements. Finally, radiotelemetry technology is providing a more humane approach for monitoring physiological functions in experimental animals. Currently, radiotelemetry can be applied in all commonly used laboratory animal species, from mice to monkeys (Kramer 2000) and even fish (Snelderwaard et al. 2006). Totally implantable miniaturized radiotelemetry systems are valuable for animal models in biomedical research studies because these systems enable the acquisition of otherwise unavailable experimental data (Meile and Zittel 2002) and present substantial reduction and refinement alternatives (Kinter and Johnsen 1999).
References Arras, M., A. Rettich, P. Cinelli, H. P. Kasermann, and K. Burki. 2007. Assessment of post-laparotomy pain in laboratory mice by telemetric recording of heart rate and heart rate variability. BMC Veterinary Research 2 (3): 16. Balakrishnan, S., and J. R. McNeill. 1996. Cardiac output mediates the antihypertensive effect of vasopressin in spontaneous hypertension. Journal of Physiology 27:H1729–H1733. Barone, F. C., D. J. Knudsen, A. H. Nelson, G. Z. Feuerstein, and R. N. Willette. 1993. Mouse strain differences in susceptibility to cerebral ischemia are related to cerebral vascular anatomy. Journal of Cerebral Blood Flow Metabolism 13 (4): 683–692. Baumans, V., J. A. Bouwknecht, H. Boere, K. Kramer, H. A. van Lith, H. A. van de Weerd, and H. van Herck. 2001. Intra-abdominal transmitter implantation in mice: Effects on behavior and body weight. Animal Welfare 10:291–302. Billing, R. L. 2005. The effect of laboratory “stress” on noradrenergic activity in sympathetic tissue of the mouse. PhD dissertation, University of Sydney, Australia. Brockway, B. P. 1998a. Fully implanted radiotelemetry for monitoring laboratory animals. Lab Animal 27:40–46. Brockway, B. P. 1998b. Fully implanted radiotelemetry for monitoring laboratory animals. Lab Animal 27:40–46. Brockway, B. P., and C. R. Hassler. 1993. Application of radio telemetry to cardiovascular measurements in pharmacology and toxicology. In New technologies and concepts for reducing drug toxicity, Ed. H. Salem and S. I. Baskin, 109–132. Boca Raton, FL: CRC Press. Brockway, R. V. 1998. Unpublished data. Data Sciences International, St. Paul, MN. ———. 1999. A new method for chronic measurement of respiratory rate in conscious freely moving rats. Paper presented at the 50th AALAS National Meeting, Indianapolis, IN. ———. 2001. A new method for chronic measurement of respiratory rate in conscious freely moving mice. Personal communication, Data Sciences International, St. Paul, MN.
84550.indb 462
11/11/10 6:27:46 PM
Applications of Radiotelemetry in Small Laboratory Animals
463
Brockway, R. V., and Huetteman, D. A. 1999. A new method for chronic measurement of respiratory rate in conscious freely moving rats. Paper presented at the 50th AALAS National Meeting, Indianapolis, IN. Brockway, R. V., and Huetteman, D. A. 2001. A new method for chronic measurement of respiratory rate in conscious freely moving mice. Personal communication, Data Sciences International, St. Paul, MN. Broom, D. M., and K. G. Johnson. 1993. Stress and animal welfare, ed. D. M. Broom. London: Chapman & Hall. Butz, G. M., and R. L. Davisson. 2001. Long-term telemetric measurement of cardiovascular parameters in awake mice: A physiological genomics tool. Physiological Genomics 5:89–97. Capdevila, S., M. Giral, J. L. Ruiz de la Torre, R. J. Russell, and K. Kramer. 2007. Acclimatization of rats after ground transport to a new animal facility. Lab Anim 41:255–262. Chin, I. 2005. Effects of environmental factors on laboratory mice. PhD dissertation, University of Sydney, Australia. Clement, J. G., P. Mills, and B. Brockway. 1989. Use of telemetry to record body temperature and activity in mice. Journal of Pharmacology Methods 21:129–140. Conley, R. K., and P. H. Hutson 2007. Effects of acute and chronic treatment with fluoxetine on stress-induced hyperthermia in telemetered rats and mice. European Journal of Pharmacology 564 (1–3): 138–145. Cotugno, M., P. Mandile, D. D’Angiolillo, P. Montagnese, and A. Giuditta. 1996. Implantation of an EEG telemetric transmitter in the rat. Italian Journal of Neurological Science 17:131–134. Davisson, R. L., D. S. Hoffmann, G. M. Butz, G. Aldape, G. Schlager, D. C. Merrill, S. Sethi, R. M. Weiss, and J. N. Bates. 2002. Discovery of a spontaneous genetic mouse model of preeclampsia. Hypertension 39 (2): 337–342. De Heer, R., et al. 2010. Simultaneous collection of behavioral and physiological data in rats: Integration of the Noldus Pheno Typer with data sciences telemetry system. In preparation. Duke J. L., T. G. Zammit, and D. M. Lawson. 2001. The effects of routine cage changing on cardiovascular and behavioral parameters in male Sprague–Dawley rats. Contemporary Topics in Laboratory Animal Science 40 (1): 17–20. Einthoven, W. 1957. The electrocardiogram. Trans. H. W. Blackburn. American Heart Journal 53:602–615. Festing, M. F. W. 1992. The scope for improving the design of laboratory animal experiments. Laboratory Animals 28:212–221. Fewell, J. G., H. Osinska, R. Klevitsky, W. Ng, G. Sfyris, F. Bahrehmand, and J. Robbins. 1997. A treadmill exercise regimen for identifying cardiovascular phenotypes in transgenic mice. American Journal of Physiology 273 (42): H1595–H1605. Fraser, T. B., S. W. Turner, G. J. Mangos, J. Ludbrook, and J. A. Whitworth 2001. Comparison of telemetric and tail-cuff blood pressure monitoring in adrenocorticotrophic hormone-treated rats. Clinical Experiments in Pharmacology and Physiology 28 (10): 831–835. Gehrmann, J., P. E. Hammer, C. T. Maguire, H. Wakimoto, J. K. Triedman, and C. I. Berul. 2000. Phenotypic screening for heart rate variability in the mouse. American Journal of Physiology—Heart and Circulatory Physiology 279 (2): H733–H740. Greene, A. N., S. L. Clapp, and R. H. Alper. 2007. Time course of recovery after surgical intraperitoneal implantation of radiotelemetry transmitters in rats. Journal of Pharmacology and Toxicology Methods 56 (2): 218–222. Gross, V., A. F. Milia, R. Plehm, T. Inagami, and F. C. Luft. 2000. Long-term blood pressure telemetry in AT(2) receptor-disrupted mice. Journal of Hypertension 18 (7): 955–961. Gutekunst, C. A., A. Hartney, H. D. Rees, and R. C. Green. 1993. Lack of strain differences in spatial learning among seizure-prone and seizure-resistant mice. Behavior Genetics 23:385–389. Hawkins, P. et al. 2004. Husbandry refinements for rats, mice, dogs and nonhuman primates used in telemetry procedures. Seventh report of the BVAAWF/FRAME/RSPCA/ UFAW Joint Working Group on Refinement, Part B. Lab Anim 38 (1): 1–10. Herd, J. A. 1991. Cardiovascular response to stress. Physiological Reviews 71 (1): 305–330. Ihmsen, H., M. Schywalsky, R. Plettke, M. Priller, F. Walz, and H. Schwilden. 2008. Concentration-effect relations, prediction probabilities (Pk), and signal-to-noise ratios of different electroencephalographic parameters during administration of desflurane, isoflurane, and sevoflurane in rats. Anesthesiology 108 (2): 276–285. Jackson, K., G. A. Head, B. J. Morris, J. Chin-Dusting, E. Jones, L. La Greca, and D. N. Mayorov 2007. Reduced cardiovascular reactivity to stress but not feeding in renin enhancer knockout mice. American Journal of Hypertension 20 (8): 893–899.
84550.indb 463
11/11/10 6:27:46 PM
464
Handbook of Laboratory Animal Science
James, J. F., T. E. Hewett, and J. Robbins. 1998. Cardiac physiology in transgenic mice. Circulation Research 82:407–415. Kaidi, S., F. Brutel, F. Van Deun, K. Kramer, R. Remie, W. Dewé, P. Remusat, A. Delaunois, and O. Depelchin. 2007. Comparison of two methods (left carotid artery and abdominal aorta) for surgical implantation of radiotelemetry devices in CD-1 mice. Lab Anim 41 (3): 388–402. Kinter, L. B. 1996. Cardiovascular telemetry and laboratory animals’ welfare: New reduction and refinement alternatives. Paper presented at the General Pharmacology/Safety Pharmacology Meeting, Philadelphia. Kinter, L. B., and D. K. Johnsen. 1999. Remote monitoring of experimental end points in animals using radiotelemetry and bioimpedance technology. In Humane end points in animal experiments for biomedical research. London: RSM Press. Kinter, L. B., D. J. Murphy, W. A. Mann, T. B. Leonard, and D. G. Morgan. 1997. Major organ system toxicology: An integrated approach to pharmacodynamic safety assessment in animals. In Toxicological testing and evaluation. New York: Elsevier. Kluger, M. J., C. A. Conn, B. Franklin, R. Freter, and G. D. Abrams. 1990. Effect of gastrointestinal flora on body temperature of rats and mice. American Journal of Physiology 258 (27): R552–R557. Kramer, K. 2000. Applications and evaluation of radiotelemetry in small laboratory animals. PhD dissertation, University of Utrecht, the Netherlands. Kramer, K., and L. B. Kinter. 2003. Evaluation and applications of radiotelemetry in small laboratory animals. Physiological Genomics 13 (3):197–205. Kramer, K., L. Kinter, B. P. Brockway, H. P. Voss, et al. 2001. The use of radiotelemetry in small laboratory animals: Recent advances. Contemporary Topics in Laboratory Animal Science 40 (1): 8–16. Kramer, K., and R. Remie. 2005. Measuring blood pressure in small laboratory animals. In Methods in molecular medicine, hypertension: Methods and protocols, ed. J. P. Fennel and A. H. Baker, 51–62. Totowa, NJ: Humana Press Inc. Kramer, K., S. A. B. E. van Acker, H-P. Voss, J. A. Grimbergen, W. J. F. van der Vijgh, and A. Baast. 1993. Use of telemetry to record electrocardiogram and heart rate in freely moving mice. Journal of Pharmacology and Toxicology Methods 30 (4): 209–215. Kramer, K., H. van de Weerd, A. Mulder, C. van Heijningen, V. Baumans, R. Remie, H. P. Voss, and B. F. M. van Zutphen. 2004. Effect of conditioning on the increase of heart rate and body temperature provoked by handling in the mouse. Alternative Laboratory Animal Journal 32 (1): 177–181. Lapray, D., J. Bergeler, E. Dupont, O. Thews, and H. J. Luhmann. 2008. A novel miniature telemetric system for recording EEG activity in freely moving rats. Journal of Neuroscience Methods 168 (1): 119–126. Lemmer, B., et al. 1997. Telemetric data acquisition of cardiovascular function in biology and pharmacology. In Animal alternatives, welfare and ethics, ed. L. F. M. van Zutphen and M. Bals. New York: Elsevier. Leon, L. R., D. Walker, D. A. DuBose, and L. A. Stephenson. 2004. Biotelemetry transmitter implantation in rodents: Impact on growth and circadian rhythms. American Journal of Physiology—Regulatory and Integrative Comparative Physiology 286:R967–R974. Mattes, A., and B. Lemmer. 1991. Effects of amlodipine on circadian rhythms in blood pressure, heart rate, and motility: A telemetric study in rats. Chronobiology International 8 (6): 526–538. Meerlo, P., S. F. De Boer, J. M. Koolhaas, S. Daan, and R. H. Van Den Hoofdakker. 1996. Changes in daily rhythms of body temperature and activity after a single social defeat in rats. Physiology & Behavior 59 (4–5): 735–739. Meijer, M. K., K. Kramer, R. Remie, B. M. Spruijt, L. F. M. van Zutphen, and V. Baumans. 2006. The effect of routine experimental procedures on physiological parameters in mice kept under different husbandry conditions. Animal Welfare 15:31–38. Meile, T., and T. T. Zittel. 2002. Telemetric small intestinal motility recording in awake rats. European Surgical Research 34:271–274. Mills, P. A., D. A. Huetteman, B. P. Brockway, L. M. Zwiers, A. J. M. Gelsema, R. S. Schwartz, and K. Kramer 2000. A new method for measurement of blood pressure, heart rate, and activity in the mouse by radiotelemetry. Journal of Applied Physiology 88 (5): 1537–1544. Mitchell, G. F., A. Jeron, and G. Koren. 1998. Measurement of heart rate and Q-T interval in the conscious mouse. American Journal of Physiology 274 (43): H747–H751. Moran, M. M., R. R. Roy, C. E. Wade, B. J. Corbin, and R. E. Grindeland. 1998. Size constraints of telemeters in rats. Journal of Applied Physiology 85 (4): 1564–1571.
84550.indb 464
11/11/10 6:27:46 PM
Applications of Radiotelemetry in Small Laboratory Animals
465
Mouri, G., E. Jimenez-Mateos, T. Engel, M. Dunleavy, S. Hatazaki, A. Paucard, S. Matsushima, W. Taki, and D. C. Henshall. 2008. Unilateral hippocampal CA3-predominant damage and short latency epileptogenesis after intra-amygdala microinjection of kainic acid in mice. Brain Research 1213:140–151. Paterson, L. M., et al. 2009. Characterization of the effects of caffeine on sleep in the rat: A potential model of sleep disruption. Journal of Psychopharmacology 23 (5): 475–486. Paylor, R., R. Tracy, J. Wehner, and J. W. Rudy. 1994. DBA/2 and C57BL/6 mice differ in contextual fear but not auditory fear conditioning. Behavioral Neuroscience 108:810–817. Russell, W. M. S., and R. Burch. 1959. The principles of humane experimental technique. London: Methuen. Saba, S., P. J. Wang, and N. A. M. Estes III. 2000. Invasive cardiac electrophysiology in the mouse: Techniques and applications. Trends in Cardiovascular Medicine 10:122–132. Sällström, J., M. Carlström, B. L. Jensen, O. Skøtt, R. D Brown, and A. E. G. Persson. 2008. Neuronal nitric oxide synthase-deficient mice have impaired renin release but normal blood pressure. American Journal of Hypertension 21 (1): 111–116. Schnell, C. R., and P. Gerber. 1997. Training and remote monitoring of cardiovascular parameters in nonhuman primates. Primate Report 49:61–70. Sgoifo, A., D. Stilli, D. Medici, P. Gallo, B. Aimi, and E. Musso. 1996. Electrode positioning for reliable telemetry ECG recordings during social stress in unrestrained rats. Physiology & Behavior 60 (6): 1397–1401. Sharp, J., T. Azar, and D. Lawson. 2005. Effects of cage enrichment program on heart rate, blood pressure, and activity of male Sprague–Dawley and spontaneously hypertensive rats monitored by radiotelemetry. Contemporary Topics in Laboratory Animal Science 44 (2): 3240. Shusterman, V., I. Usiene, C. Harrigal, J. S. Lee, T. Kubota, A. M. Feldman, and B. London. 2002. Strainspecific patterns of autonomic nervous system activity and heart failure susceptibility in mice. American Journal of Physiology—Heart Circulation Physiology 282:H2076–H2083. Snelderwaard, P. C. 2006. Surgical procedures for implanting a radiotelemetry transmitter to monitor ECG, heart rate and body temperature in small Carassius auratus and Carassius gibelio auratus under laboratory conditions. Lab Anim 40 (4): 465–468. Sommer, R., M. Meijer, K. Kramer, R. Bulthuis, F. Ohl, and V. Baumans. 2005. Simultaneous collection of behavioral and physiological data in mice: Integration of LABORAS with Dataquest A.R.T. telemetry. Paper presented at Measuring Behavior 2005, Wageningen, the Netherlands (http://www.noldus.com/mb2005). Stauss, H. M., A. Gödecke, R. Mrowka, J. Schrader, and P. B. Persson 1999. Enhanced blood pressure variability in eNOS knockout mice. Hypertension 33 (6): 1359–1363. Stemkens-Sevens, S., K. van Berkel, I. de Greeuw, B. Snoeijer, and K. Kramer 2009. The use of radiotelemetry to assess the time of acclimatization of guinea pigs after several hours of transport. Lab Anim 43:78–84. Stiedl, O., J. Radulovic, R. Lohmann, K. Birkenfeld, M. Palve, J. Kammermeier, F. Sananbenesi, and J. Spiess. 1999. Strain and substrain differences in context- and tone-dependent fear conditioning of inbred mice. Behavioral Brain Research 104 (1–2): 1–12. Stiedl, O., and J. Spiess. 1997. Effect of tone-dependent fear conditioning on heart rate and behavior of C57BL/6N mice. Behavioral Neuroscience 111 (4): 703-711. Tang, X., and L. D. Sanford. 2002. Telemetric recording of sleep and home cage activity in mice. Sleep 15 (6): 691–699. Tierney, M. J., J. A. Tamada, R. O. Potts, L. Jovanovic, S. Garg, and Cygnus Research Team. 2001. Clinical evaluation of the GlucoWatch (R) biographer: A continual, noninvasive glucose monitor for patients with diabetes. Biosensors and Bioelectronics 16 (9–12): 621–629. Til, H. P., C. F. Kuper, and H. E. Falke. 1997. Nitrite-induced adrenal effect in rats and the consequences for the no-observed-effect level. Food Chemistry Toxicology 35:349–355. Tsalikian, E. 2004. Accuracy of the GlucoWatch G2 biographer and the continuous glucose monitoring system during hypoglycemia: Experience of the diabetes research in children network. Diabetes Care 27 (3): 722–726. Van Acker, S. A. B. E., K. Kramer, E. E. Voest, J. A. Grimbergen, J. Zhang, W. J. F. van der Vijgh, and A. Bast. 1996. Doxorubicin-induced cardiotoxicity monitored by ECG in freely moving mice. A new model to test potential protectors. Cancer Chemotherapy and Pharmacology 38:95–101. Van Dongen, J. J. et al. 1990. In Manual of microsurgery on the laboratory rat. Amsterdam: Elsevier.
84550.indb 465
11/11/10 6:27:46 PM
466
Handbook of Laboratory Animal Science
Van Loo, P. L., N. Kuin, R. Sommer, H. Avsaroglu, T. Pham, and V. Baumans. 2007. Impact of “living apart together” on postoperative recovery of mice compared with social and individual housing. Lab Anim 41 (4): 441–455. Vleeming, W., A. Van De Kuil, J. D. te Biesebeek, J. Meulenbelt and A. B. T. J. Boink. 1997. Effect of nitrite on blood pressure in anaesthetized and free-moving rats. Food Chemistry Toxicology 35:615–619. Weiergräber, M., M. Henry, J. Hescheler, N. Smyth, and T. Schneider. 2005. Electrocorticographic and sleep intracerebral EEG recording in mice using a telemetry system. Brain Research Protocols 14:154–164. Yang, L., X. Tang, L. L. Wellman, X. Liu, and L. D. Sanford. 2009. Corticotropin releasing factor (CRF) modulates fear-induced alterations in sleep in mice. Brain Research 1276:112–122.
84550.indb 466
11/11/10 6:27:46 PM
Chapter 16
Immunization for Production of Antibodies
Jann Hau and Coenraad Hendriksen Contents Introduction..................................................................................................................................... 467 The Immune Response...................................................................................................................468 Antibodies....................................................................................................................................... 470 Polyclonal Antibodies versus Monoclonal Antibodies................................................................... 470 Production of Antibodies................................................................................................................ 471 The Antigen................................................................................................................................ 471 The Adjuvant.............................................................................................................................. 472 The Animal................................................................................................................................ 475 Species................................................................................................................................... 475 Species/Strain–Stock............................................................................................................. 476 Sex......................................................................................................................................... 476 Age........................................................................................................................................ 477 The Immunization Protocol....................................................................................................... 477 The Injection Route............................................................................................................... 477 The Volume of Injection........................................................................................................ 477 The Quantity of Antigen....................................................................................................... 479 Booster Immunization........................................................................................................... 479 Blood Collection........................................................................................................................480 General Aspects..............................................................................................................................480 Existing Guidelines for the Production of Polyclonal Antibodies............................................. 481 New Developments in Antibody Production............................................................................. 481 References....................................................................................................................................... 481 Introduction Many breakthroughs in biomedical science have been achieved by the use of polyclonal and monoclonal antibodies, and the development of quantitative radioimmunoassay and enzyme immunoassay technology has replaced millions of live animals, previously used in bioassays to estimate concentrations of hormones and enzymes. One of the earliest and probably best known examples of polyclonal antibodies’ importance in medical advancement is the discovery by Behring and Kitasato 467
84550.indb 467
11/11/10 6:27:47 PM
468
Handbook of Laboratory Animal Science
(1890) of the therapeutic effects of diphtheria antiserum. The history of monoclonal antibodies, however, is much more recent and dates to the pioneering work of Köhler and Milstein in 1975. Today, polyclonal antibodies (Pabs) and monoclonal antibodies (Mabs) are indispensable tools in the laboratory. They are used for immunoassays (e.g., as a diagnostic tool), for affinity chromatography, as immunomarkers (e.g., in pathology), and in basic research (e.g., to discover new proteins and to characterize complex antigenic structures). Furthermore, they are of crucial value in the clinic. Mabs are increasingly being used as carriers in drug-targeted therapy. Although antibiotics, recombinant monoclonal antibody therapy, and vaccines have replaced most therapeutic polyclonal antisera, some, such as rabies antiserum and snake antivenom, are still important in third-world countries. However, a clinical need of polyclonal antibodies for treatment of infections and various other diseases certainly remains, and polyclonal antibody-based therapeutics are still widely used in the clinic for neutralization of viruses and toxins and for replacement treatment of patients with immunoglobulin deficiencies (Newcombe and Newcombe 2007). There are welcome initiatives to develop production systems for antibodies in plants (De Muynck, Naavarre, and Boutry 2010) and human antibodies from mice carrying human chromosome fragments (Ishida et al. 2002), but the present chapter focuses on classical production of antibodies from animals immunized with an antigen and subsequently bled for antibody production. The production of Pabs as well as Mabs requires the use of substantial numbers of animals and, because of the invasiveness of the experimental procedure, the animals are protected by legislation regulating the use of laboratory animals. In the case of Pabs, animals are immunized to induce an optimal humoral immune response, and in the case of Mabs, animals are needed to obtain antigen-specific B-lymphocytes and to produce ascites if in vitro production is not possible. Production of Mabs in mice through harvest from abdominal ascites fluid is discouraged for animal welfare reasons, and there are a number of guidelines available on the Internet addressing this issue, such as NIH guidelines (http://oacu.od.nih.gov/ARAC/documents/Ascites.pdf) and guidelines issued by the Australian government (http://www.nhmrc.gov.au/_files_nhmrc/file/publications/synopses/ea19.pdf.). Animal welfare issues associated with ascites production have been addressed by Hendriksen and Hau (2003) and will not be addressed in the present chapter. The procedures employed to generate Pabs and Mabs are frequently under discussion for animal welfare reasons. In particular, certain aggressive immunostimulatory products (adjuvants) used to enhance the immune response and some routes and locations of immunization are considered to be associated with pain and distress. In addition, animals used for production of Pabs may spend a long time being regularly boosted and bled in the animal house. The aims of this chapter are to discuss the factors that influence the production of Pabs and Mabs and to provide best-practice principles for antibody production methods. Furthermore, approaches to reduce, replace, and refine the use of animals will be presented and discussed. But, first, an overview of the immune response is given to provide a basis for understanding of the mechanism underlying the induction of antibody synthesis. Additional information may be obtained from any of several immunology textbooks. After presenting and discussing the many parameters in immunization protocols and the production of antibodies, a summary will be presented of existing guidelines for in vivo Pab and Mab production. The Immune Response In the intact vertebrate organism, an immune response is triggered by contact of the organism with an antigen, which is a structure recognized by the immune system as foreign (nonself). (Note: In this text, the term antigen is used instead of the term immunogen. Immunogens are defined as substances that are able to induce an immune response, whereas an antigen is a substance that can
84550.indb 468
11/11/10 6:27:47 PM
Immunization for Production of Antibodies
469
react with an antibody. The two are not always synonymous, but often no distinction is used in the literature.) Antigens can be presented to the immune system as complex (particulate) multiantigens, such as bacteria, viruses, parasites, and artificial particles, or as single antigens such as proteins or polysaccharides (Leenaars, Claassen, and Boersma 1996). The immune system, responsible for the immune response, can be broadly divided into the innate system and the acquired (adaptive) system; each is composed of cellular and humoral elements. Interactions between the two are important for the functioning of both systems. The innate system, which appears to be evolutionarily older, does not have a high level of specificity and efficiency, but responds rapidly. Elements that are involved include macrophages, dendritic cells, and humoral factors such as the complement activation system. The adaptive-antigen specific—immune system, on the other hand, has a high level of specificity and efficiency, but it responds more slowly (Hanly, Artwohl, and Bennett 1995). The principal actors in the antigen-specific immune response are the B-lymphocytes, responsible for antibody production, and the T-cells (thymus-derived lymphocytes), responsible for cytotoxic responses (cytotoxic T-cells, Tc-cells) and T helper cells (T h-lymphocytes, Th-cells) for B-cells and Tc-cells. Finally, the maintenance of the immune system requires intercellular communication mediated through cell-to-cell contact (e.g., by the production of cytokines) and the support of so-called accessory cells such as fibroblasts and endothelial cells. The antigen-specific immune response can be divided into three phases: the inductive phase, the effector phase, and the establishment of immunological memory (McCullough, Hendriksen, and 1997). In the inductive phase, an antigen surviving the primary defenses (e.g., skin, mucosa) of the organism is recognized as “foreign” by so-called “antigen-presenting-cells” (APCs: monocytes, dendritic cells, and B-lymphocytes). Dendritic cells recognize all antigens, while B-lymphocyte APCs employ a specific receptor-dependent form of antigen binding. The APC binds and internalizes the antigen and processes it into peptides that, together with MHC class molecules, are then exposed on the surface of the APC for presentation to the antigen receptors on the T-cells. This contact leads to activation of the T-cells. After antigenic contact, only a small percentage of the lymphoid cells will become involved in the immune response—initially, less than 10 per 1,000,000 cells, increasing to 1 in 1,000 cells within 8 days of antigen-dependent activation, if the antigen is a potent immunogen. The rest of the cells remain dormant. The next step in the immune response is the effector phase. Depending on the type of antigen, activation of T-cells leads to cell-mediated responses (mainly cytotoxic, such as is the case for viral antigens) or to T-cell help for B-cells (i.e., in case of protein antigens). Once activated, T h-cells become more sensitive to the action of growth factor cytokines, such as IL-1 and IL-2. In the mouse, at least three kinds of T h-cell subsets can be recognized: Th0 -, T h1-, and T h2-cells. These cells also produce cytokines; each subpopulation generates different sets of cytokines, which modulate the immune response. A T h1 response is predominantly cell mediated, whereas a T h2 response is dominated by humoral factors. Cytokines induce the proliferation and, ultimately, differentiation of B-lymphocytes into antibody-producing plasma cells. Each plasma cell is genetically encoded to produce an exactly defined antibody with specificity against a single antigen epitope. The final phase of the antigen-specific immune response is the induction of memory after primary contact with the antigen. Memory, which is based on the differentiation of B-cells, gives the secondary immune response its characteristics of speed, higher magnitude, and avidity compared to the primary response, and dominance of the IgG antibody subclasses over IgM (McCullough et al. 1997). In the mouse, Th1-cells bias the secondary immune response toward cell-mediated immunity and the IgG2a subclass of antibody, while Th2-cells bias the response toward humoral immunity and IgE plus the IgG1 subclass of antibodies.
84550.indb 469
11/11/10 6:27:47 PM
470
Handbook of Laboratory Animal Science
Antibodies Antibodies, also called immunoglobulins or gamma-globulins, are of fundamental importance for the identification and elimination of foreign material that enters the organism. In mammals, antibodies belong to one of five immunoglobulin (Ig) classes: IgA, IgD, IgE, IgG, or IgM; each has a specific range of activities and characteristics. In avian species, immunoglobulins belong to one of three classes: IgA, IgM, or IgG/IgY. The IgG/IgY molecule is functionally similar to, but differs structurally from, the mammalian IgG molecule. IgA and IgG may be further subdivided into subclasses in some species. The core structure of all Ig molecules comprises two identical, covalently linked heavy (H) chains and two identical light (L) chains. Each chain has a variable (V) and a constant (C) region. Antibody molecules are bifunctional, having an antigen binding site and a biological effector function site. The heavy and light chain V regions’ domains form the antigen-binding sites and differ in structure from one antibody specificity to the next (between clone variations). The heavy chain C regions (Fc segment) are relatively constant in structure within each subclass of antibodies and contain the biological effector functions, such as complement activation. Polyclonal Antibodies versus Monoclonal Antibodies The basic difference between Mabs and Pabs is implied in the terminology. Mabs are antibodies obtained from one plasma cell clone. By definition, antibodies produced by a single clone are monospecific; they have specificity directed against only one antigen determinant. Generally, however, an antigen epitope consists of perhaps 5–10 amino acids. And, because most antigens are much larger, they normally constitute a number of different epitopes, each inducing a plasma cell clone (polyclonal) producing antibodies against a particular epitope. Therefore, the antibody response against such antigens normally consists of a combination of a number of monoclonal antibodies (sometimes hundreds to thousands). Some characteristics of Pabs and Mabs are given in Table€16.1. Mabs are antibodies produced by a single B-cell clone when an antigen epitope interacts with the immune system. Normally, B-cells are not able to survive outside the body. However, in 1975, Köhler and Milstein discovered that by fusion of a B-cell with a myeloma cell, they were able to immortalize the cell. These cells, called hybridoma cells, could produce virtually unlimited quantities of (monoclonal) antibodies.
Table€16.1╅Characteristics of Polyclonal and Monoclonal Antibodies Characteristic
Polyclonal Antibodies
Monoclonal Antibodies
Specificity Sensitivity for antigenic modification (e.g., genetic polymorphism, heterogeneity of glycosylation, denaturation) Affinity/avidity
Low to high Low
High High
Avidity equals sum of affinity of individual specificities; increases with time Unlikely after absorption Usually 4–8 weeks Yes
Avidity = affinity; constant
Cross reactivity with other antigens Production time Batch-to-batch variability
Likely Up to 3–6 months In principle, not
Source: From Hendriksen, C. F. M., and J. Hau. 2003. In Handbook of Laboratory Animal Science, vol. I: Essential Principles and Practices, ed. J. Hau and G. Van Hoosier. Boca Raton, FL: CRC Press.
84550.indb 470
11/11/10 6:27:47 PM
Immunization for Production of Antibodies
471
Mabs have become important tools in the laboratory and in the clinic and are used as therapeutic agents, components of kits for immunoassay and affinity chromatography, as carriers in drugtargeted therapy, and in basic research. There are more than 100,000 different Mabs available, and most are produced in small quantities (<0.1 g) solely for bench-related purposes. Larger amounts are often required for use in diagnostic kits and reagents (0.1–0.5 g), for routine diagnostic procedures, in preclinical evaluation studies (0.5–10 g), and for prophylactic and therapeutic purposes (>10 g). When making a choice between the generation of Pabs or Mabs, the desired application of the antibody and the time and money available for its production should be taken into consideration. A polyclonal antiserum can be obtained within a month or two, whereas it takes about 3–6 months to produce Mabs (Leenaars and Hendriksen 2005) In research, while many questions can be answered utilizing Pabs, Mabs are preferred for diagnostic purposes and large-scale in vitro production systems. Single Mabs cannot cross-link antigens and are thus not usable in precipitation assays. However, this problem can be overcome by mixing different Mabs with specificities against different epitopes raised against the antigen.
Production of Antibodies Scientists performing an immunization procedure often want a number of conditions to be met, such as that the antibody titer should be high and be obtained within a reasonable time, the antibody response should be specific, the antibodies induced should be of high avidity, and the amount of antibody produced should be sufficient for the intended purpose. The aim of this section is to identify these factors and to ensure proper immunization procedures, which combine acceptable immunological results with minimal pain and suffering for the animals. Generally speaking, an immunization procedure includes a number of critical factors that may influence the outcome of the immunization procedure and the welfare of the animals. They include the following: the antigen the use of an adjuvant the choice of animal immunization protocol (route of injection, volume, boosting schedule) blood collection
The Antigen Antigens can be single molecules like proteins or particulate complex multiantigens. The group of particulate complex antigens includes intact microorganisms like bacteria, viruses, parasites, protozoa, and mammalian cells, as well as artificial particles. The single antigens include proteins, peptides, and polysaccharides. Antibodies to carbohydrates are generally more difficult to produce than antibodies against proteins and peptides because of the T-cell independent response to carbohydrates (Heimburg-Molinaro and Rittenhouse-Olson 2009). Some antigen features that influence immunogenicity are shown in Table€16.2. The intensity of the immune response is directly related to the size of the antigen and the degree of foreignness. Particulate antigens are generally very immunogenic, while single antigens are less immunogenic. Small antigens (<3–4 kDa) are not able to elicit an antibody response in themselves (Poulsen, Hau, and Kollerup 1987; Poulsen et al. 1990) and need a protein-carrier molecule to activate T-cells. A rule of thumb is that the bigger the molecule is, the better the antibody response will be. The molecule should be in a form such that it can be easily processed by the cells of the
84550.indb 471
11/11/10 6:27:47 PM
472
Handbook of Laboratory Animal Science
Table€16.2╅ General Features Influencing Immunogenicity of Antigens Parameter Source Chemical nature Size Complexity Stability Degradability
Immunogenicity Xenogenic > allogenic > syngenic > autologous Protein > polysaccharide > lipid High molecular weight > low molecular weight Complex > simple Stable > unstable Degradable > undegradable
Source: From Hendriksen, C. F. M., and J. Hau. 2003. In Handbook of Laboratory Animal Science, vol. I: Essential Principles and Practices, ed. J. Hau and G. Van Hoosier. Boca Raton, FL: CRC Press.
immune system; but on the other hand, it should have a certain structural stability and be present in more or less the same form in booster injections. In order to be a good immunogen, the molecule must possess at least one epitope that can be recognized by the cell surface antibody found on B-cells, and it must have at least one surface structure that can be recognized simultaneously by a class II protein and a T-cell receptor. Furthermore, the greater the phylogenetic difference between the antigen donor and the animal to be immunized is, the better the immune response that is evoked will be. An important aspect to consider is the quality of the antigen. When raising Pabs, as opposed to Mabs, the purity of the antigen is of major importance. Minute impurities may prove to be immunodominant and will lower the immune response wanted (Leenaars and Hendriksen 2005). It can also be time consuming and laborious to purify the antigen. However, the resources are usually well spent because it is much easier to render an antiserum only weakly contaminated with antibodies against impurities functionally monospecific than to remove a lot of unwanted specificities by extensive absorption procedures. Antigen preparations can be toxic due to contamination with endotoxins or chemical residues used to inactivate microorganisms (e.g., formaldehyde or propiolactone) or an extreme pH (Leenaars et al. 1999). There may be human health and safety issues associated with handling and administration of certain antigens, and antigen preparations should ideally be prepared and stored according to the rules of good laboratory practice. The Adjuvant Adjuvants (from the Latin word adjuvare = help) are substances used to enhance and modulate the immune response. They have been used since the 1920s (Ramon 1925), and the ideal adjuvant can be characterized as a product that stimulates high and sustainable antibody titers, is efficient in a variety of species, is applicable to a broad range of antigens, is easily and reproducibly prepared in an antigen mixture, is easily injectable, is effective in a small number of injections, has a low toxicity for the immunized subject, and is not hazardous to the investigator. Unfortunately, the adjuvant that meets all of these criteria remains to be identified. There are more than 100 known adjuvants, but only some of these are routinely utilized. Depending on their mode of action or composition, adjuvants can be categorized in a number of groups (Table€16.3). The adjuvant products most frequently used are the water-in-oil (W/O) emulsions—in particular, Freund’s adjuvant (Freund, Casals, and Hosmer 1937), as can be seen from a survey performed in the Netherlands (Figure€16.1). Freund’s incomplete adjuvant (FIA) is an oilin-water emulsion composed of mineral oil (Bayol F) at 85–90% and a detergent (Arlacel A; mannide monooleate) at 10–15%. Freund’s complete adjuvant (FCA) is FIA to which heat-killed, dried Mycobacterium species—usually M. tuberculosis, M. smegmatis, or M. butyricum—has been added. The use of FCA is favored because it gives rise to long-term persistence of the immune responses;
84550.indb 472
11/11/10 6:27:47 PM
Immunization for Production of Antibodies
473
Table€16.3╅Overview of Categories of Adjuvants That May Be Used for Routine Polyclonal Antibody Production Category
Examples
Mineral salts Oil emulsions (W/O, O/W, W/O/W)
Al(OH)3, AlPO4 FIA, Specol, Montanide
Microbial (like) products
LPS, MDP, MPL, TDM
Saponins
Quil-A
Synthetic products
DDA Iscoms
Liposomes NBP Cytokines
IL-2, IL-1, IFN-γ
Adjuvant formulations
FCA, TiterMaxTM, RIBI, Gerbu, Softigen
Mode of Action Vehicle, depot effect Vehicle, depot effect, activation of macrophages Stimulation B- or T-cells, activation of macrophages, enhanced antigen uptake Facilitate cell–cell interaction, aggregation of antigen Activation of macrophages and complement Facilitate cell–cell interaction, stimulation of T-cells, aggregation of antigen Vehicle, enhanced antigen uptake Activation of macrophages and complement Growth and differentiation of B- and T-cells Combinations of the preceding
Source: From Hendriksen, C. F. M., and J. Hau. 2003. In Handbook of Laboratory Animal Science, vol. I: Essential Principles and Practices, ed. J. Hau and G. Van Hoosier. Boca Raton, FL: CRC Press. Notes: W/O = water in oil; O/W = oil in water; W/O/W = water in oil in water. FIA = Freund’s incomplete adjuvant; LPS = lipopolysaccharide, MDP = muramyl dipeptide; MPL = monophosphoryl lipid A; TDM = trehalose dimycolate; DDA = dimethyldioctadecylammonium bromide; ISCOMs = immunostimulating complexes; NBP = nonionic block polymer; FCA = Freund’s complete adjuvant.
although this is not common practice in research laboratories, frequent booster doses (Hohmann 1998) are not always needed because of the depot effect of FCA and FIA. When antigenic material in a water solution is injected into an animal, a prompt and rapid dissemination of the injected material occurs. Alum precipitation or adsorption of antigens, as in many vaccines used in humans, causes only a slight retention of the antigen. However, if the same material is injected as a water-in-oil emulsion (with or without mycobacteria), the oil vesicles can be retained
% Responders that use Adjuvant
45 40
FC A
35
IFA
30
Alum
25
Iscoms
20
TiterMax
15
Specol
10
MPL
5 0
Quil-A Adjuvant
Figure 16.1â•…The use of adjuvant. Percentage of responders using a specific adjuvant product in mice. (From Leenaars, P. P. A. M., and C. F. M. Hendriksen. 1999. With permission.)
84550.indb 473
11/11/10 6:27:50 PM
474
Handbook of Laboratory Animal Science
at the site of injection for many months while the material undergoes slow degradation. The rate of antigen elimination from the injection site has been reported to have a half-life of approximately 14 days. The antibody response to emulsified protein antigens seems to be relatively constant for up to a year (Talmage and Dixon 1953). However, the persistence of cell-mediated reactions, particularly at the site of the original depot, may lead to tissue inflammation and granulomas, which may become necrotic (Kittell, Banks, and Hadick 1991). Moreover, an additional finding might be the presence of granulomas in lungs, kidneys, liver, heart, lymph nodes, and skeletal muscle due to adjuvant/antigen dissemination (Hau 1988; Kittell et al. 1991). This phenomenon might be explained by uptake of the emulsion by macrophages and transport by the lymphatics (Leenaars 1997). FCA is widely used to induce autoimmune diseases in animal models, and obviously its use in immunization may result in different autoimmune abnormalities in the immunization animals. In chickens, the use of FCA should be discouraged because it results in a reduction in the egg laying frequency and thus antibody productivity, as compared to chickens immunized with other adjuvants including FIA (Bollen and Hau 1999). Freund’s adjuvant-induced arthritis constitutes the only animal model of chronic pain that has been validated to a significant extent (Colpaert 1987; Erb and Hau 1994), and the side effects of adjuvants, in particular FCA, have led to serious concern for the welfare of the animals. A number of studies have been undertaken to replace FCA by adjuvant products with less irritant properties. Unfortunately, only a few products have been demonstrated to have less undesirable effects while being as immunogenic as FCA. However, efficient alternative adjuvants are now available not only for stimulating humoral immunity but also for stimulating cell-mediated immunity (Herbert 1967; Lindblad and Hau 2000) and for immunomodulation (Table€16.3). These include the RIBI adjuvant system containing squalene; monophosphoryl lipid A; trehalose dimycolate; saponins; dextran polymer particles; ethylene-vinyl acetate polymers; muramyl dipeptide (MDP); pegylated C8/C19 mono- and diglycerides; cholera toxin B subunit; Montanide; Specol; and ISCOMs (immunostimulating complexes). Adjuvant products like the mineral salts (e.g., AlPO4), the only adjuvants allowed for human application, are almost without adverse effects, but the immunostimulatory effects are not as impressive as for many of the adjuvants used for experimental purposes. More detailed information on adjuvants can be obtained from several review articles and books (Lindblad and Sparck 1987; Claassen and Boersma 1992; Stewart-Tull 1995; Jennings 1995). Some evidence suggests that FCA is a human health hazard because of unpleasant complications for sensitized personnel who accidentally injure themselves while immunizing the animals. Even if the skin is pricked accidentally and no injection made, this can result in a painful and dramatic swelling of the area, and the wound may heal slowly and discharge up to 2 years after the incident (Chapel and August 1978). One of the first questions that might be asked when an immunization procedure is set up is whether an adjuvant is needed. In case of particulate antigens, it might be possible to obtain a high antibody titer without the use of an adjuvant. When an adjuvant is required, it is up to the investigator to decide which type is the most appropriate. When specific immunomodulation is part of the immunization objectives, ISCOMs, Quil-A, and other modern adjuvants should be considered. However, the overall welfare of the laboratory animal to be immunized must be taken into consideration. Because it is known that oil adjuvants combined with bacterial components (e.g., FCA) may induce considerable side effects and production of many irrelevant antibodies against bacterial components, the use of these adjuvants should be discouraged. If they are used, then the administration of the adjuvant should comply with certain restrictive conditions, such as a limitation to the volume injected, the location of injection, and the number of booster injections. FCA should never be injected twice, and it would normally be administered with the first immunization of the animal. Repeated injections of FCA may lead to severe tissue reactions and may result in anaphylactic shock if the animal has become sensitized to any of the bacterial immunogens. The antibodies against
84550.indb 474
11/11/10 6:27:50 PM
Immunization for Production of Antibodies
475
the bacterial components of complete adjuvants can disturb the use of the antiserum, which is an additional reason for avoiding the use of these types of adjuvants. When cholera toxin is used as an adjuvant, the antibody response against the adjuvant and the antigen seems to be positively correlated (Mayo, Royo, and Hau 2005). The preparation of the antigen and adjuvant mixture should occur lege artis. First, the mixture should be prepared aseptically, to minimize the risk of possible contamination. When an oil emulsion is used, the stability and quality of the emulsion should be checked. In case of a water-in-oil emulsion, drops of the mixture placed on the surface of water in a dish will remain as discrete white drops on or just below the surface, indicating that the water phase containing the antigen is entirely enclosed within the oil (Lindblad and Sparck 1987). If, on the other hand, the emulsions form a cloud of tiny particles when dropped in the water, it is an oil-in-water emulsion. Water-in-oil emulsions that are not properly prepared are ineffective as adjuvants (Freund 1947). The time between the preparation of the mixture and its administration should ideally be short. If stored, the mixture should be put in a freezer; the emulsion should be checked after it thaws and before the injection is given. The Animal Species An essential issue to consider when producing Pabs is the selection of the animal species and animal strain. The rabbit is by tradition the most widely used species, but the choice of species should depend on the purpose of the immunization. Many factors may influence which species will be the most appropriate for a specific purpose: for example, the volume of antibody or antiserum needed and the ease to obtain large blood samples, the phylogenetic relationship between the recipient and the donor of the antigen, the character of the antibody synthesized by the recipient species, and the intended use of the antibody (Hanly et al. 1995). When large amounts of antibody are needed, farm animals such as sheep or goats are preferred. Mice and rabbits are most frequently used when small amounts are required, particularly because these animal species are easy to bleed compared to, for example, guinea pigs and hamsters. Because of the continuous transovarian passage of antibodies from blood to egg yolk in birds (Bollen and Hau 1997), it is convenient to harvest and purify antibodies from the egg yolk of the domestic fowl, and several fairly simple methods have been described (Jensenius et al. 1981; Svendsen et al. 1995; Svendsen Bollen et al. 1996; Schade et al. 1996, 2001; Bizhanov, Jonauskiene, and Hau 2004). Chicken egg antibodies (IgYs), which are phylogenetically a progenitor of IgG antibodies, can be extracted from the egg to concentrations of approximately 100 mg IgY/egg. Compared to the IgG productivity of the rabbit (approximately 200 mg Ig/bleeding), a chicken produces about 10 times as much IgY. In addition, IgY does not cross-react with mammalian immunoglobulins, thereby reducing the risk for false positive results in, for example, an ELISA. Antibody production against highly conserved antigens (such as intracellular proteins) requires a wide phylogenetic distance between the recipient and the donor animal (e.g., the use of chickens for producing antibodies to mammalian proteins). Chicken antibodies raised against a protein in one mammalian species will often react against the analogous protein in other mammalian species (Hau et al. 1980, 1981). The use of chickens for production of antibodies is attractive from an ethical viewpoint and with respect to Russell and Burch’s (1959) principle of the three Rs to replace, reduce, and refine the use of laboratory animals when possible. Mammals can thus be replaced by a species with a lower degree of neurophysiological sensitivity, and the number of animals needed can be considerably reduced. Oral immunization techniques are being developed (Persdotter Hedlund and Hau 2001; Mayo, Tufuesson, et al. 2005; Mayo, Royo, et al. 2005; Mayo et al. 2009), thus eliminating restraint and distress associated with administration of antigen and harvest of antibodies and refining the
84550.indb 475
11/11/10 6:27:50 PM
476
Handbook of Laboratory Animal Science
methodology so that this production of antibodies may no longer be considered an experimental procedure from a legal point of view. However, although some commercial companies have specialized in the production of avian antibodies, the use of IgY technology is not widespread. This is probably due to a number of factors, such as tradition, the limited availability of conjugated antibodies directed against IgY, and the infrequent use of the chicken as a laboratory animal, in general, and its specific requirements for housing. Species/Strain–Stock The antibody response to most antigens seems to be genetically determined and under polygenic control involving histocompatibility antigens (Long 1957; Cannat et al. 1979; Lifschitz, Schwartz, and Mozes 1980; Mayo, Tufuesson, et al. 2005). A species and also often strain difference are thus to be expected. Minimal interindividual variation in antibody response is observed if inbred strains of animals are used in immunization series using the same antigen and immunization procedure. If this minimal variation is important, mice, rats, or chickens, which are all readily available as inbred strains, may be preferred. Outbred stocks like rabbits, although selected through generations for high antibody response, have been documented to exhibit a remarkable interindividual variation in their immune response (Harboe and Ingild 1983). Even animal strains may differ in their immune response. For example, BALB/c mice tend to be Th2-like responders, while C57BL/6 mice are Th1 responders. In certain instances, the genetic control of the immune response seems to be dependent on antigen dose, and increasing doses can transform a poorly responding mouse strain into a good responder (Young and Atassi 1982). For Mab production, the choice of animal species and strains as immune spleen cell donors for fusion is largely dependent on the myeloma cell line available and the origin of the antigen. The mouse is the animal most commonly used for immunization because a variety of mouse myeloma cell lines are available. BALB/c mice are generally used for immunization because many of the myeloma cells available for fusion are of BALB/c origin. In case BALB/c mice are unable to produce the B-cell clone of interest or in specific cases, B-cells can also be obtained from other species, such as the rat, hamster, and human. The immunization protocols are quite similar to those for the production of Pabs. However, at the end of the immunization period (generally 3 days after the booster immunization), the animal is killed, and the spleen is removed for isolation of B-cells. Instead of the spleen, other lymphoid tissue, such as mesenteric or peripheral intestinal lymph nodes, can also be used. Sex Although there are no scientific reasons for not using male animals for antibody production (with the exception of chickens), traditionally, female animals are preferred. Females are generally more docile and less aggressive in social interactions and can therefore be group housed more successfully. Group housing has also been reported for castrated male rabbits, but there may be ethical problems associated with castrating animals in order to make them easier to group house. Modern guidelines for the care of laboratory animals like the CoE (ETS 123; Appendix A, recently revised) advocate group housing whenever possible. However, group housing of male animals is not unproblematic and may be associated with stress and elevated corticosteroid levels. A number of reports have demonstrated that social stress may impair the initial immune response in immunized animals (Abraham et al. 1994). A sex difference in the magnitude of antibody response within inbred strains has been observed (Kaplan, Caperna, and Garvey 1981), and the humoral immune response is suppressed during pregnancy in the mouse (Poulsen and Hau 1990).
84550.indb 476
11/11/10 6:27:50 PM
Immunization for Production of Antibodies
477
Age The outcome of an immunization procedure depends on the immune status of the animal. In general, young adult animals are better responders because they are naive to immunostimulatory agents in their environment. The immune response declines with advancing age after young adulthood. Chickens should be of egg-laying age by the time antibody is harvested, but there is no great difference in titers obtained in young egg-layers compared with older hens about to cease egg-laying (Bollen and Hau 1996). The following minimum ages are recommended for Pab production: mice and rats at 6 weeks; rabbits and guinea pigs at 3 months; chickens at 3–5 months; goats at 6 or 7 months; and sheep at 7–9 months. The Immunization Protocol The Injection Route The location at which the antigen is deposited in part determines the lymphoid organs activated and the type of antibody response that will be induced. However, other aspects need to be taken into consideration when the route of injection is selected: the species used, the quantity of the antigen and adjuvant mixture, the choice of the adjuvant, and the welfare of the animals. The choice often seems mostly determined by tradition. Frequently used routes are subcutaneous (s.c.), intramuscular (i.m.), intraperitoneal (i.p.), intravenous (i.v.), and intradermal (i.d.). Some characteristics of these routes as well as advantages and disadvantages are summarized in Table€16.4. Alternative routes that are being used include oral administration, intranasal inoculation, footpad injection, and intrasplenic injection. Oral administration is preferred when using specific adjuvants, such as lactobacillus, and when the requirement is a peripheral as well as a mucosal immune response. Intranasal administration, with antigen in aqueous solution, is used for the induction of tolerance. Animals usually have to be anesthetized and the material is given by pipette or microcanula (Nielsen, Poulsen, and Hau 1989). Footpad injection (particularly when an oil adjuvant is used) and intrasplenic injection give rise to serious animal welfare concerns. The intrasplenic route can be used when only small amounts of antigen are available and a direct delivery of antigen to lymphoid tissue is needed. Footpad injection is used when high numbers of B-cells are required from a local lymph node. However, these routes are generally not necessary for routine antibody production and should be justified on a case-bycase basis. If a (subcutaneous) footpad injection is given, only one hind foot should be used, and the animals should be housed on soft bedding. The Volume of Injection In principle, injection volumes should be as small as possible to limit the level of side effects of the antigen and adjuvant, but also because the immune response is generally stronger against an antigen administered in high concentration in a small volume than a low concentration in a large volume of vehicle. An important aspect to consider is the adjuvant used. In particular, oil adjuvants in combination with microbial products and viscous gel adjuvants form a depot at the injection site and induce a local inflammation, potentially inducing sterile abscesses. Therefore, emulsion antigens should be administered in smaller volumes than aqueous antigens. Information on recommended volumes for emulsions and aqueous products is given in Tables€16.5a and 16.5b, respectively. In case of s.c. injection, multiple small volumes should be given in multiple sites, instead of one large volume. This may reduce the intensity of side effects (Halliday et al. 2000). However, the number of injection sites should be limited.
84550.indb 477
11/11/10 6:27:51 PM
478
Handbook of Laboratory Animal Science
Table€16.4╅ Injection Routes: Advantages and Disadvantages Injection Route
Details
Advantage
Most frequently used route Route preferred Do not inject material in part of animal used for restraint Limit locations to less than five Skeletal muscles are well vascularized Use B. femoris in small animals Not recommended for injection of oil adjuvant in rodents
Relatively large volumes can be administered Inflammatory processes can be easily monitored Rapid absorption, in particular, with muscular activity In large animals, relatively large volumes can be administrated
i.p.
An efficient route for antigen delivery (easy access to lymphatics, large intercellular clefts in lymphatics, transport help by respiratory activity)
Relatively large volumes of inoculums can be accommodated
i.v.
Antigen is delivered primarily to spleen and secondarily to lymph nodes Route of choice for particulate antigen Limit number of injections to less than five per animal
Rapid distribution of antigen
s.c.
i.m.
i.d.
Disadvantage
Efficient processing of antigen due to high density of Langerhans dendritic cells Small quantities of antigen already effective
Slow absorption
Antigen and adjuvant can spread along interfacial planes and nerve bundles and may damage sciatic nerve and have other serious side effects Local reactions can be easily overlooked Relatively high percentage of injection failures Oil adjuvant induces peritonitis Risk for anaphylactic shock at booster injection of aqueous antigen No oil or viscous gel adjuvant can be used High risk for anaphylactic shock at booster immunization Use of oil adjuvants leads to ulcerative processes
Table€16.5a╅ Maximum Volume of Injection Used for Injection of Emulsions per Route of Injection for Different Animal Species Volume (mL) Species
s.c.
i.m.
i.p.
i.v.
i.d.
Mice Rats Guinea pigs Rabbits Sheeps/goats Cattle Poultry
0.1 0.2 0.2 0.25 0.5 0.5 0.25
0.05 0.1 0.2 0.25 0.5 0.5 0.5
N.R. N.R. N.R. N.R. N.R. N.R. N.R.
N.A. N.A. N.A. N.A. N.A. N.A. N.A.
0.05 0.05 0.05 0.05 0.05 0.05 0.05
Source: From Hendriksen, C. F. M., and J. Hau. 2003. In Handbook of Laboratory Animal Science, vol. I: Essential Principles and Practices, ed. J. Hau and G. Van Hoosier. Boca Raton, FL: CRC Press. Notes: N.R. = not recommended; N.A. = not acceptable.
84550.indb 478
11/11/10 6:27:51 PM
Immunization for Production of Antibodies
479
Table€16.5b╅ Maximum Volume of Injection Used for Injection of Aqueous Antigens per Route of Injection for Different Animal Species Volume (mL) Species Mice Rats Guinea pigs Rabbits Sheeps/goats Cattle Poultry
s.c.
i.m.
i.p.
i.v.
i.d.
0.5 0.5–1.0 1.0 1.5 2.0 2.0 0.5
0.05 0.1 0.2 0.2–0.5 2.0 2.0 1.0
1.0 5.0 5–10 10–20 N.A. N.A. N.A.a
0.2 0.5 0.5–1.0 1–5 N.G. N.G. N.G.
0.05 0.05 0.05 0.05 0.05 0.05 0.05
Source: From Hendriksen, C. F. M., and J. Hau. 2003. In Handbook of Laboratory Animal Science, vol. I: Essential Principles and Practices, ed. J. Hau and G. Van Hoosier. Boca Raton, FL: CRC Press. Notes: N.A. = not acceptable; N.G. = not given. a Air sacs and liver can be easily damaged.
The Quantity of Antigen Guidelines on the quantity of antigen that should be used are difficult to give. The quantity depends on the inherent properties of the antigen, whether the antigen is purified or a component in a mixture of antigens, the adjuvant used, the route and frequency of injection, etc. In general terms, >25–50 µg up to milligram quantities of protein antigen in conjunction with an adjuvant are needed to ascertain a high-titer antibody response. Although smaller doses may be used for smaller animals, the antigen dose is not increased or decreased in proportion to body weight. A better way is to think in terms of the number of lymphoid follicles to which the antigen will be distributed. Recommended antigen doses in combination with FCA are 10–200 µg for mouse and 250–5000 µg for goat and sheep. For each antigen, there is a dose range called the “window of immunogenicity.” Too much or too little antigen may induce suppression, sensitization, tolerance, or other unwanted immunomodulation. Very low doses (<1–5 µg) are used to induce hypersensitivity (allergy) (Poulsen, Hau, and Kollerup 1987; Nielsen et al. 1989) and should be avoided—not least because booster injections may result in an anaphylactic shock in the animals. Based on the finding that high antigen doses activate a large number of low-affinity B-cell clones, while a low dose activates only the high-affinity B-cell clones, it is recommended that a lower dose be used at booster immunizations. Booster Immunization Generally, it is not sufficient to immunize only once, even with the help of an adjuvant. The time interval for booster immunization is critical, but fixed periods are difficult to give. As a rule of thumb, a booster can be considered after the antibody titer has plateaued or begins to decline. Usually, this is 3 weeks after the primary immunization in case of an aqueous antigen and 4 weeks when a depot-forming adjuvant is used. The number of booster immunizations should be limited to a maximum of two or three. If antibody responses are still insufficient, the animal should, in general, be taken out of the protocol. However, in case of antigens of low molecular weight, more boosters might be needed. Also, when animals are used as antibody donors over a long period of time, frequent booster immunizations might be necessary. If a water-in-oil emulsion has been administered in the initial immunization, a depot of antigen has been established in the animal, and booster immunizations may be carried out
84550.indb 479
11/11/10 6:27:51 PM
480
Handbook of Laboratory Animal Science
without the use of an adjuvant. The depot will remain in the animal for many months. Too frequent immunizations with very low amounts of antigen might lead to tolerance. Intermittent bleeding of a hyperimmunized animal or plasmaphoresis appears to help maintain a high serum antibody titer. Booster antigen mixtures should never be inoculated at the site of the previous injections and certainly not in granulomas or swellings. Furthermore, booster immunizations might advantageously be given by a different route than the primary immunization because the object of this procedure is to stimulate as many memory cells as possible. Blood Collection A number of conditions should be met when collecting blood from an animal. The technician should be experienced with the technique, and the procedure has to be performed in a quiet surrounding. If the animal is not stressed during bleeding, the use of sedative to facilitate blood sampling is usually unnecessary. However, in case of exsanguination, general anesthesia is a prerequisite and, at the end of the procedure, the animal should be subjected to euthanasia following appropriate standards. Blood should normally be collected only from the sites recommended in guidelines (e.g., by the BVA/FRAME/RSPCA/UFAW Joint Working Group on Refinement, 1993). These guidelines also specify the maximum amount of blood that can be collected. When maximum volumes are taken, the frequency should be less than once a fortnight. Table€16.6 specifies maximum amounts of blood to be collected for common immunization species. As a general rule, a maximum of approximately 8 mL/kg body weight can be removed once a fortnight (van Zutphen, Baumans, and Beynen 2001). The needle used should be matched for the vessel size and should preferably be of relatively large bore to facilitate rapid collection and to prevent hemolysis. The volume to be removed per bleeding should not exceed 15% of the total blood volume, which equals about 1% of the animal’s body weight. General Aspects During the entire experiment, immunized animals should be closely monitored for general appearance, for food and water intake, and for local reactions at the injection site. In case no information is available about the effects of the antigen or adjuvant, as well as of the immunization period and booster interval, it might make sense to perform a pilot study. Macroscopic and microscopic examination should be part of such a study (Leenaars et al. 1998). A postmortem analysis of animals dying spontaneously during an immunization period should be performed. Table€16.6â•… Maximum Volumes of Blood That Should Be Collected Animal Species Mouse Hamster Rat Guinea pig Rabbit Monkey Sheep Horse
Volume (mL) 0.3 0.3 2.0 5.0 15a 20–200a 200–600a 500–700a
Source: Data from Leenaars, P. P. A. M. et al. 1999. ATLA 27:79. a Depending on body weight.
84550.indb 480
11/11/10 6:27:51 PM
Immunization for Production of Antibodies
481
Existing Guidelines for the Production of Polyclonal Antibodies Several guidelines for the production of Pabs have been established by various national control authorities, organizations, and institutions, such as those of the Canadian Council on Animal Care (2002), the NIH (Grumstrupp-Scott and Greenhouse 1988), and the Danish (1990) and Dutch (1993/2000) Inspectorates. An overview was prepared by Leenaars and coauthors (1999). Other guidelines can be found on the Internet. The general aim of the guidelines is to ensure appropriate procedures that combine the scientific interest in producing satisfactory results and the animal welfare aspect. The guidelines can be used as a tool for scientists, animal technicians, animal welfare officers, and ethical review committees. Most of the guidelines are not mandatory, but they can be used as a reference to assure proper treatment of animals. Differences between the guidelines occur, but most agree about a number of aspects. Thus, there is general consensus that footpad injection is not necessary and the use of FCA is discouraged, while the use of alternative adjuvant products is recommended. Furthermore, most guidelines set limitations to the volumes to be injected and the blood samples taken.
New Developments in Antibody Production The use of egg-laying chickens for the production of egg yolk (IgY) antibodies can be seen as both a reduction and a refinement alternative (Hau and Hendriksen 2005). Due to the high IgY concentration in the egg yolk and the egg-laying time, large amounts of antibodies can be produced in one animal. Consequently, the number of animals can be reduced. This is a refinement alternative because animals do not have to be restrained and bled. For isolation of IgY from egg yolk, easy-to-use commercial kits are now available, and also the number of secondary antichicken conjugate antibodies has increased. This makes the IgY technology an attractive alternative to mammalian antibody production. Combined with oral immunization, this technology would eliminate all procedural stress from antibody-producing animals. However, it seems difficult to develop efficient simple oral administration regimes that induce sufficient memory in chickens to allow for effective booster effect when the chickens receive subsequent oral antigen (Mayo et al. 2009). Another approach to refine conventional Pab production is the induction of a systemic humoral response by transcutaneous immunization. For example, by using cholera toxin as an adjuvant, the antigen can be applied with a skin plaster, facilitating transcutaneous uptake. Oral immunization eliminates invasive injections and is also a novel trend that should be welcomed from an animal welfare aspect. The available in vitro techniques for Pab production as routine methodology are still limited. Cloning and production of immunoglobulins with specificity against a particular antigen by nonvertebrate and plant hosts are indeed a possibility, but only time will tell whether this technology develops into a practical adaptable technique for smaller scale and experimental antibody production.
References Abraham, L., D. O’Brien, O. M. Poulsen, and J. Hau. 1994. The effect of social environment on the production of specific immunoglobulins against an immunogen (human IgG) in mice. In Welfare and science, ed. J. Bunyan, Royal Society of Medicine Press Ltd., London. 165. Behring, E. von, and Kitasato. 1890. Uber das Zustandekommen der Diphtherie-Immunität und der TetanusImmunität bei Tieren [Production of diphtheria immunity and tetanus immunity in animals]. Deutsche Medizinische Wochenschrift 49:113.
84550.indb 481
11/11/10 6:27:51 PM
482
Handbook of Laboratory Animal Science
Bizhanov, G., I. Jonauskiene, and J. Hau. 2004. A novel method, based on lithium sulphate precipitation for purification of chicken egg yolk immunoglobulin Y, applied to immunospecific antibodies against Sendai virus. Scandinavian Journal of Laboratory Animal Science 31:121–130. Bollen, L. S. and J. Hau. 1996. Comparison of immunospecific antibody response in young and old chickens immunized with human IgG. Lab Anim 33:71. ———. 1997. Immunoglobulin G in the developing oocytes of the domestic hen and immunospecific antibody response in serum and corresponding egg yolk. In Vivo 11:395. ———. 1999. Freund’s complete adjuvant has a negative impact on egg laying frequency in immunized chickens. In Vivo 13:107. BVA/FRAME/RSPCA/UFAW Joint Working Group on Refinement. 1993. Removal of blood from laboratory animals and birds. Lab Anim 27:1–22. Cannat, A., N. Feingold, J. C. Caffin, and A. Serre. 1979. Studies on the genetic control of murine humoral response to immunization with a peptidoglycan-containing fraction extracted from Brucella melitensis. Annals of Immunology 130c:675. CCAC (Canadian Council on Animal Care). 2002. Guidelines on antibody production. Ottawa, Ontario, Canada. Chapel, H. M., and P. J. August. 1976. Report on nine cases of accidental injury to man with Freund’s complete adjuvant. Clinical and Experimental Immunology 24:558. Claassen, E., and W. J. A. Boersma. 19692. Characteristics and practical use of new generation adjuvants as an acceptable alternative for Freund’s complete adjuvant. Research in Immunology 143:475. Colpaert, F. C. 1987. Evidence that adjuvant arthritis in the rat is associated with chronic pain. Pain 28:201. De Muynck, B., C. Naavarre, and M. Boutry. 2010. Production of antibodies in plants: Status after twenty years. Plant Biotechnology Journal 8:1–35. Erb, K., and J. Hau. 1994. Monoclonal and polyclonal antibodies. In Handbook of laboratory animal science, vol. I, ed. P. Svendsen and J. Hau, 293–309. Boca Raton, FL: CRC Press. Freund, J. 1947. Some aspects of active immunization. Annals of Reviews of Microbiology 1:291. Freund, J., J. Casals, and E. P. Hosmer. 1937. Sensitization and antibody formation after injection of tubercle bacilli and paraffin oil. Proceedings of the Society for Experimental Biology and Medicine 37:509. Grumstrupp-Scott, J., and D. D. Greenhouse. 1988. NIH intramural recommendations for the research use of Freund’s complete adjuvant. ILAR News 2:9. Halliday, L. C. et al. 2000. Physiologic and behavioral assessment of rabbits immunized with Freund’s complete adjuvant. Contemporary Topics in Laboratory Animal Science 39:8. Hanly, W. C., J. E. Artwohl, and B. T. Bennett. 1995. Review of polyclonal antibody production procedures in mammals and poultry. ILAR News 37:93. Harboe, N. M. G., and A. Ingild. 1983. Immunization, isolation of immunoglobulins, and antibody titer determination. Scandinavian Journal of Immunology 17 (Suppl. 10): 245. Hau, J. 1988. The rabbit as antibody producer—Advantages and disadvantages. In Symposium über Zucht, Haltung und Ernährung des Kaninches. Das Kanin als Modell in der Biomed. Forschung. Altromin & Akademi für tierärztliche Fortbildung, p. 68. Hau, J., and C. F. M. Hendriksen. 2005. Refinement of polyclonal antibody production by combining oral immunization of chickens with harvest of antibodies from the egg yolk. ILAR Journal 46:294–299. Hau, J., J. G. Westergaard, P. Svendsen, A. Bach, and B. Teisner. 1980. Comparison between pregnancy-associated murine protein-2 (PAMP-2) and human pregnancy-specific beta 1-glycoprotein (SP-1). Journal of Reproduction and Fertility 60:115. ———. 1981. Comparison of the pregnancy-associated murine protein-1 and human pregnancy zone protein. Journal of Reproductive Immunology 3:341. Heimburg-Molinaro, J., and K. Rittenhouse-Olson. 2009. Development and characterization of antibodies to carbohydrate antigens. In Methods in molecular biology, glycomics: Methods and protocols, vol. 534, ed. N. H. Packer and G. Karlsson. Totowa, NJ: Humana Press. Hendriksen, C. F. M., and J. Hau. 2003. Production of monoclonal and polyclonal antibodies. In Handbook of laboratory animal science, vol. I: Essential principles and practices, ed. J. Hau and G. Van Hoosier. Boca Raton, FL: CRC Press. Herbert, W. J. 1967. Methods for preparation of water-in-oil, and multiple, emulsions for use as antigen adjuvants; and notes on their use in immunization procedures. In Handbook of experimental immunology, ed. D. M. Weir. Oxford, England: Blackwell Scientific.
84550.indb 482
11/11/10 6:27:51 PM
Immunization for Production of Antibodies
483
Hohmann, A. W. 1998. Freund’s adjuvant and immune responses. The use of immuno-adjuvants in animals in Australia and New-Zealand. Glen Osmond, SA: Australia ANZCCART. Ishida, I., K. Tomizuka, H. Yoshida, T. Tahara, N. Takahashi, A. Ohguma, S. Tanaka, et al. 2002. Production of human monoclonal and polyclonal antibodies in TransChromo animals. Cloning Stem Cells 4:91–102. Jennings, V. M. 1995. Review of selected adjuvants used in antibody production. ILAR Journal 37:119. Jensenius, J. C., I. Andersen, J. Hau, M. Crone, and C. Koch. 1981. Eggs: Conveniently packaged antibodies. Methods for purification of yolk IgG. Journal of Immunology Methods 46:63. Kaplan, P. J., T. J. Caperna, and J. S. Garvey. 1981. Bovine serum albumin humoral immune response in aged Fischer 344 rats. Mechanisms of Ageing and Development 16:61. Kittell, C. L., R. E. Banks, and C. L. Hadick. 1991. Raised skin lesions in rabbits after immunization. Lab Anim 25:16. Köhler, G., and C. Milstein. 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495. Leenaars, M. 1997. Adjuvants in laboratory animals. Thesis, Erasmus University Rotterdam, Ponsen & Looijen BV, Wageningen, the Netherlands. Leenaars, M., E. Claassen, and W. J. A. Boersma. 1996. Modulation of the humoral immune response; antigens and antigen presentation. In Immunology methods manual, ed. I. Lefkovits, 989. New York: Academic Press. Leenaars, P. P. A. M., and Hendriksen, C. F. M. 1999. Ontwikkeling Van “Comparative Adjuvant Selection System,” RIVM report 623860006. Leenaars, M., and C. F. M. Hendriksen. 2005. Critical steps in the production of polyclonal and monoclonal antibodies: Evaluation and recommendations. ILAR Journal 46:269–279. Leenaars, P. P. A. M., C. F. M. Hendriksen, W. A. De Leeuw, F. Carat, P. Delahaut, R. Fischer, M. Halder, et al. 1999. The production of polyclonal antibodies in laboratory animals. ATLA 27:79. Leenaars, P. P. A. M., M. A. Koedam, P. W. Wester, V. Baumans, E. Claassen, and C. F. M. Hendriksen. 1998. Assessment of side effects induced by injection of different adjuvant/antigen combinations in rabbits and mice. Lab Anim 32:387. Lifschitz, R., M. Schwartz, and E. Mozes. 1980. Specificity of genes controlling immune responsiveness to (T,G)-A-L and (Phe, G)-A-L. Immunology 41:339. Lindblad, E. B., and J. Hau. 2000. Escaping from the use of Freund’s complete adjuvant. In Progress in the reduction, refinement and replacement of animal experimentation, Proceedings from the 3rd World Congress on Alternatives and Animal Use in the Life Sciences 1999, ed. M. Balls, A.-M. van Zeller, and M. E. Halder, 1681–1685. Amsterdam: Elsevier. Lindblad, E. B., and J. V. Sparck. 1987. Basic concepts in the application of immunological adjuvants. Scandinavian Journal of Laboratory Animal Science 14:1. Long, D. A. 1957. The influence of the thyroid gland upon immune responses of different species to bacterial infection. CIBA Found, Colloq. Endocrinology 10:287. Mayo. S., H. E. Carlsson, A. Zagon, F. Royo, and J. Hau. 2009. Enhancement of anamnestic immunospecific antibody response in orally immunized chickens. Journal of Immunology Methods 342:58–63. Mayo, S. L., F. Royo, and J. Hau. 2005. Correlation between adjuvanticity and immunogenicity of cholera toxin B subunit in orally immunized young chickens. APMIS 113:284-287. Mayo, S. L., M. Tufvesson, H. E. Carlsson, F. Royo, S. Gizurarson, and J. Hau. 2005. RhinoVax is an efficient adjuvant in oral immunization of young chickens and cholera toxin B as an effective oral primer in traditional subcutaneous immunization with Freund’s incomplete adjuvant. In Vivo 19:375–382. McCullough, K. C., C. F. M. Hendriksen, and T. Seeback. 1997. In vitro methods in vaccinology. In Veterinary vaccinology, ed. P.-P. Pastoret, J. Blancou, P. Vannier, and C. Verschueren, 69. New York: Elsevier. Newcombe, C., and A. R. Newcombe. 2007. Antibody production: Polyclonal-derived biotherapeutics. Journal of Chromatography B 848:2–7. Nielsen, B. R., O. M. Poulsen, and J. Hau. 1989. Reagin production in mice: Effect of subcutaneous and oral sensitization with untreated bovine milk and homogenized bovine milk. In Vivo 3:271. Persdotter Hedlund, G., and J. Hau. 2001. Oral immunization of chickens using cholera toxin B subunit and Softigen as adjuvants results in high antibody titre in the egg yolk. In Vivo 15:384. Poulsen, O. M., and J. Hau. 1990. Suppression of humoral antibody response during pregnancy in mice. In Vivo 4:381. Poulsen, O. M., J. Hau, and J. Kollerup. 1987. Effect of homogenization and pasteurization on the allergenicity of bovine milk analyzed by a murine anaphylactic shock model. Clinical Allergy 17 (5): 449–458.
84550.indb 483
11/11/10 6:27:52 PM
484
Handbook of Laboratory Animal Science
Poulsen, O. M., B. R. Nielsen, A. Basse, and J. Hau. 1990. Comparison of intestinal anaphylactic reactions in sensitized mice challenged with untreated bovine milk and homogenized bovine milk, Allergy 45:321. Ramon, G. 1925. Sur l’augmentation anormale de l’antitoxine chez les chevaux producteurs de serum antidiphterique. Bull. Soc. Centr. Med. Vet., 101:227. Russell, W. M. S., and R. L. Burch. 1959. The principles of humane experimental technique. London: Methuen. Schade, R., I. Behn, M. Erhard, A. Hlinak, and C. Staak, eds. 2001. Chicken egg yolk antibodies, production and application. IgY-technology. Berlin: Springer. Schade, R., C. Staak, C. Hendriksen, M. Erhard, H. Hugl, G. Koch, A. Larsson, et al. 1996. The production of avian (egg yolk) antibodies: IgY. ECVAM Workshop Report 21, ATLA 24:925. Stewart-Tull, D. E. S. 1995. The theory and practical application of adjuvants. New York: John Wiley & Sons. Svendsen, L., A. Crowley, L. H. Ostergaard, G. Stodulski, and J. Hau. 1995. Development and comparison of purification strategies for chicken antibodies from egg yolk. Laboratory Animal Science 45:89. Svendsen Bollen, L., A. Crowley, G. Stodulski, and J. Hau. 1996. Antibody production in rabbits and chickens immunized with human IgG. A comparison of titer and avidity development in rabbit serum, chicken serum and egg yolk using three different adjuvants. Journal of Immunology Methods 27:191. Talmage, J., and F. J. Dixon. 1953. The influence of adjuvants on the elimination of soluble protein antigens and the associated antibody responses. Journal of Infectious Disease 93:176. Van Zutphen, L. F. M., V. Baumans, and A. C. Beynen, eds. 2001. Principles of laboratory animal science, 322. Amsterdam: Elsevier. Young, C. R., and M. Z. Atassi. 1982. Genetic control of antibody response to antigenic sites by increasing the dose of antigen used in immunization. Journal of Immunogenetics 9:343.
84550.indb 484
11/11/10 6:27:52 PM
Chapter 17
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
Ludo J. Hellebrekers and Patricia Hedenqvist Contents Introduction..................................................................................................................................... 486 Administration of Analgesics and Anesthetics............................................................................... 487 Topical Administration.............................................................................................................. 487 Administration by Injection....................................................................................................... 488 Administration by Inhalation..................................................................................................... 489 Analgesic and Anesthetic Drugs..................................................................................................... 490 Opioids....................................................................................................................................... 490 NSAIDs...................................................................................................................................... 491 Local Anesthetics....................................................................................................................... 491 Anticholinergics......................................................................................................................... 492 Sedatives and Tranquilizers....................................................................................................... 493 Volatiles and Gases.................................................................................................................... 494 Hypnotics................................................................................................................................... 496 Alpha-2 Agonists........................................................................................................................ 498 Dissociative Anesthetics............................................................................................................ 499 Muscle Relaxants....................................................................................................................... 499 Anesthetic Management.................................................................................................................500 Preanesthetic Considerations.....................................................................................................500 Monitoring during Anesthesia...................................................................................................500 Postanesthetic Care.................................................................................................................... 503 Pain Management........................................................................................................................... 503 Acute Pain.................................................................................................................................. 503 General Preventative and Therapeutic Strategies in Acute Pain...............................................504 Chronic Pain.............................................................................................................................. 505 Analgesic Modalities for Treating Acute or Chronic Pain........................................................ 505 Operator Safety...............................................................................................................................506 Anesthetic Gases........................................................................................................................506 Gas Cylinders.............................................................................................................................506 Controlled Drugs........................................................................................................................506 485
84550.indb 485
11/11/10 6:27:52 PM
486
Handbook of Laboratory Animal Science
Analgesia and Anesthesia of Different Species..............................................................................507 Analgesia and Anesthesia in Rodents........................................................................................507 Analgesia...............................................................................................................................507 Local Anesthetics.................................................................................................................. 508 Antidepressants..................................................................................................................... 508 Anesthesia............................................................................................................................. 508 Analgesia and Anesthesia in Rabbits......................................................................................... 513 Analgesia............................................................................................................................... 513 Anesthesia............................................................................................................................. 514 Analgesia and Anesthesia in Pigs.............................................................................................. 516 Analgesia............................................................................................................................... 516 Anesthesia............................................................................................................................. 516 Analgesia and Anesthesia in Small Ruminants......................................................................... 519 Analgesia............................................................................................................................... 519 Anesthesia............................................................................................................................. 519 Analgesia and Anesthesia in Cats, Dogs, and Ferrets............................................................... 520 Analgesia............................................................................................................................... 520 Anesthesia............................................................................................................................. 520 Analgesia and Anesthesia in Primates....................................................................................... 522 Analgesia............................................................................................................................... 522 Anesthesia............................................................................................................................. 522 Analgesia and Anesthesia—Birds, Fish, Reptiles, and Amphibia............................................. 523 Analgesia............................................................................................................................... 523 Anesthesia............................................................................................................................. 524 Euthanasia....................................................................................................................................... 526 Goals and Definition of Terms in Euthanasia............................................................................ 526 Methods of Euthanasia............................................................................................................... 526 Pharmacological–Chemical Methods................................................................................... 526 Physical Methods.................................................................................................................. 527 Special Considerations with Regard to Scientific Purpose................................................... 527 Euthanasia of Different Species................................................................................................. 528 Euthanasia of Rodents and Rabbits....................................................................................... 528 Euthanasia of Small Ruminants and Pigs............................................................................. 528 Euthanasia of Cats, Dogs, and Ferrets.................................................................................. 529 Euthanasia of Primates.......................................................................................................... 529 Euthanasia of Birds, Fish, Reptiles, and Amphibians........................................................... 529 References....................................................................................................................................... 530 Introduction Legislation and humane use of live vertebrate animals in scientific research require that experiments must be performed in a way that minimizes pain and suffering. As a rule, painful experiments should be performed under local or general anesthesia whenever possible. One of the major problems in laboratory animal anesthesia and pain control lies in the fact that, under many circumstances, it is difficult to ascertain adequacy of anesthesia or analgesia. The major obstacles one faces are the recognition and quantification of pain, less so in the animal under anesthesia than in the wake animal potentially experiencing pain. A plethora of parameters of physiological or behavioral background has been named for their relevance for recognizing and quantifying pain in animals. Although many of these parameters
84550.indb 486
11/11/10 6:27:52 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
487
will change under painful circumstances, their discriminative power is limited to the extent that they cannot definitively differentiate between pain and other causes involving a stress response. Complicating matters further, some species are known to mask signs of pain, based upon their potential role as victims or bait of predatory animals, making detection of pain even more difficult. Under many circumstances involving laboratory animals, the principle of analogy is employed. This approach ascribes pain to an animal in the case where it may be expected, based on human experience, in all cases where pain cannot definitely be excluded based on findings from the animal involved. Despite the risk of anthropomorphism, this approach provides the “benefit of the doubt” to the animal, which will consequently receive analgesic treatment when indicated. Our present-day understanding of the physiology of pain clearly demonstrates the relevance of a proactive (i.e., preventative) approach to pain. Through early detection and installation of effective treatment, prevention of the development of peripheral and central sensitization is pursued. With this approach, the development of pain (intensity and duration) will be reduced and thus easier to treat. The distress to the animal will be reduced and a full and fast recovery promoted. In discussing laboratory animal anesthesia, a differentiation can be made between those animals that are anesthetized for reasons of instrumentation and those that are anesthetized for allowing the execution of an experiment. Under the latter circumstances, scientific data collection is performed during anesthesia. This automatically necessitates an in-depth evaluation of potential interactions between the data to be collected and the anesthetic protocol. Of major importance under these circumstances is the precise definition of the actual parameters under investigation. Thereby, the evaluation of the potential, direct, or indirect effects of the anesthetics used can be achieved based upon available physiologic and pharmacological knowledge. The scope of this chapter is to give an overview of the current views on laboratory animal analgesia and anesthesia. For more detailed information, the reader is advised to consult the different laboratory animal or veterinary anesthesia handbooks. Administration of Analgesics and Anesthetics Refinement of techniques for administration of substances has been reported by Morton and co-workers (2001). Topical Administration Local anesthetics can be applied as gel, cream, solution, or aerosol and are used for blocking pain in the eye and ear canal, on mucous membranes, and on the skin. Local anesthetics can be used for analgesic supplementation during general anesthesia in the form of regional or epidural block or be deposited directly around exposed nerves during surgery. Lidocaine gel or spray is used for mucous membranes of the mouth, nose, pharynx, larynx, vagina, urethra, and rectum for analgesia during surgery. A 2–4% solution produces an effect in 5 min and lasts for 30 min (Skarda 1996). Topical local anesthetic solution can be used during nasal, aural, or extraocular surgery. EMLA® cream, a combination of lidocaine and prilocaine, is effective in reducing pain from venipuncture or skin biopsy in larger species (cat, dog, pig, and rabbit; Nolan 2000). After the fur or coat is clipped, the cream is applied for 60 min under an adhesive plastic bandage, resulting in 1 h of full skin anesthesia (Figure€17.1). Patches for transdermal administration of the opioid fentanyl (25, 50 µg/h) can be used in larger species (cat, dog, pig, sheep, and goat) for control of postoperative or chronic pain (Nolan 2000). Even though this method involves topical administration, the aim is to achieve a systemic analgesic effect. Because fentanyl plasma concentrations vary considerably with patches, they are best used
84550.indb 487
11/11/10 6:27:53 PM
488
Handbook of Laboratory Animal Science
Figure 17.1â•…EMLA cream on rabbit ears for cannulation of central ear artery.
to deliver a background level of analgesia in combination with a rigid evaluation of effect. Skin temperature and perfusion are among the factors that will influence fentanyl absorption rate. General anesthetics and analgesics may sometimes be applied on the mucous membranes of the mouth or rectum in order to avoid painful injections or immediate hepatic drug metabolism, which occurs after oral and intraperitoneal administration. Examples are rectal administration of ketamine or buccal administration of buprenorphine in cats (Hanna, Borchard, and Schmidt 1988; Robertson et al. 2001). Before limb amputation, exposed nerves are bathed in local anesthetic for 2–3 min prior to transaction (Dobromylskyj et al. 2000). This will significantly decrease postoperative pain and possibly reduce the development of phantom limb pain. Administration by Injection Intravenous injection of anesthetics is the safest injection route because it allows for dose adjustment in the individual animal. Repeated injections (boluses) may be used to prolong duration of anesthesia. However, to achieve a stable plane of anesthesia and, consequently, more stable hemodynamics, continuous intravenous infusion is preferred. A precise control over anesthetic depth can be achieved by using continuous infusion of drugs that are rapidly metabolized (e.g., propofol). By changing the infusion rate, the level of anesthesia is rapidly changed, which gives a control comparable to that of inhalation anesthesia. Another benefit of continuous intravenous access is that fluids, analgesic supplementation, specific antagonists, or emergency drugs can be administered directly and with an immediate effect. Peripheral veins for IV injections may be difficult to access in several rodent species, like the hamster and guinea pig. Injection of anesthetics by the intraperitoneal, intramuscular, or subcutaneous route does not allow for the same flexibility in dose adjustment as intravenous administration. Therefore, only substances with a broad safety margin should be applied via these routes. Intraperitoneal administration is commonly used in rodent anesthesia. It should not be used in pregnant rodents or animals larger than rodents. Rapid absorption due to the large uptake area is considered an advantage, but care must be taken not to inject into abdominal organs. A short needle stop, which can be made from the protective needle sheath, prevents the needle from entering the peritoneal cavity to a depth of more than 4 or 5 mm (Claassen 1994).
84550.indb 488
11/11/10 6:27:54 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
489
Drugs absorbed from the peritoneal cavity by the portal system are subjected to hepatic firstpass elimination. This is in contrast to drugs administered by the intramuscular or the subcutaneous route, which reach the systemic circulation without initial hepatic metabolism. Textbooks on veterinary anesthesia often recommend the same drug dose for IM, SC, and IP injection in rodents and thus do not take into consideration the possible effects of first-pass hepatic metabolism. When routes of administration are changed from intraperitoneal to intramuscular or subcutaneous injection, drug doses should be adjusted accordingly. Intramuscular injections should be avoided in rodents because they can cause unnecessary muscle damage and pain (Wixson and Smiler 1997). Irritant substances such as ketamine may cause muscle necrosis and should be diluted and injected at several sites to limit this effect. When IM injection is indicated, the cranial thigh muscle mass (quadriceps) is most suitable. The epaxial muscles along the spine may also be used in larger animals. Injections into the caudal thigh muscle mass incorporate the risk of damaging the ischiadic nerve. Subcutaneous injection is often an alternative because it is easy to perform and less painful than IM injection. The rate of absorption is often comparable with that after IM injection. Irritant substances will cause damage when injected SC, which can be avoided by dilution with saline or water for injection (Morton et al. 2001). Administration by Inhalation Inhalation anesthesia is useful for animals of most species and ages. Advantages are fast induction and recovery, greater control than with most injectable anesthetics, and generally more stable physiological effects. The concurrent oxygen supplementation contributes to limiting the overall anesthetic risk. Due to minimal metabolism of most modern inhalants, they are less likely to be a significant variable in the research project (Brunson 1997). Disadvantages are high costs and the need for scavenging systems as well as for greater technical competence. The safest way to administer inhalants is by using an anesthetic machine. With help of a calibrated vaporizer, the volatile agent can be delivered at a precise and accurate concentration. For brief procedures in small animals, a simple bell jar was used in the past (see rodent anesthesia). This technique necessitates great caution because dangerously high vapor concentrations may be produced with modern volatile agents. With present-day systems, the volatile anesthetic is delivered to the animal by means of a carrier gas. If oxygen alone is used as carrier, the result will be increased arterial blood O2 levels. Oxygen mixed with air or nitrous oxide (N2O) will produce more physiological arterial O2 levels. Nitrous oxide has the advantage of contributing to the analgesic effect without causing any respiratory depression and of reducing the concentration of the volatile agent needed by approximately 20%. This limits the cardiovascular depression of the volatile agent. If the animal is allowed to breathe spontaneously, arterial blood carbon dioxide partial pressure may increase due to the respiratory depressant effect of the inhalant. By mechanical ventilation, more physiological CO2 levels can be maintained during anesthesia. Induction of inhalation anesthesia can be achieved by means of an induction chamber or a face mask. Chamber induction is useful for rodents, kittens, pups, piglets, birds, and other small animals not accustomed to manual restraint. Mask induction should only be used in animals that are accustomed to being handled. Anesthesia is maintained by keeping the animal on the mask or by gas delivery via an endotracheal tube. Effective scavenging systems are necessary in order to prevent exposure of staff to waste anesthetic gas (see “Operator Safety”). Induction of anesthesia with inhalants should preferably not be performed in nonsedated rabbits because they are reluctant to inhale vapors and respond with extended breath holding and violent struggling (Flecknell, Roughan, and Hedenqvist 1999; Hedenqvist et al. 2001). Instead, a shortacting injectable agent can be used for induction, followed by intubation and inhalation anesthesia.
84550.indb 489
11/11/10 6:27:54 PM
490
Handbook of Laboratory Animal Science
Analgesic and Anesthetic Drugs Opioids Opioids are used for treating moderate to severe pain; morphine is the most well-known potent opioid drug for treatment of severe pain. Synthetic opioids act in a similar manner to morphine and can likewise induce some degree of sedation in high doses. The analgesic effect is mainly mediated via µ-receptors, located in the central and peripheral nervous system and on inflammatory cells. Analgesia can also be mediated by kappa receptors, which are abundant in the spinal cord. Opioids are mostly used for their central effects but have been shown to have additional peripheral analgesic effects (e.g., in mice) after intraplantar injection or topical application on the tail (Kolesnikov, Chereshnev, and Pasternak 2000; Baamonde et al. 2000). Opioids reduce the amount of anesthetic needed for surgery and have been shown to have a preemptive analgesic effect in dogs and rodents (Lascelles et al. 1995, 1997). Side effects seen with higher dosages may include respiratory depression as a result of a decreased responsiveness of the respiratory center to carbon dioxide. This effect seems to be less significant in animals compared with humans. Older opioids such as morphine and pethidine may induce vasodilation and inhibition of baroreceptor reflexes. When the most potent agents are used IV, atropine can be given to prevent or treat severe bradycardia (Nolan 2000). All opioids decrease hydrochloric acid secretion in the stomach and prolong gastric emptying time. They also decrease propulsive activity in the gastrointestinal tract and may induce urinary retention. This does not seem to be clinically significant in rodents and rabbits. Opioid drugs are generally well absorbed from the gut and after IP, SC, and IM injection. The bioavailability after oral dosing is poor due to substantial first-pass hepatic metabolism, which also reduces bioavailability after IP administration. Repeated use over a prolonged period may induce tolerance, necessitating increased dosages to maintain efficacy. All opioid drugs can be reversed with naloxone, which has antagonistic actions on all opioid receptors. Its duration of action is short (30–60 min), and it may be used to reverse the opioid effects after overdosing or to speed up recovery. Because no analgesia remains after naloxone reversal, analgesia must be provided by other means (e.g., nonsteroidal anti-inflammatory drugs [NSAIDs]) if pain is expected. Morphine has a short elimination half-life (1–2 h). Due to longer persistence in the cerebrospinal fluid, dogs need readministration after approximately 4 h. In cats, metabolism of morphine is slow, and duration of action is approximately 6 h. Morphine has no ceiling effect, meaning that the higher the dose is, the greater the analgesic is, as well as the unwanted effect. When it is applied IV, slow injection is mandatory to avoid histamine release and consequent hypotension. Morphine may cause vomiting in dogs and cats, but not when it is given after surgery or to a traumatized cat or dog (Nolan 2000). Morphine has been shown to reduce natural killer cells and phagocytic activity (Heavner 1997). Fentanyl, alfentanil, and sufentanil are µ-agonists suitable for providing intraoperative analgesia because of their high potency and short duration of action. Preadministration with atropine or glycopyrrolate is recommended depending on the species and preexisting vagal tone. The onset of action is quick (2–5 min) and duration between 5 and 10 min (alfentanil) and 20 and 30 min (fentanyl and sufentanil) (Nolan 2000). Depending on the dose administered, respiratory support may be needed. One advantage of opioids is that they do not reduce myocardial contractility or coronary blood flow; this makes them useful for cardiac surgery and for taking physiologic measurements in cardiovascular protocols (Swindle 1998). During general surgery, they need to be combined with other anesthetics, such as inhalants, to provide full anesthesia. Fentanyl is available in the United States in combination with droperidol (Innovar®) for use in dogs, cats, rabbits, and rodents for 30 min of potent analgesia. Duration of sedation is considerably longer. Another fentanyl combination is with fluanisone in Hypnorm®, which is used for sedation
84550.indb 490
11/11/10 6:27:54 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
491
and analgesia in rodents and rabbits. Hypnorm® in combination with a benzodiazepine, such as midazolam, is used to attain 20–40 min of surgical anesthesia (Nolan 2000). Fentanyl or sufentanil can be combined with medetomidine for the induction of a completely reversible anesthesia in rats (Hu, Flecknell, and Liles 1992; Hedenqvist et al. 2000b). Due to the severe hypoxia produced by these combinations, oxygen supplementation is strongly recommended. Anesthesia with infusion of sufentanil in combination with inhalation of nitrous oxide anesthesia has been described in dogs (Van Ham et al. 1996). Buprenorphine and butorphanol are mixed agonists primarily used for providing postoperative pain relief. Although buprenorphine is 25 times as potent as morphine, the maximum analgesic effect is less (Heavner 1997). Peak plasma levels are reached after 30–60 min and duration is up to 8 h in most species. Sheep need readministration more often (approximately every 3 h). Buprenorphine can be used to reverse pure µ-agonists such as morphine or fentanyl, while providing long-lasting analgesia. Side effects are minimal when compared to those of morphine. Buprenorphine can safely be given before induction with inhalation anesthesia. Butorphanol has effects on µ- and κ-receptors and is useful to control mild to moderate pain (Dobromylskyj et al. 2000). Its duration of action is very short (1–2 h in dogs and cats). Like buprenorphine, it may be used to reverse the effects of µ-agonists. It is four times as potent as morphine, and peak plasma levels are reached after 45–60 min. NSAIDs Nonsteroidal anti-inflammatory drugs are a group of weak organic acids with anti-inflammatory, analgesic, antipyretic, and antihyperalgesic properties (Nolan 2000). They act by inhibiting enzymes (cyclo-oxygenases) and are well absorbed after oral, SC, and IM injection. Metabolism occurs in the liver. The most common side effects of NSAIDs are gastrointestinal ulceration and renal function disturbances, but they are less frequently seen with the more modern drugs. In therapeutic doses, most NSAIDs do not impair clotting time or prolong bleeding time. With the possible exception of carprofen, NSAIDs should be avoided in animals with impaired renal blood flow (e.g., during hypovolemia). Due to their embryotoxicity, NSAIDs are contraindicated in pregnant animals. Salicylates (e.g., aspirin) stimulate the respiratory center directly (medullary effect) and indirectly (production of CO2) (Heavner 1997). They interfere with platelet aggregation, thereby prolonging bleeding time. Aspirin is given by mouth and may be used to treat mild pain. Metabolism includes conjugation with glucoronic acid. In cats, which lack the enzyme for conjugation, half-life is considerably prolonged. Ketorolac (acetic acid derivate) is pharmacologically similar to aspirin, except that it does not interfere with platelets (Nolan 2000). It is readily absorbed following oral or intramuscular administration. Ketoprofen (proprionic acid derivate) has stronger analgesic properties than aspirin and can be used in dogs, cats, swine, rabbits, and rodents (Heavner 1997). It is available in tablet and injection preparations and may be used alone for postoperative analgesia or in combination with an opioid. Carprofen is the most commonly used NSAID for perioperative analgesia and is often effective for controlling postoperative pain (Nolan 2000). It may be combined with an opioid to achieve the most efficient pain relief. Half-life varies considerably in different species (dog, 8 h; sheep, 24 h). It is suitable for administration once or twice daily by mouth or by SC injection and is safe to administer preoperatively in a number of species (e.g., rat, dog). Local Anesthetics Local anesthetics are weak bases that, in the ion form, stop nerve transmission by blocking Na+ channels (Skarda 1996). Duration of action is dependent on the drug’s lipid solubility. Bupivacaine,
84550.indb 491
11/11/10 6:27:55 PM
492
Handbook of Laboratory Animal Science
for example, is more lipid than lidocaine and therefore of longer duration. Toxicity is primarily systemic and related to absorption from the injection site to the circulation. By adding a vasoconstrictor like adrenaline, the risk of toxicity is reduced, and intensity as well as duration of analgesia is extended. Adrenaline must not be included if local anesthetics are used on extremities (e.g., toe, ear, or penis) because necrosis and loss of the extremity may result from vasoconstriction-induced hypoperfusion. Not only sensory nerves, but also sympathetic and motor nerves will be blocked by local anesthetics, with a resulting muscle relaxation. When local anesthetics are used during surgical procedures, the level of general anesthesia can be kept at a lighter plane, thus reducing the extent of the side effects. The combination of a sedative and local anesthesia can potentially replace general anesthesia for minor procedures. With the use of local anesthetics, analgesia can be provided during and after surgery and thus prevent severe physiological effects of surgery. The need for other analgesics will also be reduced. Local anesthetics can be used topically, by local infiltration, in the area of a nerve, intrapleurally, and epidurally (Nolan 2000). Intravenous injection of high doses may result in convulsions, hypotension, ventricular arrhythmia, and myocardial depression. Gels may be used for application in the urethra with catheterization and aerosols for spraying the airways before intubation or bronchoscopy. Intrapleural administration is used after thorax surgery or in animals with rib fractures, and subcutaneous infiltration before skin suturing. The use of local anesthesia is essential before limb amputation to avoid development of phantom pain. The most commonly used drugs are lidocaine and bupivacaine. The duration of action for lidocaine is 60–90 min. If adrenaline is included in a 1:200,000 concentration, duration will be increased by 50%. Due to a high rate of protein binding, which prevents absorption, bupivacaine has a longer duration of action (2–6 h). The long duration makes bupivacaine suitable for control of postoperative pain. Addition of adrenaline has little effect on duration. Ropivacaine has similar properties to bupivacaine but is less cardiotoxic. Metabolism of all three drugs occurs in the liver with cytochrome P450. Safe maximum doses in most species are 4 mg/kg for lidocaine and 1–2 mg/kg for bupivacaine (Dobromylskyj et al. 2000). The LD50 for bupivacaine in adult rats is 30 mg/kg (Van Ham et al. 1996). For epidural analgesia or anesthesia, local anesthetics as well as opioids, ketamine, or xylazine may be used. EMLA cream is a mixture of prilocaine and lidocaine that will induce full skin anesthesia after 60 min of application (Nolan 2000). Its main use is in the prevention of pain with venipuncture in cats, dogs, pigs, and rabbits. MS-222 (tricaine) is a soluble local anesthetic used for sedation, immobility, and anesthesia in fish and amphibians (Bowser 2001). The drug is delivered via the water or respiratory system and has local as well as systemic effects. Onset and recovery are rapid. Anticholinergics Anticholinergics are parasympathetic antagonists that reduce salivation, limit or correct a decrease in heart rate, and dilate bronchi (Lukasik 1999). Their use is contraindicated in animals with an elevated heart rate, and they should be used with caution in species that normally have a very high heart rate (e.g., birds). Anticholinergics are used to block the effects of an increased vagal tone, which slows the heart rate during induction of anesthesia and intubation. Routine incorporation in every anesthetic protocol is not warranted; instead, they are recommended in combination with drugs that cause vagal bradycardia (opioids) or increased salivation (ketamine). An overdose of anticholinergics will precipitate seizures. Atropine has a faster onset and a shorter duration than glycopyrrolate (Brunson 1997).. In the dog, the duration of action is 30–60 min for atropine and between 2 or 3 h (reduction of heart rate) and 7 h (reduction of secretions) for glycopyrrolate. Glycopyrrolate is a large polar quaternary ammonium
84550.indb 492
11/11/10 6:27:55 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
493
molecule that will not diffuse across the blood–brain barrier or the placenta and thus has minimal effects on the CNS and fetus. Administration routes are SC, IM, or IV. Several animal species (rats, rabbits, cats) can destroy large amounts of atropine with atropine esterase in the liver. Sedatives and Tranquilizers These agents are administered to relieve anxiety, decrease stress, and ensure animal and staff safety during restraint. They are often given before induction of anesthesia to allow a smoother induction and recovery and to reduce the need for anesthetics, with the result that side effects are minimized. Upon administration, the animal should be left in a quiet area for 10–30 min to allow the treatment to take effect. The group includes benzodiazepines, phenothiazine, and butyrophenone derivates, and they are often used in conjunction with a dissociative or opioid drug to provide general anesthesia. Benzodiazepines (diazepam, midazolam, and zolazepam) are weak bases with sedative, muscle relaxant, and anticonvulsant properties (Brunson 1997). They exert their effect via potentiation of GABA (gamma-aminobutyric acid) through binding on specific GABA-receptor sites. The result is a postsynaptic inhibition with modulation of the release of norepinephrine, dopamine, and serotonin. Cardiovascular as well as respiratory systems are minimally affected and no analgesia is provided. Sometimes when a benzodiazepine is administered as the first or only drug, the animal reacts by getting more difficult to handle. By adding an opioid, sedation is increased and handling facilitated. The effects of benzodiazepines can be reversed with the specific antagonist flumazenil. The most common formulation of diazepam is in a propylene glycol base, making it insoluble in water. Uptake after SC or IM administration is unpredictable, so IV injection is preferable. The halflife of diazepam varies between 1 and 7 h in the rat, guinea pig, rabbit, and dog. Midazolam is slightly more potent than diazepam and is water soluble, which makes it suitable for intravenous, subcutaneous, and intramuscular injection. It is compatible with atropine, fentanyl, Hypnorm, and ketamine. Combinations of zolazepam and the dissociative agent tiletamine (Zoletil® or Telazol®) are useful for attaining surgical anesthesia in rats, ferrets, gerbils, dogs, and cats. They have been reported to cause nephrotoxicity in rabbits and tachycardia in dogs. Due to respiratory depression, oxygen supplementation and assisted ventilation may be necessary in dogs and cats (Lukasik 1999). Acepromazine is a phenothiazine derivate acting as a sedative by central dopamine blockade, and it has peripheral alpha-adrenergic antagonistic effects (Brunson 1997). At lower doses, it causes sedation and drowsiness and, at higher doses, ataxia and somnolence. It has no analgesic effects. Acepromazine reduces the occurrence of [nor]adrenaline-induced ventricular arrhythmias. Side effects include vasodilation, hypotension, hypothermia, and a lowered CNS seizure threshold. It should not be used in animals with hypotension or in pediatric animals. It can be administered IM or SC. Butyrophenone derivates include droperidol, fluanisone, and azaperone, with similar properties to the phenothiazine derivates (Brunson 1997). Droperidol is combined with the opioid fentanyl in Innovar-Vet® and fluanisone with fentanyl in Hypnorm. In dogs, Innovar-Vet causes profound analgesia for about 30 min and is useful for minor procedures, but undesirable CNS stimulation may occur in cats. In rats, muscle relaxation is poor. Adverse effects may include bradycardia, hypotension, respiratory depression, hypoxia, hypercapnia, and acidosis. Anticholinergic premedication eliminates some of these side effects. Hypnorm is commonly used in rodents and rabbits for its sedative and analgesic effects and may be administered SC, IM, or IV. It is often used in combination with a benzodiazepine such as midazolam for surgical anesthesia. During Hypnorm + midazolam anesthesia in rats, cardiac output is almost double that of pentobarbital anesthetized rats (Skolleborg et al. 1990). Azaperone has minimal cardiovascular effects and produces immobilization in pigs lasting about 20–40 min (Swindle 1998). It is not recommended for use in other species.
84550.indb 493
11/11/10 6:27:55 PM
494
Handbook of Laboratory Animal Science
Figure 17.2â•…Isoflurane vaporizer.
Volatiles and Gases Inhalants are halogenated hydrocarbons (halothane), halogenated ethers (iso-, des-, en-, sevoflurane), or inorganic gases (N2O). Volatile anesthetics (halo-, iso-, des-, en-, sevoflurane) enhance the inhibitory neurotransmitter GABA (gamma-aminobutyric acid) by interacting with CNS GABAA receptors. Each volatile has a unique range of temperature-dependent vapor pressures, usually giving rise to concentrations much higher than those needed for anesthesia. Thus, a calibrated vaporizer is necessary for controlled delivery (Figure€17.2). Minimum alveolar concentration (MAC), a measure of inhalation anesthetic potency, is the concentration that produces immobility in 50% of subjects during standard noxious stimulation and corresponds to a light level of anesthesia. MAC is similar between animals of one species and only varies 10–20% between animal species (Ludders 1999). To reach an adequate depth of anesthesia reliably, with 95% of the subjects unresponsive to a noxious stimulus, 1.5 MAC is needed when only a volatile is administered to the subject in question. A deep level of anesthesia is represented by 2.0 MAC and may constitute an overdose (Steffey 1996). MAC decreases with age, and the highest values are found in neonates. Premedication with opioids, alpha-2 agonists, or tranquilizers will reduce the required MAC. The high MAC value for N2O in animals indicates that anesthesia cannot be maintained with N2O alone because more than 100% is needed to suppress reaction to a noxious stimulus. Nitrous oxide is more effective in humans (MAC 104%). By measuring end-tidal gas concentrations in intubated animals during inhalation anesthesia, anesthetic depth can be checked and controlled throughout the procedure, and multiple animals can
84550.indb 494
11/11/10 6:27:56 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
495
Table€17.1╅MAC Values in Some Laboratory Animal Species Species Mouse Rat Rabbit Cat Dog Primate
Halothane 0.96–1.0 0.81–1.23 â•›0.8–1.56 â•›0.82–1.19 â•›0.86–0.93 â•›0.89–1.15
Isoflurane
Sevoflurane
N2O
1.35–1.41 1.17–1.52 2.05 1.61–1.63 1.28–1.39 1.28–1.46
— 2.4–2.5 3.7 2.58 2.10–2.36 2
150–275 155–235 — 255 188–297 200
Source: Data from Steffey, E. P. 1996. In Lumb & Jones’ Veterinary Anesthesia, 3rd ed., ed. J. C. Thurmon, W. J. Tranquilli, and G. J. Benson. Baltimore, MD: Williams & Wilkins.
be studied under the same anesthetic conditions (Brunson 1997). This is more difficult to achieve with injection anesthesia. Chronic exposure to trace anesthetic gases is a health hazard; therefore, scavenging of excess gas is essential and, in many countries, required by law. Human exposure to trace amounts of haloÂ� thane is more hazardous than exposure to isoflurane, due to halothane’s high degree of metabolism. Especially with mask and box induction, the risk of exposure is high, and isoflurane a better choice. Most countries have regulations on the use of inhalants to ensure staff safety (see section on operator safety); as a consequence, the use of halothane has been abolished in many countries. Onset, depth, and recovery of anesthesia are dependent on the concentration of the inhalant in the brain and spinal cord. The speed of induction is related to the solubility of the inhalant in the blood. Isoflurane, for example, is less soluble than halothane and produces a quicker induction. By increasing the concentration of a less soluble agent and thus the gradient from alveolus to the blood, the speed of induction can be increased. With halothane, stable anesthesia is reached within 10 min in most laboratory species (Brunson 1997). Likewise, recovery is slower with the more lipid-soluble agents. Halothane is stored in fatty tissue and, upon recovery, the storage will leave the fat and slow the return to consciousness. Inhalants are mainly eliminated by the lungs. The degree of hepatic metabolism is highest for halothane (20%), followed by enflurane (2%), isoflurane (0.2%), and desflurane (0.02%). All inhalants depress cardiopulmonary function and renal blood flow in a dose-dependent manner (Steffey 1996). Nitrous oxide (N2O) is poorly soluble in blood, oil, and fat and easily passes through biological membranes. Therefore, uptake and equilibration throughout the body are rapid. During induction with N2O and a volatile, the rate of uptake of the inhalant increases; this is especially valuable during mask induction. By adding 60–70% N2O to the inspired O2, the MAC of halothane or isoflurane is reduced by approximately 20–25%, and, consequently, cardiopulmonary function is less depressed (Brunson 1997). The lowest O2:N2O ratio should be 1:2 so that the O2 concentration in the inspiration is at least 33% (Ludders 1999). At the end of anesthesia, N2O delivery has to be discontinued first, and 100% oxygen must be given for 5–10 min to wash out the N2O and prevent the development of diffusion hypoxia. Nitrous oxide is most useful in animals with a body weight over 2 kg because, in smaller animals, volatile anesthetics equilibrate rapidly without N2O (Brunson 1997). Its use is not recommended in herbivores because it readily diffuses into gas-filled intestinal segments. Halothane induces a pronounced dose-dependent depression of myocardial contractility and sensitizes the heart for the arrhythmogenic effects of [nor]adrenaline (Brunson 1997). Cardiac output and total peripheral resistance are decreased, and cerebral blood flow is increased. At deep anesthetic levels, ventilation becomes inadequate and blood pressure falls. Hepatic metabolism of halothane during hypoxia can result in the formation of radicals, which give rise to hepatotoxicity, by reaction with cell membrane proteins (Ludders 1999). Radical formation is less likely with isoflurane or desflurane, due to their low degrees of metabolism. Because halothane is degraded to toxic end products by sunlight, it is contained in dark bottles containing a preservative (thymol).
84550.indb 495
11/11/10 6:27:56 PM
496
Handbook of Laboratory Animal Science
Isoflurane maintains overall cardiovascular functions slightly better than halothane but is a stronger respiratory depressant and has greater vasodilatory effects (Brunson 1997). It is a more potent coronary vasodilator than halothane or enflurane and does not sensitize the heart to catecholamines. In all, the safety margin is greater than with halothane. The insignificant amount of metabolism reduces the potential for hepatic and renal injury. Induction and recovery are faster than with halothane. Because deep anesthesia is rapidly achieved, care must be taken not to overdose the animal. The solubility in blood is even lower for sevoflurane than for isoflurane, which results in faster induction and recovery. Cardiovascular and respiratory effects are similar to those of isoflurane. The properties of desflurane are equal to those of sevoflurane, with the exception that the very low boiling point of desflurane makes the use of a heated vaporizer necessary for controlled delivery. Toxicity is very low. Enflurane has the potential of causing nephrotoxicity and may induce seizures in dogs (Brunson 1997). Hypnotics By definition, hypnotics are drugs that induce sleep and they have no intrinsic analgesic properties (Brunson 1997). Low doses are potentially sedative, although initial excitation is sometimes seen when they are used as monoanesthetics; higher doses can produce a light plane of sleep or anesthesia. They are used for immobilization, minor superficial surgery (in conjunction with analgesic treatment), or induction of anesthesia. They can be administered IV by injection or infusion or IP (in rodents). Barbiturates, propofol, and chloral hydrate belong to this group. Barbiturates enhance GABA-mediated inhibition of synaptic transmission and can be classified as ultrashort acting (thiopental, methohexital), short acting (pentobarbital), or long acting (Inactin®). They are poor analgesics. Sleep time is affected by sex, age, strain, time of day, room temperature, and nutritional status, as well as other factors. Administration of glucose or adrenergic agents to animals recovering from barbiturate anesthesia may result in reanesthetization (Brunson 1997). Barbiturates decrease cerebral metabolism and may, when administered repeatedly, induce development of tolerance due to induction of hepatic enzymes. Thiopental and methohexital are ultrashort-acting barbiturates that are highly lipid soluble and rapidly cross the blood–brain barrier, inducing loss of consciousness in 10–20 sec upon IV injection (Reid and Nolan 1999). Due to redistribution from the CNS to other, primarily fatty tissues, duration of action is only 5–30 min. Recovery is prolonged with repeated doses of thiopental due to storage in body fat and slow hepatic clearance. The rate of metabolism is higher for methohexital; recovery from anesthesia is faster, with fewer hangover effects, even after repeated boluses. Both agents are used for short procedures or to enable endotracheal intubation followed by inhalation anesthesia. Side effects include hypotension and respiratory depression, whereby apnea may follow induction and necessitate mechanical ventilation. Thiopental is available in a 1–2.5% solution and is highly irritating if not injected intravenously. Methohexital is less irritating. Pentobarbital is rapidly absorbed from most sites and metabolized by the liver cytochrome P450 enzyme system. Metabolism is slow, with a half-life of 38 min in the mouse and 200 min in the dog (Brunson 1997). The 6% alkaline solution is irritating and should be diluted with water for injection, except when administered IV. Pentobarbital has been shown to produce inadequate and inconsistent anesthesia in several species. It is a respiratory and cardiovascular depressant, potentially inducing hypoxemia, hypercapnia, and prolonged hypotension. Side effects include reduced blood pressure, stroke volume, pulse pressure, and central venous pressure. It impairs myocardial contractility, has a narrow margin of safety, and is especially hazardous to use in rabbits (Brunson 1997). There are no antagonists available, and recovery is slow.
84550.indb 496
11/11/10 6:27:56 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
497
Propofol is an alkylphenol that is formulated as an emulsion in soybean oil and glycerol. It produces a dose-dependent level of sleep by enhancement of the GABA-receptor function. The distribution half-life is 1–6 min, and the elimination half-life is 16–55 min (Brunson 1997). Upon IV injection, consciousness is lost in 30–60 sec, with a duration of effect of 5–10 min. IV induction often results in respiratory depression and a fall in blood pressure. Apnea can be avoided by slow injection. Intramuscular injection results in sedation only. Due to propofol’s minimal accumulation in body tissues and rapid metabolism in the liver and other internal organs, it can be used for long, light anesthesia by continuous infusion. Recovery is rapid following discontinuation of infusion. At antinociceptive doses, hypotension and a decrease in arterial blood pressure and heart rate are produced. Propofol decreases cerebral blood flow and cerebral oxygen consumption and reduces intracranial pressure (Reid and Nolan 1999). In rabbits, respiratory arrest may follow high doses (Glen 1980). By preadministration with an alpha-2 agonist, the propofol dose may be reduced by approximately 80%, and physiological variables can be well maintained (Hellebrekers et al. 1997). In swine, propofol decreases myocardial contractility. Tribromoethanol is an anesthetic primarily used in the production of genetically altered mice. After IP injection, surgical anesthesia is rapidly produced and maintained for 10–35 min (Brunson 1997). Respiration and cardiovascular functions are depressed, and high mortality has been reported after repeated anesthesia with tribromoethanol. Metabolism occurs in the liver by conjugation with gluconoronic acid. Decomposition of the drug with formation of lethal products may follow improper storage. Chloral hydrate is reduced to the active metabolite trichloroethanol in the liver. Sedative doses have a depressant effect on the cerebrum, with minimal effects on medullary centers and cardiorespiratory functions; anesthetic doses are severely depressive (Brunson 1997). Motor and sensory nerves are not affected, and the analgesic properties appear minimal. Intraperitoneal injections of high concentrations in rats may cause adynamic ileus, and large repeated doses may cause hepatic and renal damage. Intravenous administration in the dog sensitizes the heart to sudden vagal arrest or arrhythmias. Like chloral hydrate, alpha-chloralose is metabolized to trichloroethanol in the liver. The hypnotic effect is of long duration, 8–10 h in the rat (Brunson 1997). Low doses have minimal effects on reflexes and cardiorespiratory function. There are no controlled studies of the anesthetic properties, and analgesia seems to be poor. Induction is rough and seizure-like activity can be seen in dogs and cats. After IP injection in rats and guinea pigs, a severe inflammatory reaction may occur (Silverman and Muir 1993). It is only useful for nonsurvival procedures when chemical restraint with minimal cardiac and respiratory depression is needed. Urethane (ethyl carbamate) produces 8–10 h of anesthesia after IV injection and has a wide margin of safety. Intraperitoneal injection in rats results in peritoneal effusion due to toxic effects on the mesenteric vasculature (Brunson 1997). During anesthesia, cardiovascular and respiratory depression is minimal, spinal reflexes are well maintained, and effects on neurotransmission are minimal, which is why urethane is frequently used in neurophysiological studies (Albrecht and Davidowa 1989). The analgesic properties are sufficient to permit surgery in small rodents. Urethane is metabolized to carbon dioxide, ammonium, and ethanol and has carcinogenic and mutagenic properties. Metomidate and etomidate are imidazole derivatives that have GABA-mimetic effects and are useful for long-term anesthesia because of their minimal cumulative effect and good preservation of cardiovascular function (Brunson 1997). Metomidate has strong central muscle relaxing effects. Etomidate decreases intracranial pressure and brain oxygen consumption and is therefore beneficial for neurosurgery procedures. It is not recommended in dogs because of the risk of serious endocrine effects, leading to Addisonian crisis (Lukasik 1999). For surgical anesthesia, additional analgesia is required. In mice, duration and depth of metomidate anesthesia is improved by additional SC administration of fentanyl (Green et al. 1981).
84550.indb 497
11/11/10 6:27:56 PM
498
Handbook of Laboratory Animal Science
Alpha-2 Agonists Alpha-2 adrenoceptors are found in the CNS, gastrointestinal tract, uterus, and kidney, and on platelets. Their activation by alpha-2 agonistic drugs results in sedation, analgesia, muscle relaxation, and anxiolysis by inhibition of presynaptic calcium influx and neurotransmitter release. Side effects are initial hypertension due to peripheral effects, followed by slight hypotension, bradycardia, and decreased cardiac output. Intravenous fluids may be needed to counteract hypotension when it is significant. If anticholinergic agents are given before the alpha-2 agonist, they may partially prevent bradycardia, but there is also the risk that the initial hypertension is potentiated. Alpha-2 agonists depress the release of insulin, leading to elevated blood glucose levels (Lukasik 1999). They also decrease the release of antidiuretic hormone and have a direct renal tubular effect, leading to marked (nonosmotic) diuresis. Hypothermia is another common feature seen; in cats, vomiting occurs, usually within 15 min of administration. To prevent aspiration of stomach contents, induction of anesthesia should be delayed until after vomiting. Requirements of intravenous and volatile anesthetics are reduced by as much as 80% after premedication with an alpha-2 agonist. The alpha-2 antagonist atipamezole reverses the effects of all alpha-2 agonists within 5–10 min after IM or SC injection. Intravenous administration is not recommended. Xylazine is a mixed alpha-2/alpha-1 agonist, with an onset of action 3–5 min after IV injection and 10– 15 min after SC or IM injection (Lukasik 1999). Analgesia lasts for 15–30 min and sedation for 1–2 h, whereas complete recovery may take 2–4 h. In most species, xylazine alone does not produce sleep or loss of the righting reflex. It should be avoided in pregnant animals due to an increase in uterus tone. Medetomidine is highly selective for the alpha-2 adrenoceptor—more potent than xylazine and eliminated more quickly (Brunson 1997). It can be administered SC, IM, and IV. In the rat, peak brain levels are reached 15–20 min after subcutaneous administration and the elimination half-life is 1.6 h. Deep sedation and loss of righting reflex are produced in dogs, cats, and rats, but not in mice and rabbits (Figure€17.3). Medetomidine has antinociceptive actions, but they are weak in rabbits, guinea pigs, and hamsters. The addition of an opioid or a benzodiazepine will enhance analgesia and sedation. The combination with ketamine for general anesthesia offsets the bradycardia caused by medetomidine and induces a state of pronounced hypertension throughout anesthesia (Lukasik 1999).
Figure 17.3â•…Cat sedated with medetomidine.
84550.indb 498
11/11/10 6:27:57 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
499
Dissociative Anesthetics This group of lipophilic cyclohexamines has antagonistic action at excitatory N-methyl-daspartate (NMDA) receptors and produces a “dissociative state” by depression of cortical associative areas. By stimulation of the sympathetic nervous system, blood pressure is maintained at above-normal levels, despite direct myocardial depression. This is potentially useful in trauma models and in animals with hypovolemia or circulatory depression. Bronchodilation is also produced, and muscle tonus is increased. The degree of analgesia induced varies dose dependently, and metabolism occurs in the liver. Ketamine is one of the most widely used anesthetics due to its wide safety margin and compatibility with other drugs. It is a racemic mixture; l-ketamine is the likely cause of the psychomimetic side effects that occur in humans, presented as hallucinations and nightmares. Ketamine can cause tonic-clonic movements and psychomimetic emergence reactions in nonhuman primates when it is used alone (Popilskis and Kohn 1997). It can induce seizures and hallucinatory behavior in cats and dogs during emergence from anesthesia. This may be avoided by premedication with a benzodiazepine or an alpha-2 agonist. Ketamine increases hemodynamic variables and produces good analgesia except in rodents and rabbits. Surgical anesthesia can be attained by combination with an alpha-2 agonist in rabbits and rodents or with a benzodiazepine in rabbits. Absorption of ketamine is rapid after IV, IM, SC, and IP injection. Rectal administration has been described in cats (Hanna et al. 1988). Induction and recovery are rapid with IV administration. Ketamine solution has a pH between 3.3 and 5.5 and has been reported to cause tissue necrosis in rodents and rabbits when it is administered with xylazine IM (Smiler et al. 1990; Beyers, Richardson, and Prince 1991). Salivation and lacrimation are increased during ketamine anesthesia, with persisting laryngeal and swallowing reflexes. Metabolism by cytochrome P450 leads to formation of the hypnotic metabolite norketamine, which is responsible for the prolonged recovery and drowsiness seen after large ketamine doses (Lukasik 1999). Tolerance development has been demonstrated in rats (Brunson 1997). Tiletamine is closely related to ketamine, but it is longer lasting and more potent and has more pronounced side effects. It is marketed for use in combination with the benzodiazepine zolazepam (Telazol or Zoletil) and may be used for surgery in cats, dogs, ferrets, and rats. Nephrotoxicity has been reported in rabbits (Brunson 1997). Muscle Relaxants Muscle relaxants act by blocking acetylcholine receptors on the motor end plates, in a depolarizing or nondepolarizing fashion, without affecting the CNS. Their structure is related to that of acetylcholine. The use of muscle relaxants must be limited to procedures that necessitate a fully relaxed animal during general anesthesia. The degree of blockade should be kept as low as the procedure will allow. Because muscle relaxants prevent the animal from moving or showing withdrawal responses (consciously or unconsciously) during noxious stimulation, great care must be taken with their use with respect to adequacy of anesthetic depth. To ensure that the anesthetic regimen is appropriate for the specific procedure, species, or strain, it should be performed without the muscle relaxant first in some animals (Hildebrand 1997). Additionally, an adequate depth of anesthesia must be established before the muscle relaxant is added. It is not acceptable to start an infusion with a muscle relaxant and surgery before a stable plane of surgical anesthesia, such as those described in some studies, has been achieved (Van Woerkens et al. 1992; Schou et al. 1998). Finally, the muscle relaxant should be allowed to wear off intermittently in order to ensure adequate depth of anesthesia throughout the procedure.
84550.indb 499
11/11/10 6:27:57 PM
500
Handbook of Laboratory Animal Science
Signs of insufficient anesthesia during the use of muscular relaxants include increase of pulse rate unrelated to hemorrhage or slowing of the heart (bradycardia), increase or drop in blood pressure, pallor of the mucous membranes, pupil dilation, tear formation (lacrimation), increased salivation, and twitching of the tongue and facial muscles (Jones 1999). In research protocols including hemodynamic instability or cardiopulmonary bypass, blood pressure and heart rate cannot be used to ensure adequate depth of anesthesia, and the use of muscle relaxants should therefore be avoided. Intubation and positive pressure ventilation must be provided when muscle relaxants are used, as well as a means of monitoring relaxation (response to nerve stimulus). In addition to potentiating the effect of muscle relaxants, inhalation anesthesia produces muscle paralysis by a direct action on the acetylcholine receptor. In humans, this is seen increasingly with concentrations over 1 MAC and after 30 min of anesthesia (saturation effect) (Stenqvist 2000). Succinylcholine is the only depolarizing paralytic in use. Duration of this hydrophilic agent is short (5–8 min), due to quick hydrolysis by pseudocholine-esterase in the plasma and liver. Because of its depolarizing effect, initial muscle fasciculations are seen. Side effects include cardiac arrhythmia, histamine release, and muscular pain upon recovery. Succinylcholine is a potent initiator of malignant hyperthermia in pigs, dogs, and cats with a genetic predisposition, especially in combination with halothane (Hildebrand 1997). Pancuronium is a nondepolarizing muscle relaxant with a long duration of action (30 min) that blocks peripheral uptake of noradrenaline and thus causes hypertension and tachycardia. Part of the drug is excreted with the urine (70%), and part is metabolized in the liver (30%) (Jones 1999). Its effects can be reversed with neostigmine. Vecuronium is similar to pancuronium, but with a shorter duration of action and no hemodynamic effects. It is eliminated in the bile. Anesthetic Management Preanesthetic Considerations Proper preparation of the anesthetic protocol begins with the evaluation of the animal to be anesthetized. Aspects such as overall condition, underlying pathology or preexisting instrumentation may be of influence in determining the best possible anesthetic protocol for the animal and for the experiment. Consideration with respect to the experimental protocol is relevant not only concerning immediate aspects of anesthesia and instrumentation, but also relating to late effects of anesthesia on relevant physiologic functions such as neuroendocrine balance. Proper preanesthetic assessment of the experimental animal and in-depth evaluation of the experimental protocol, with regard to possible interference from the anesthetic protocol, provide the basis for a successful experiment. Depending on the animal species and character and the amount of handling needed before anesthesia, preanesthetic sedation may be considered. Preemptive analgesia should also be considered, depending on the level of invasiveness before, during, and after anesthesia. Next to the pharmacological considerations, general supportive measures such as aspects of housing and handling, temperature control, and fluid balance may exert an important influence on the overall well-being of the animal. Together with the design and execution of the anesthetic protocol, these factors will profoundly influence the quality of recovery and thus morbidity and mortality. Monitoring during Anesthesia During any anesthesia, some form of monitoring must take place. In short and relatively lowrisk procedures, it can be limited to simple, nonelectronic monitoring of pulse rate, respiration
84550.indb 500
11/11/10 6:27:57 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
501
Figure 17.4â•…Rat monitored with pulse oxymeter.
rate, body temperature, and arterial oxygenation by observing the color of the skin or mucous membranes. A simple method for obtaining relevant but only semiquantitative information on the functioning of the circulatory and the respiratory systems involves the use of a pulse oxymeter (Figure€17.4). It provides information on the arterial pulse frequency as well as the level of arterial oxygenation and thus basal information on the circulatory status and the efficiency of gas exchange. Practical applicability is limited during conditions of peripheral vasoconstriction, when the reduced signal strength prohibits proper functioning of the apparatus. Also, during oxygen supplementation, a situation of inadequate ventilation with concurrent high CO2 partial pressures may exist while arterial oxygenation is still normal or even above normal. For more invasive procedures of long duration, involving potentially high-risk interventions, more sophisticated monitoring is indicated. Electronic monitoring of heart rate and rhythm by way of a continuous electrocardiogram provides essential information about the electrical activity of the heart but not on the mechanical performance. For more detailed information on cardiac function, there are different hemodynamic monitoring options available. Direct arterial blood pressure measurement at different sites of the body provides such data. However, the more detailed the hemodynamic picture needs to be, the more sophisticated, expensive, and complicated the necessary apparatus will be. In principle, it can be stated that although modern hemodynamic monitoring techniques available in human medicine can be applied in animals as well, the size of the most frequently used animal species, such as rodents and rabbits, limits their application. Respiratory monitoring in its simplest form consists of respiration rate and rhythm. More functional information is derived from parameters such as tidal volume (expiration volume) and the percentage of end-tidal (end expiratory) CO2 and O2. Information specifically relevant to anesthesia further includes the concentration of end-tidal N2O and the volatile agent. For all these parameters, respiratory monitors are available, but as with hemodynamic monitoring, the small size of most laboratory animals makes their application difficult. Specialized equipment for use in small animal species is increasingly becoming available on the market. Further to monitoring the different essential organ functions, the overall quality of anesthesia must be evaluated during the procedure. Under all circumstances, assurances must be given to the adequacy of sleep (hypnosis) and pain control (analgesia). Traditionally, these aspects of
84550.indb 501
11/11/10 6:27:59 PM
502
Handbook of Laboratory Animal Science
Figure 17.5â•…Testing pedal withdrawal response in rat.
anesthesia have been evaluated by observing the trends in respiration (respiration rate, breathing pattern), circulation (heart rate and rhythm, arterial and venous blood pressures, and other hemodynamic variables), and autonomic function and reflexes such as eyelid and corneal reflexes, position and movement of the eyeball (depending on species), swallowing reflex, and withdrawal reflexes (Figures€17.5 and 17.6). More recently, newer techniques for evaluating specific elements of anesthetic quality have become available. These include several neurophysiological techniques such as the evaluation of the raw EEG, as well as methods for specifically assessing sleep (auditory-evoked responses) or analgesia (somatosensory-evoked responses). Although a lot of work as well as progress has been made in this field of research, the clinical applicability is still limited. Further work is needed before these methods can be easily applied to produce effective and reliable information.
Figure 17.6â•…Assessing tail pinch reflex in mouse.
84550.indb 502
11/11/10 6:28:01 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
503
Postanesthetic Care First and most importantly, when it comes to postanesthetic care, is the consideration of postÂ� intervention pain control. For this goal, opioids, NSAIDs, local analgesics, or combinations can be employed. More detailed information is provided in relevant sections of this chapter. Under certain circumstances, such as postanesthesia restlessness or automutilation, sedation may be indicated. This is more commonly seen in large animal species than in rodents and rabbits. Side effects of sedation, such as an increased risk of delayed recovery and hypothermia, are also potentially hazardous in small animals. Related to the risk of hypothermia is the need for good temperature control. This starts with proper monitoring of the body temperature until the animal is fully recovered. Supporting measures, such as warm water heating blankets, infrared heating lamps, and incubators for recovery, can dramatically reduce the incidence of life-threatening hypothermia. Installation of these measures does not, however, eliminate the need for regular measurement of body temperature. For all animal species, proper housing conditions and easy access to food and water are easily realized aspects. For rodents, this may be accomplished by assuring that food is available on the floor of the cage and by using water bottles with extra-long drinking nipples. Similarly, fluid balance should be monitored during and following anesthesia, until the animal is fully recovered. Fluid therapy before and during anesthesia can greatly improve the fluid status in the postoperative period. Last, but not least, general supportive care will improve the quality and speed of the recovery period. The matter of which measures should be part of the supportive regime depends on the circumstances and the animal species involved (Figures€17.7 and 17.8). Rodents and rabbits are often best left alone, whereas larger animal species may well benefit from extra attention and “nursing care.” Pain Management Acute Pain Recognition of acute pain in animals is commonly facilitated by the fact that the behavioral response is related in time to the actual tissue damage. Typical response characteristics include avoidance behavior, vocalization, or defensive behavior. Specific response patterns vary greatly among
Figure 17.7â•…Rats in incubator recovering after surgery.
84550.indb 503
11/11/10 6:28:03 PM
504
Handbook of Laboratory Animal Science
Figure 17.8â•…Rat provided with soft bedding and soaked diet after surgery.
species, so for a proper evaluation of the presence and extent of the pain, an in-depth knowledge of the normal behavior of the specific species is needed (ACLAM Task Force 2007). In the acute phase of initial tissue trauma, the accompanying pain can be considered to serve a physiological function of alarming the body of the possible deleterious effects of the trauma, as well as having protective qualities, thus limiting the risk of further tissue damage. Related to the physiological function of acute pain, it has been suggested that its perception is proportional to the intensity of the noxious stimulus. Under more chronic circumstances and following related neural system changes involved in recognition and processing of pain signals, the pain becomes increasingly pathological. The perception of pain no longer relates to the initial pain intensity; rather, it is by far greater, lasts longer, and has spread over an increased area. In the acute phase, the nociceptive stimulus is transported from the peripheral nociceptor to the spinal cord by way of afferent sensory fibers of the A delta and C class. The A delta fibers are myelinated and small in diameter and transport signals at high speed. These fibers are responsible for the immediate, sharp, and often intense pain; through direct connection with motor neurons at the dorsal horn level, they induce an instantaneous motor response (i.e., a withdrawal reflex). The activation of C fibers is held responsible for the longer-lasting, dull pain that follows the initial sharp pain sensation. Relevant physiological aspects of acute pain include the nocifensive behavioral response and autonomic responses such as increase in heart rate and blood pressure, a change in respiratory rate and rhythm, and endocrine changes. The autonomic responses are traditionally used as pain indicators but are unreliable as such. The unreliability of these parameters for detecting pain and differentiating it from other potential “negative” influences lies in the limited discriminative power of individual parameters. Only under circumstances where these parameters do not show any relevant changes can the absence of pain be postulated. The aforementioned parameters can at best be viewed as “circumstantial evidence” of pain being present. General Preventative and Therapeutic Strategies in Acute Pain Effective analgesic strategies for acute pain are primarily focused on preventing pain from occurring, rather than approaching pain once it has been established. The main advantage of preemptive analgesia is that sensitization has not yet developed. Early intervention prevents the development of
84550.indb 504
11/11/10 6:28:04 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
505
the inflammatory cascade and makes pain treatment more effective, both in the short term (efficacy) and in the long term (necessary duration). Consideration should be given to the different modalities of analgesic therapy, including local analgesia, systemic use of opioid or nonsteroidal anti-inflammatory drugs, and the alpha-2 adrenergic drugs used for sedation and general anesthesia. The choice of one (single therapy) or multiple (polymodal therapy) analgesic drugs should be determined on the basis of species- and model-dependent characteristics as well as on the underlying pathology. Each class of analgesic drugs has its own specific characteristics and efficacy, depending on the circumstances (see section on analgesic drugs). Chronic Pain The pathophysiological background of chronic pain involves not only temporal aspects (i.e., long duration) of this class of pain, but also functional changes in the neurophysiological pain system. The functional changes result in an increased intensity of nociceptive stimulus perception, as well as an enlargement of the sensitive area. In the present day, understanding of the consequences of continued nociceptive activation of the peripheral and central nervous system (i.e., pain sensitization) plays a crucial role. Following initial nociceptor stimulation and recognition of the painful character of stimulation, the repetition of stimulation initiates a process of production and release of inflammatory mediators at the site of injury (i.e., the stimulation site) (Woolf and Chong 1993). Such prolonged nociceptor stimulation may be the consequence of continued tissue trauma, of long-duration sympathetic stimulation, or the presence of the inflammatory response. Vasoactive amines released from the damaged tissue as well as inflammatory cells and neuropeptides released from the injured nerve endings in the traumatized area are collectively responsible for an increased sensitivity of peripheral nociceptors. The peripheral sensitization results in an increased responsiveness of these nociceptors (hyperalgesia) and also in non-noxious stimuli to be perceived as painful (allodynia). Next to peripheral sensitization, continued nociceptor input into the dorsal horn of the spinal cord changes the responsiveness of the dorsal horn neurons, thereby increasing the responsiveness to incoming stimuli. Consequently, low-intensity stimulation and even activation of nervous system structures normally involved in tactile sense (A-beta receptors and fibers) now lead to a potentially intense, painful sensation (Woolf and Chong 1993). The overall result of chronic pain is that, through peripheral and central sensitization, low-intensity stimulation of a body area previously involved in painful stimulation now produces a response that is more intense, of longer duration, and with the painful sensation spreading over a larger area. The consequence is obvious for the clinical effectiveness of analgesic therapy. Once sensitization has set in, effective pain treatment will be harder to achieve and require a longer duration (Hellebrekers 2000). Analgesic Modalities for Treating Acute or Chronic Pain The primary aspect to consider in the treatment of acute pain involves timely administration, i.e., preferably before the pain occurs. This is an option in a properly designed anesthetic protocol for performing a surgical or investigational procedure. In order to prevent pain from occurring before, during, or after surgery, it is of utmost importance to initiate treatment early and to continue treatment during and after the intervention at correct (pharmacologically determined) intervals. The aspect of the choice of drug (class) is determined by the pathophysiological background of the pain, allowing a primary choice between opioid analgesics and NSAID therapy (Lascelles 2000). Furthermore, specific treatment modalities such as local analgesic supplementation may be applied under specific circumstances.
84550.indb 505
11/11/10 6:28:04 PM
506
Handbook of Laboratory Animal Science
For the alleviation of immediate postsurgical pain, most evaluations and comparative studies show opioids to be most effective during the first 24–48 h after surgery. Following this initial period, the inflammatory component of pain becomes more prominent, and NSAID therapy can be applied with good success. In circumstances where the pathophysiological background of the pain is more complex, or when single-mode therapy has proven to be inadequate, multimodal therapy with a combination of an opioid and an NSAID is indicated. Additional local analgesic supplementation might further support the reduction of pain and stress in the animal. For specific details (dosage, interval, and route of administration) of the different analgesic treatment modalities, the reader is referred to the relevant sections of this chapter. Operator Safety Safety aspects related to anesthesia concern exposure to anesthetic gases, handling of gas cylinders, and possible misuse of drugs. Anesthetic Gases Chronic exposure to trace anesthetic gases has been linked to hepatic and renal disease, abortion, infertility, birth defects, and neoplasia in humans (Ludders 1999). Nitrous oxide can inhibit bone marrow function and cause abortion. Several countries have legislation related to the use of anesthetic gases. In the United Kingdom, the Control of Substances Hazardous to Health Regulations regulate the control of anesthetic gas pollution. Regulations include risk assessment, prevention and control of exposure, installation and maintenance of control measures, monitoring of exposure, and health surveillance. Information and training of staff is also included, as are exposure limits. In the United States, safety standards are issued by the National Institute for Occupational Safety and Health (Brunson 1997). Set exposure limits for anesthetic gases and recommendations on how to limit exposure are included. Low fresh gas flows and scavenging systems minimize exposure to trace gases. Using double mask systems connected to the exhaust system and ventilated benches and hoods reduces exposure further. Old-type vaporizers should only be filled with volatiles in ventilated hoods. If exposure cannot be avoided, the use of safer anesthetics, like isoflurane, is recommended before using less safe ones, like halothane. Activated charcoal systems have a limited use and will absorb halogenated volatiles but not nitrous oxide. Gas Cylinders Gas cylinders are filled to high pressures, and explosions are possible if they are mechanically damaged by falling over or being dropped or exposed to heat (Clutton 1999). Cylinders must not be stored near sources of heat or combustible material. Oxygen cylinder valves and associated equipment must not be lubricated and must be free from carbon-based oils and greases. The combination of these with high-pressure oxygen may result in explosion. Smoking or naked lights must not be allowed within the vicinity of a cylinder or pipeline outlet. Controlled Drugs Many of the opioid drugs are subject to control under national legislation. In the United Kingdom, for example, all pure opioid agonists must be kept in locked cupboards and records kept of purchase and usage (Nolan 2000). The drugs are subject to special prescription requirements and legislation on destruction of unwanted stock. Safekeeping requirements also apply to buprenorphine. Butorphanol is subjected to controlled legislation in the United States but not in
84550.indb 506
11/11/10 6:28:04 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
507
the United Kingdom. Other drugs with a potential for misuse are ketamine, benzodiazepines, and barbiturates. Careful storage and record of usage are recommended for these drugs as well. Analgesia and Anesthesia of Different Species Analgesia and Anesthesia in Rodents Analgesia Like other species, rodents in painful conditions benefit from analgesic treatment. Postoperative pain will cause inappetence and prolong recovery in rats (Liles and Flecknell 1992). Recent research focuses on postoperative recovery and recognition of pain-related behavior (Roughan and Flecknell 2001). Studies on the effects of perioperative opioid or NSAID treatment are increasing (Flecknell, Roughan, and Stewart 1999; Roughan and Flecknell 2001; Jablonski, Howden, and Baxter 2001; Hayes et al. 2000). Morphine is a potent analgesic of short duration (2–4 h) in the rat, causing sedation and respiratory depression (Wixson and Smiler 1997). Apart from central effects, morphine appears to exert a peripheral effect. In a mouse model of visceral pain, a low dose of morphine provided efficient analgesia after IP but not IV administration (Reichert et al. 2001). Butorphanol is 30 times more potent than morphine in rats. It can be used to treat mild to moderate pain and has a duration of action lasting 1–4 h (Gades et al. 2000). By combination with an NSAID, more potent analgesia of longer duration is provided. Buprenorphine is about 25 times more potent than morphine but has less maximum analgesic effect, which is reached 30 min after injection. Duration of action is between 3 and 12 h, depending on species and test used (Wixson and Smiler 1997; Gades et al. 2000). Clinical experience suggests that readministration is needed every 8–12 h. Low-dose buprenorphine has a beneficial effect on the postoperative recovery in rats (Liles and Flecknell 1994; Flecknell, Roughan, and Stewart 1999; Jablonski et al. 2001). Systemic buprenorphine has been shown to be superior to local infiltration with bupivacaine in terms of postoperative recovery following laparotomy in rats (Liles and Flecknell 1993). Reported side effects from buprenorphine in rats include ingestion of bedding and gastric distension and obstruction, seemingly related to dose and strain (Clark et al. 1997; Jacobson 2000). Buprenorphine can be administered SC or orally. Higher doses are needed for oral administration due to first-pass hepatic metabolism, but high levels of buprenorphine have been documented in the circulation following oral ingestion (Kalliokoski et al. 2009). Rats can be trained to eat fruit, meat-flavored jelly cubes (1 cm3 per 200 g rat), or commercially available nut paste containing buprenorphine for postoperative oral treatment (Figure€17.9) (Flecknell, Roughan, and Stewart 1999; Dobromylskyj et al. 2000; Kalliokoski et al. 2009; Goldkuhl, Hau, and Abelson 2010). Preadministration of buprenorphine reduces the need of isoflurane for surgery by approximately 20% (Flecknell 1999). When injection anesthesia is used, it is safer to administer buprenorphine at the end of the procedure, when anesthesia reaches a light plane. After extensive surgery, buprenorphine may be combined with an NSAID as well as a local anesthesia. Buprenorphine may also be administered to reverse the effect of fentanyl after the use of fentanyl/fluanisone (Hypnorm) + midazolam for surgery, while at the same time providing postoperative analgesia. NSAIDs are indicated after minor to intermediate surgery. Premedication with carprofen or ketoprofen before abdominal surgery in rats will reduce pain-related behavior for 4–5 h and speed recovery (Liles and Flecknell 1993; Roughan and Flecknell 2001). Clinical experience suggests that analgesia may last longer. Ibuprofen treatment in mice results in a more rapid return of activity and water intake after abdominal surgery (Hayes et al. 2000). There are no reports on adverse reactions with the use of the modern NSAIDs, but prolonged use should be avoided (Dobromylskyj et al. 2000).
84550.indb 507
11/11/10 6:28:04 PM
508
Handbook of Laboratory Animal Science
Figure 17.9â•…Rat eating buprenorphine jelly.
The initial dose can safely be repeated after 24 h. Carprofen is one NSAID that is proven safe to use preoperatively; other NSAIDs can cause renal toxicity if hypotension occurs during anesthesia. Local Anesthetics In most species, maximum safe doses are 4 mg/kg of lidocaine and 1–2 mg/kg of bupivacaine (Dobromylskyj et al. 2000). The duration of action of bupivacaine seems shorter in rats than in larger species. Reports of toxicity in rodents relate dose rates similar to those toxic in dogs and cats. Overdosing is more easily accomplished in small animals and can be avoided by accurately calculating the dose rate. Doses can be diluted to provide a more appropriate volume. General toxic dose rates are 10–20 mg/kg of lidocaine and 4 mg/kg of bupivacaine IV. In adult rats, the LD50 of bupivacaine is 30 mg/kg (Van Ham et al. 1996). Addition of adrenaline should be avoided in ring blocks of appendages. Immersion of the mouse tail in a dimethyl sulfoxide (DMSO) solution with lidocaine for 2 min provides effective analgesia in the heat tail-flick test (Kolesnikov et al. 2000). Antidepressants Tricyclic antidepressants (TCAs) or serotonin reuptake inhibitors (SSRIs) can be considered for treatment of pain when opioids and NSAIDs are contraindicated. The analgesic effect of TCAs and SRRIs is due to a blockade of serotonergic reuptake. Amitriptyline and imipramine (TCAs) have been shown to decrease signs of pain perception in neuropathic pain models for up to 2 weeks in mice and rats, without development of tolerance or adverse effects (Wixson and Smiler 1997). The SSRI fluvoxamine shows an antinociceptive effect in the hot plate test in mice (Roughan, Ojeda, and Flecknell 1999). Anesthesia Premedication The use of premedication is commonly excluded in rodent anesthesia. Sedatives and tranquilizers are usually administered together with injectable anesthetics. The use of atropine is frequently not a routine but should be considered with anesthetics causing bradycardia (opioids) or
84550.indb 508
11/11/10 6:28:05 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
509
excessive salivation (ketamine). In a number of circumstances, premedication with analgesics or sedatives could be advantageous in rodent anesthesia. Sedatives may be used in combination with local anesthetics for minor superficial surgery or before induction with intravenous anesthetics. Preadministration with an alpha-2 agonist (e.g., medetomidine) before induction with propofol IV or inhalation anesthesia reduces stress and decreases the anesthetic dose needed by as much as 80% in other species (Hellebrekers et al. 1997; Lukasik 1999) and probably also in rodents. The alpha-2 agonistic effect can be reversed if sedation is unwanted after recovery. Fentanyl + fluanisone (Hypnorm), administered SC, will provide analgesia and sedation sufficient for minor surgery or, if combined with a benzodiazepine, surgical anesthesia. Premedication with buprenorphine has been shown to reduce the pentobarbital dose needed for provision of surgical anesthesia, but it does not improve respiration (Roughan et al. 1999). Buprenorphine administration before isoflurane anesthesia will reduce MAC by approximately 20% (Flecknell 1999). Buprenorphine before ketamine + medetomidine anesthesia has proven dangerous in rats due to severe respiratory depression, although this may be considered dose dependent (Hedenqvist et al. 2000a). Among the NSAIDs, carprofen and ketoprofen are usually safe for preanesthetic administration (Hanna et al. 1988). Inhalation Anesthesia Most volatile agents are safe to use in rodents. Hepatotoxicity has been reported after low concentrations of halothane in guinea pigs (Wixson and Smiler 1997). Hemodynamic parameters are often more stable during inhalation anesthesia compared with injectable anesthesia (Chaves, Weinstein, and Bauer 2001); due to safety and stable physiological effects, inhalation anesthesia is recommended in mouse cardiovascular, cerebral, and noninvasive imaging studies (Rao and Verkman 2000; Balaban and Hampshire 2001). Premedication with buprenorphine, carprofen, or medetomidine will enhance analgesia and reduce the volatile dose needed for surgical anesthesia. Table€17.2╅Dose Rates for Analgesics in Rodents Drug
Mouse
Morphine Butorphanol Buprenorphine
2–5 mg/kg SC; 4 hourly 1–2 mg/kg SC; 4 hourly 0.05–0.1 mg/kg SC; 8–12€hourly
Carprofen
5 mg/kg SC or by mouth; 24€hourly
Ketoprofen Ibuprofen Lidocaine Bupivacaine Amitriptyline Imipramine
30 mg/kg by mouth; 24€hourly 4 mg/kg or 0.4 mL/kg of a 1% solution 1–2 mg/kg or 0.4–0.8 mL/kg of a 0.25% solution 1.2–5 mg/kg SC or IP; 3–12€hourly 2.3 mg/kg SC or IP; 12–24€hourly
Rat 2–5 mg/kg SC; 4 hourly 1–2mg/kg SC; 4 hourly 0.01–0.05 mg/kg SC or IV; 0.1–0.25 mg/kg by mouth; 8–12 hourly 5 mg/kg SC or by mouth; 24€hourly 5 mg/kg SC or by mouth; 24€hourly 15 mg/kg by mouth; 24 hourly
Guinea Pig 2–5 mg/kg SC; 4 hourly 2 mg/kg SC 0.05 mg/kg SC; 8–12€hourly — — —
1–10 mg/kg SC or IP; 3–12€hourly 10 mg/kg SC or IP; 12–24€hourly
Source: Data from Dobromylskyj, P. et al. 2000. In Pain Management in Animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders; and Wixson, S. K., and K. L. Smiler. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
84550.indb 509
11/11/10 6:28:05 PM
510
Handbook of Laboratory Animal Science
Table€17.3╅Drug Dose Rates for Premedication in Rodents Premedication Atropine Acepromazine Diazepam Midazolam Xylazine Medetomidine Fentanyl + fluanisone (Hypnorm) Buprenorphine Carprofen
Mouse
Rat
Guinea Pig
40 µg/kg SC or IM 2.5 mg/kg IM 5 mg/kg IM or IPa 5 mg/kg IM or IPa 5–10 mg/kg IP 30–100 µg/kg SC 0.5 mL/kg IM or IPa
40 µg/kg SC or IM 2.5 mg/kg IM 2.5 mg/kg IM or IPa 5 mg/kg IP 1–5 mg/kg IM or IPa 30–100 µg/kg SC 0.5–1.0 mL/kg IM, SC or IPa
50 µg/kg SC or IM 2.5 mg/kg IM 2.5 mg/kg IM or IPa 5 mg/kg IM or IPa 5–40 mg/kg IP — 0.5 mL/kg IM
0.05–0.1 mg/kg SC 5 mg/kg SC or by mouth
0.01–0.05 mg/kg SC 5 mg/kg SC or by mouth
0.05 mg/kg SC —
Sources: Data from Wixson, S. K., and K. L. Smiler. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press; and Lukasik, V. M. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. a IM and SC administration of a drug are likely to give a greater effect compared to IP injection, due to possible first-pass hepatic metabolism after IP injection.
By using local anesthesia as an adjunct, the concentration of the volatile agent can be kept to a minimum, with less cardiorespiratory depression as a result. Induction is best achieved in a plexiglas anesthetic chamber equipped with a gas inlet and an outlet for exhaled and excess gases (Figure€17.10). With halothane or isoflurane, induction is complete after 2–3 min. Anesthesia can be maintained by gas delivery via a face mask (Figure€17.11) or an endotracheal tube. Successful intubation has been described in rats, guinea pigs, hamsters, and mice (Figure€17.12) (Dalkara et al. 1995; Berul et al. 1996; Wixson and Smiler 1997). After 30 min of isoflurane or halothane anesthesia, animals will regain their righting reflex within 5–10 min. A low fresh gas flow is beneficial for keeping costs down and protecting the environment. In an open breathing system, a flow of three times the animal’s minute volume is necessary in order to meet fresh gas requirements. The minute ventilation in rats and mice is approximately 700 mL/kg/min (Steffey 1996), resulting in a required fresh gas flow of approximately 0.06 L/min for a 30 g mouse and 0.6 L/min for a 300 g rat. Some vaporizers (e.g., Mark III tecs) deliver accurate concentrations of volatile anesthetics at very low flows (0.2 L/min) (Clutton 1999). Another type of
Figure 17.10â•…Rat in induction chamber: inlet for anesthetic gas (bottom) and outlet for scavenging (top).
84550.indb 510
11/11/10 6:28:06 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
511
Figure 17.11â•…Rat maintained on face mask.
rodent anesthetic machine, managing even lower gas flows (0.05 L/min), works by injection of the volatile anesthetic into the fresh gas stream (Univentor 400 Anaesthesia Unit, www agnthos.se). Oxygen may be used as sole carrier gas, except in very long procedures. If near physiological blood oxygen levels are needed, air or nitrous oxide can be mixed with oxygen in a 2:1 ratio. In order to prevent the development of hypoxia, the inspired gas should always consist of 33% oxygen as a minimum. Arterial CO2 levels will be elevated during anesthesia, unless mechanical ventilation is used. For brief procedures in mice, anesthesia may be induced with a volatile agent in a glass jar. Close observation is mandatory because ambient concentration of the inhalant within the jar may vary and result in an increased anesthetic risk. Under standard temperature and pressure conditions, 1 mL of liquid isoflurane will produce 182 mL of gas (Brunson 1997). In a 500 mL jar, 0.11 mL of liquid isoflurane will quickly evaporate to reach a concentration of 4%. A mouse placed in the jar will be anesthetized within a few minutes and wake up 30–60 sec after removal from the jar. The animal must never come in contact with the liquid anesthetic, or local irritation will occur (Wixson and Smiler 1997). If several mice are anesthetized one after the other in the jar, care must be taken to prevent a shortage of oxygen. For protection of staff, this method should only be used in a ventilated hood.
Figure 17.12â•…Rat on Harvard ventilator.
84550.indb 511
11/11/10 6:28:09 PM
512
Handbook of Laboratory Animal Science
Table€17.4╅ Induction and Maintenance Concentrations for Volatile Anesthetics in Rodents Volatile Agent Isoflurane Sevoflurane Halothane
Induction Concentration 2–4% 4–8% 2–4%
Maintenance Concentration 5–3% 3–4% 1.5–2%
Sources: Data from Wixson, S. K., and K. L. Smiler. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press; and Brunson, D. B. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
Injection Anesthesia Because most injectable anesthetics depress ventilation, oxygen supplementation is recommended during anesthesia. Intubation and mechanical ventilation may be necessary during long-term anesthesia. First choice combinations for surgical procedures, from the aspect of safety and efficiency, are Hypnorm + midazolam, or ketamine + medetomidine. Each combination can be mixed and administered with one injection IP or SC. The recommended IP dose should be reduced by 30–50% for SC administration because no immediate hepatic metabolism occurs after SC injection. Hypnorm (fentanyl + fluanisone) provides sedation and analgesia sufficient for minor superficial procedures, although muscle relaxation is poor. In combination with midazolam or diazepam, surgical anesthesia with good muscle relaxation is produced. Unlike midazolam, the diazepam preparation is not compatible with Hypnorm and must be administered separately. Duration of surgical anesthesia for either combination is 20–40 min. Additional SC doses of Hypnorm may be given to prolong anesthesia for several hours. The effects of fentanyl can be antagonized with buprenorphine or butorphanol when postoperative pain is expected. Ketamine produces immobility but poor analgesia. In combination with medetomidine or xylazine, surgical anesthesia is consistently achieved, except in guinea pigs (Wixson and Smiler 1997). Duration of surgical anesthesia is 15–30 min in rats and up to 80 min in mice, depending on dose and strain. Side effects are less severe than during pentobarbital anesthesia (see later discussion). If anesthesia proves insufficient, it is best improved by adding a low concentration of inhalation anesthesia. Medetomidine and xylazine can be reversed with atipamezole and, if surgery has been undertaken, an analgesic should be provided before reversal. Long-term anesthesia (12 h) with ketamine and xylazine IV infusion has been described in rats (Simpson 1997). Pentobarbital is best used to provide light anesthesia only, due to its narrow safety margin. Dilution with saline is recommended before IP injection to prevent peritoneal irritation. Severe respiratory depression and moderate circulatory depression are produced even at light planes of anesthesia. Local anesthetics may be added to provide analgesia, or anesthesia may be improved by the addition of an inhalant agent (e.g., isoflurane). Intravenous administration of thiopental, methohexital, and propofol can be used for short-term anesthesia. Propofol can be given repeatedly IV (or, better, by continuous infusion) for procedures lasting at least 3 h. Analgesia is poor with propofol but can be improved by premedication with Hypnorm in rats (Brammer, West, and Allen 1993). Alphaxalone + alphadolone (Saffan®) is another anesthetic useful for repeated IV injections or long-term infusion, except in guinea pigs, which may develop pulmonary edema. Intravenous anesthesia with isoflurane has been described in mice. An average infusion rate of 1.6 µL/min of isoflurane in Intralipid was required for maintenance (Eger and MacLeod 1995). Chloral hydrate can be used in rats and mice to produce surgical anesthesia but with a narrow safety margin. Severe acidosis and cardiovascular and respiratory depressions are produced. Tribromoethanol produces approximately 15 min of surgical anesthesia after IP administration in mice. Induction and recovery are fast. Exposure to light or improper storage will cause tribromoethanol to decompose and
84550.indb 512
11/11/10 6:28:09 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
513
then it will cause inflammation if injected. Proper storage is in the dark, at 4°C (Papaioannou and Fox 1993). Urethane will produce prolonged deep surgical anesthesia and 24 h of sleep after IP injection in rats (Wixson and Smiler 1997). Progressive side effects are acidosis, hypotension, bradycardia, peritoneal fluid accumulation, and decline in renal function. Urethane is a proven carcinogen and mutagen in rodents. Analgesia and Anesthesia in Rabbits Analgesia Few controlled clinical studies have been published on NSAIDs and opioids in rabbits, but experience shows that a single dose of the NSAID carprofen provides up to 24 h of postoperative analgesia and the opioid buprenorphine 8–2 h (Dobromylskyj et al. 2000). Rabbits seem to benefit from a single dose of buprenorphine, administered toward the end of a surgical procedure. After more extensive surgery, a combination of an NSAID, an opioid, and, if needed, a local anesthetic will provide excellent analgesia. Local anesthetics may be used in similar ways as in other species. EMLA cream provides analgesia before venipuncture or catheter placement in the ear. After the fur is clipped, the cream is applied for 1 h under an adhesive tape, providing skin-deep anesthesia for 1 h (see Figure€17.1). Table€17.5â•… Injectable Anesthetics in Rodents Drug Fentanyl + fluanisone (Hypnorm) + midazolam Fentanyl + fluanisone (Hypnorm) + diazepam Ketamine + medetomidine Atipamezole Ketamine + xylazine Pentobarbital Thiopental Methohexital Propofol Alphaxalone + alphadolone (Saffan) Tribromoethanol Chloral hydrate
Alpha-chloralose Urethane
Mouse
Rat
Guinea Pig
10–13 mL/kg IP of a premixed solutiona 0.4 mL/kg IP + 5 mg/ kg IP 50–75 mg/kg + 1–10 mg/kg IP 1–2.5 mg/kg IM or SC 80–100 mg/kg + 5–10 mg/kg IP 40–50 mg/kg IP 30–40 mg/kg IV 10 mg/kg IV 26 mg/kg IV in repetitive boluses 14 mg/kg IV + 4–6 mg/ kg every 15 min
2.7–4 mL/kg IP of a premixed solutiona 0.6 mL/kg IP + 2.5 mg/ kg IP 75 mg/kg + 1 mg/kg IP
8 mL/kg IP of a premixed solutiona 1 mL/kg IP + 2.5 mg/kg IP
1 mg/kg IM or SC 75–100 mg/kg + 10 mg/ kg IP 40–50 mg/kg IP 10–15 mg/kg IV 10–15 mg/kg IV 10 mg/kg IV + 44–55 mg/kg/h IV 10–15 mg/kg IV + 0.25–0.45 mg/kg/min IV of a 1:10 dilution 300 mg/kg IP (0.25% solution) 300–450 mg/kg IP + 40 mg/kg/h IP (5% solution) 31–65 mg/kg IP (5% solution) 1200–1500 mg/kg IP (50% solution)
1 mg/kg IM or SC 40 mg/kg + 5 mg/kg IP
250 mg/kg IP (1.2% solution) 450 mg/kg IP + 40 mg/ kg/h IP (5% solution) 114 mg/kg IP (5% solution) —
40 mg/kg + 0.5 mg/kg IP
37 mg/kg IP — — — 16–20 mg/kg IV; no repetition — 400 mg/kg IP
— —
Sources: Data from Wixson, S. K., and K. L. Smiler. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press; Flecknell, P. A. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA; Cruz, J. I., J. M. Loste, and O. H. Burzaco. 1998. Laboratory Animals 32:18; and Tanaka, N. et al. 1996. Circulation 94:1109. a Mix one part Hypnorm with two parts water for injection first, then add one part midazolam (5 mg/mL).
84550.indb 513
11/11/10 6:28:09 PM
514
Handbook of Laboratory Animal Science
Table€17.6╅Dose Rates for Analgesics in Rabbits Analgesic Buprenorphine Butorphanol Morphine Carprofen Ketoprofen Lidocaine Bupivacaine
Dose 0.01–0.05 mg/kg IM, SC, or IV; 6–12 hourly 0.1–0.5 mg/kg IM, SC, or IV; 4 hourly 2–5 mg/kg IM or SC; 4 hourly 4 mg/kg SC daily or 1.5 mg/kg by mouth; twice daily 3 mg/kg SC; daily 4 mg/kg or 0.4 mL/kg of a 1% solution 1–2 mg/kg or 0.4–0.8 mL/kg of a 0.25% solution
Source: Data from Dobromylskyj, P. et al. 2000. In Pain Management in Animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders.
Anesthesia Premedication Most preanesthetic drugs may be used in rabbits. Many rabbits have high plasma levels of atropinase, so glycopyrrolate may be a better choice if an anticholinergic is indicated. Glycopyrrolate elevates the heart rate for 60 min in rabbits and will prevent bradycardia during ketamine + xylazine anesthesia (Olson et al. 1994). Inhalation Anesthesia Rabbits react with extended breath holding and severe struggling upon mask or chamber induction with most volatile agents (Flecknell, Roughan, and Hedenqvist 1999; Hedenqvist et al. 2001). The result of breath holding is a marked decrease in heart rate and arterial oxygen levels, as well as an elevation in blood carbon dioxide levels; thus, induction with inhalation anesthesia is best avoided in rabbits. Premedication with a sedative or tranquilizer will eliminate struggling but not always prevent breath holding. Anesthesia is best induced with an injectable agent, after which the rabbit can be intubated and maintained on inhalation anesthesia (Figure€17.13).
Table€17.7╅Drug Dose Rates for Premedication in Rabbits Drug Acepromazine Atropin Glycopyrrolate Fentanyl + fluanisone (Hypnorm) Diazepam Midazolam Xylazine Medetomidine
Dose 0.25–0.75 mg/kg IM or SC 40 µg/kg SC or IM 0.1 mg/kg SC or IM or 0.01 mg/kg IV 0.2–0.5 mL/kg IM or SC 1–2 mg/kg IM or IV 1–2 mg/kg IV or IM 3–5 mg/kg IM or IV 0.1–0.5 mg/kg IM or SC
Sources: Data from Flecknell, P. A. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA; and Lipman, N. S., R. P. Marini, and P. A. Flecknell. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
84550.indb 514
11/11/10 6:28:09 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
515
Figure 17.13â•…Blind endotracheal intubation of an anesthetized rabbit.
Injection Anesthesia Fentanyl + fluanisone (Hypnorm) alone may be sufficient for minor superficial surgery. When they are used in combination with midazolam, anesthesia sufficient for invasive surgery is produced. It is best to administer Hypnorm SC first; after 10–15 min, the ear vein can be cannulated and midazolam administered IV, until a plane of surgical anesthesia is achieved. After completion of surgery, fentanyl can be reversed with butorphanol or buprenorphine. Another commonly used combination is ketamine + medetomidine (Henke et al. 2005; Grint and Murison 2008). The drugs can be mixed and injected together, or medetomidine can be administered SC first for sedation, followed by ketamine IM, SC, or IV. Arterial oxygen levels are significantly reduced by this combination, and supplemental oxygen is recommended (Hellebrekers et al. 1997; Hedenqvist et al. 2001). The heart rate is reduced, but arterial blood pressure is maintained. Medetomidine may be reversed with atipamezole. If surgery has been undertaken, an analgesic should be administered before reversal. Ketamine may also be combined with acepromazine, midazolam, or diazepam for provision of light to moderate surgical anesthesia. Each sedative may be administered prior to ketamine or mixed with ketamine. Table€17.8â•…Doses for Injectable Anesthesia in Rabbits Drug Fentanyl + fluanisone (Hypnorm) + midazolam Ketamine + medetomidine Atipamezole (antidote) Ketamine + xylazine Ketamine + acepromazine Ketamine + midazolam Ketamine + diazepam Thiopentone Methohexital Propofol
Dose 0.3 mL IM or SC + 1–2 mg/kg IV 5–15 mg/kg + 0.25–0.35 mg/kg IM or SC 1 mg/kg IM 35–50 mg/kg + 5–10 mg/kg IM 50 mg/kg IM + 1 mg/kg IM 25 mg/kg IM + 5 mg/kg IM 25 mg/kg IM + 5 mg/kg IM 1–2% dilution IV to effect (6–50 mg/kg) 5–15 mg/kg of a 1% solution IV 7–10 mg/kg IV + 0.9 mg/kg/min IV
Source: Data from Flecknell, P. A. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA.
84550.indb 515
11/11/10 6:28:11 PM
516
Handbook of Laboratory Animal Science
Pentobarbital is hazardous in rabbits and should not be used. Thiopental and methohexital can be administered IV for short procedures or to allow intubation for maintenance with inhalation anesthesia. Propofol administered IV produces light anesthesia that is insufficient for surgery. Injections can be repeated or better, continuous infusion administered (Martinez, Murison, and Love 2009). In combination with local anesthesia, superficial surgery may be performed. By premedication with medetomidine, less propofol is needed and surgical anesthesia can be provided. Analgesia and Anesthesia in Pigs Analgesia The NSAID carprofen may be used to control postsurgical pain in pigs by IV or SC administration for up to 3 days without side effects (Dobromylskyj et al. 2000). Acetylsalicylic acid and acetaminophen may be administered orally for acute or chronic, mild to moderate pain. Ketorolac (NSAID) has been reported effective in relieving pain after major surgery (Andersen et al. 1998). The opioid buprenorphine has been shown to be effective in thermal analgesiometry tests (Dobromylskyj et al. 2001). Buprenorphine and butorphanol are long acting, with few side effects. For minor surgery, repeated administration of analgesics may not be necessary, whereas a few days of opioid treatment alone or in combination with an NSAID are needed after major surgery. Local anesthetics may be used as in other species. Combinations of local anesthetics infiltrated along the incision line with parenteral opioid analgesics and general anesthesia reduce effects from noxious stimulation as well as recovery times in pigs (Swindle 1998). Anesthesia Premedication Atropine and glycopyrrolate are used to decrease secretion, abolish the vaso-vagal reflex during intubation, and prevent bradycardia (Swindle 1998). Acepromazine, diazepam, midazolam, and azaperone can be used to relieve anxiety and will decrease the amount of general anesthetic required. The duration of action is 8–24 h for acepromazine and 20–50 min for azeperone. Inhalation Anesthesia Isoflurane, desflurane, and sevoflurane are all good for use in pigs (Swindle 1998). They provide better control of anesthesia and shorter recovery periods than many of the injectable agents, and isoflurane provides the least myocardial depressant effect of all the inhalants. Mask induction followed by endotracheal intubation is possible in a well-ventilated area, for protection of staff (Figure€17.14). Nitrous oxide may be used in 1:1 or 1:2 combinations with oxygen in order to reduce the amount of inhalant needed (20–25% reduction). Injection Anesthesia Endotracheal intubation should always be performed during general anesthesia in pigs (Figure€17.15) (Swindle 1998). Ketamine and tiletamine + zolazepam (Telazol or Zoletil) are the most common injectables used in swine anesthesia. Ketamine is not sufficient for surgical procedures (Boschert et al. 1996); however, in combination with acepromazine, diazepam, midazolam, azaperone, xylazine, or medetomidine, it may be used for minor procedures. Intravenous infusion with ketamine and midazolam can provide visceral analgesia suitable for major surgery (Swindle 1998), but it also produces
84550.indb 516
11/11/10 6:28:11 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
517
Figure 17.14â•…Face mask induction of piglet with isoflurane.
profound hypothermia. A loading dose of ketamine precedes continuous infusion. The combination of iletamine + zolazepam provides 20 min of anesthesia for minor surgery. Barbiturates can be administered IV after induction of anesthesia with ketamine. Thiopental, administered by IV infusion, has the least cardiovascular depressant effect of all barbiturates. Pentobarbital infusion can be used for nonsurvival procedures in artificially ventilated animals. By adding fentanyl, profound analgesia is provided. For cardiac surgery, a combination of IV opioid infusion and inhalation anesthesia may be used (Swindle 1998; Husby et al. 1998, Kaiser et al. 2003). Midazolam or ketamine is given preoperatively to allow catheter placement. Opioid infusions have the advantages of not decreasing myocardial contractility and coronary blood flow. By starting the infusion before the bolus is given, muscle rigidity and sudden bradycardia are avoided. Propofol has a narrow margin of safety in swine and can produce severe hypotension (Schoffmann et al. 2009). Paralytic agents may be indicated for some procedures, but should not be administered until it is established that adequate surgical anesthesia has been obtained. Paralyzed animals must be monitored for changes in heart rate or blood pressure as an indication of inadequate anesthesia.
Figure 17.15â•…Endotracheal intubation of piglet with a 3.0 mm endotracheal tube.
84550.indb 517
11/11/10 6:28:13 PM
518
Handbook of Laboratory Animal Science
Table€17.9╅Dose Rates for Analgesics in Pigs Drug
Dose
Carprofen Acetylsalicylic acid (enteric-coated aspirin) Ketorolac Buprenorphine Butorphanol Lidocaine Bupivacaine
2–4 mg/kg SC or IV daily 10 mg/kg by mouth; 4–6 hourly 1 mg/kg IM or IV 0.005–0.05 mg/kg IV or IM; 6–12 hourly 0.1–0.3 mg/kg; 6 hourly 4 mg/kg or 0.4 mL/kg of a 1% solution 1–2 mg/kg or 0.4–0.8 mL/kg of a 0.25% solution
Sources: Data from Dobromylskyj, P. et al. 2000. In Pain Management in Animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders; and Swindle, M. 1998. Surgery, Anesthesia & Experimental Techniques in Swine. Ames: Iowa State University Press.
Table€17.10╅Dose Rates for Premedication in Pigs Drug Atropine Glycopyrrolate Acepromazine Diazepam Midazolam Azaperone
Dose 0.05 mg/kg IM or 0.02 mg/kg IV 0.004–0.01 mg/kg IM 0.1–0.2 mg/kg IM, SC, or IV 0.5–1 mg/kg IM or 0.44–2 mg/kg IV, 1 mg/kg/h IV infusion 0.1–0.5 mg/kg IM or IV, 0.6–1.5 mg/kg/h IV infusion 2–5 mg/kg IM
Source: Data from Swindle, M. 1998. Surgery, Anesthesia & Experimental Techniques in Swine. Ames: Iowa State University Press, 1998.
Table€17.11╅Dose Rates for Injectable Anesthetics in Swine Drug Ketamine Ketamine + acepromazine Ketamine + diazepam Ketamine + azaperone Ketamine + xylazine Ketamine + medetomidine Ketamine + midazolam Fentanyl + isoflurane Sufentanil + isoflurane Pancuronium Vercuronium
Dose 11–33 mg/kg IM, 3–10 mg/kg/h IV infusion 25–30 mg/kg + 1.1 mg/kg IM 15 mg/kg + 2 mg/kg IM 15 mg/kg + 2 mg/kg IM 20 mg/kg + 2 mg/kg IM 10 mg/kg + 0.05–0.1 mg/kg IM 33 mg/kg IV bolus, 8–33 mg/kg/h IV + 0.5–1.5 mg/kg/h IV infusion 0.03–0.1 mg/kg/h IV infusion + 0.050 mg/kg IV bolus + 0.5% isoflurane 0.015–0.030 mg/kg/h IV infusion + 0.007 mg/kg IV bolus + 0.5% isoflurane 0.02–0.15 mg/kg IV, 0.003–0.030 mg/kg/h IV infusion 1 mg/kg IV
Source: Data from Swindle, M. 1998. Surgery, Anesthesia & Experimental Techniques in Swine. Ames: Iowa State University Press.
84550.indb 518
11/11/10 6:28:13 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
519
Table€17.12╅Dose Rates for Analgesics in Sheep Drug Acetylsalicylic acid Carprofen Buprenorphine Butorphanol + diazepam Xylazine
Dose 50–100 mg/kg by mouth; 6–12 hourly 1.5–2.0 mg/kg SC or IV; daily 0.005–0.01 mg/kg IM or IV; 4 hourly 0.05–0.1 mg/kg IV + 0.05–0.20 mg/kg IV 0.08–0.10 mg/kg IM, 0.03–0.04 mg/kg IV
Source: Data from Dobromylskyj, P. et al. 2000. In Pain Management in Animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders.
Analgesia and Anesthesia in Small Ruminants Analgesia NSAIDs can be used safely and efficiently in ruminants (Nolan 2000). Due to rapid metabolism, opioids have a short duration of action in sheep. Butorphanol in combination with diazepam is useful for sedation and pain relief of small ruminants. Continuous infusion of alpha-2 agonists can provide prolonged analgesia, but severe hypoxia may develop (Grant, Summersides, and Kuchel 2001). Anesthesia Ruminants have a large capacity to redistribute anesthetic drugs in their large gastrointestinal tract (Dunlop and Hoyt 1997). Thus, thiopental has a shorter duration of action in ruminant than in nonruminant species. Unlike goats, sheep are not easily stressed by handling, show no excitement during recovery from anesthesia, and do not need routine premedication. Preadministration of xylazine increases the incidence of regurgitation and should be avoided before endotracheal intubation. Diazepam or local anesthesia may be used to prevent excitement during recovery in goats. Xylazine and medetomidine are sedatives that can be used in combination with local anesthesia for surgical procedures and followed by reversal with atipamezole. Xylazine reduces fetal oxygenation and can cause premature parturition if used during the last trimester. Midazolam can
Table€17.13╅Dose Rates for Premedication, Anesthetics, and Antidotes in Small Ruminants Drug Diazepam Xylazine Medetomidine Atipamezole Midazolam Flumazenil Thiopental Propofol Ketamine + xylazine Ketamine + diazepam
Dose 0.2–0.4 (sheep) or 5–10 (goats) mg/kg IV 0.02–0.15 mg/kg IV 0.003–0.01 mg/kg IV 0.02–0.04 mg/kg IV 0.4–1.0 mg/kg IV 1 mg total dose IV in sheep 10–15 mg/kg IV 4–6 mg/kg IV for induction, then 0.4–0.6 mg/kg/min 2.0–7.5 + 0.1–0.2 mg/kg IV 2.0–5.0 + 0.5–1.0 mg/kg IV
Source: Data from Dunlop, C. I., and R. F. Hoyt. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
84550.indb 519
11/11/10 6:28:14 PM
520
Handbook of Laboratory Animal Science
also be used for sedation and is reversed with flumazenil. Ketamine is commonly combined with xylazine or diazepam for 15–30 min of general anesthesia in small ruminants. Intravenous thiopental injection produces 15 min of anesthesia and is used to allow intubation, followed by inhalation anesthesia. Induction with propofol IV followed by continuous infusion may be used for long-term anesthesia. Additional analgesia is necessary if surgery is undertaken. Face mask induction of halothane or isoflurane anesthesia can easily be accomplished after sedation. This method necessitates efficient gas scavenging. During maintenance with inhalation anesthesia, a cuffed tube should be used to prevent ruminal content or the large amounts of saliva that are produced from reaching the airways. Analgesia and Anesthesia in Cats, Dogs, and Ferrets Analgesia Dogs are particularly sensitive to the adverse renal effects of NSAIDs during hypotension, and cats have an increased susceptibility to the toxic effects of acetylsalicylic acid and other NSAIDs (Dobromylskyj et al. 2000). Toxicity is avoided by carefully selecting a drug and keeping correct dosing intervals. Morphine is effective for the control of severe postoperative pain. Buprenorphine is less effective but has a longer duration of action (6–8 h) and may be used for at least 24 h in cats and 48 h in dogs without side effects. Low doses of ketamine can be used for short-lasting analgesia postoperatively; it has been shown to prevent hyperalgesia in dogs, but side effects such as increased muscle activity may limit its use. Anesthesia Premedication Anticholinergics are used to prevent bradycardia or increased salivation. Commonly used sedatives and tranquillizers are acepromazine, diazepam, and midazolam (Lukasik 1999). Opioids used for premedication are morphine, butorphanol, and buprenorphine, usually in combination with a tranquilizer. They increase sedation and provide pre-, intraoperative and, sometimes, also postoperative analgesia. Preoperative drug combinations used in young healthy dogs and cats are, for instance, acepromazine + butorphanol or acepromazine + buprenorphine. Pediatric, geriatric, and physiologically compromised animals are best premedicated with a benzodiazepine and butorphanol or buprenorphine. Table€17.14â•…Dose Rates for Analgesics in Cats, Dogs, and Ferrets Drug Acetylsalicylic acid Carprofen Ketoprofen
Morphine Buprenorphine
Dog 10–25 mg/kg orally; 8–12€hourly 4 mg/kg SC or IV 2 mg/kg SC daily for up to 3€days; 1 mg/kg orally daily for up to 5 days 0.1–1 mg/kg IM, SC, or IV; 4–6 hourly 0.005–0.020 mg/kg IM, SC, or IV; 6–12 hourly
Cat
Ferret
10–25 mg/kg orally; 48€hourly 2–4 mg/kg SC or IV
20 mg/kg orally
0.1–0.2 mg/kg IM, SC or IV; 6–8 hourly
0.5–2.0 mg/kg IM, SC; 4–6€hourly 0.01–0.03 mg/kg IM, SC, or IV; 6–12 hourly
— —
Sources: Data from Dobromylskyj, P. et al. 2000. In Pain Management in Animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders; and Harvey, R. C. et al. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
84550.indb 520
11/11/10 6:28:14 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
521
Table€17.15╅Dose Rates for Premedication in Cats, Dogs, and Ferrets Drug Atropine Acepromazine Diazepam Midazolam Morphine Buprenorphine Butorphanol Acepromazine + buprenorphine Midazolam + buprenorphine Xylazine Medetomidine
Dog
Cat
0.02–0.04 mg/kg SC or IM or 0.01–0.02 mg/kg IV 0.02–0.075 mg/kg SC or IM, total maximum dose: 3 mg 0.1–0.6 mg/kg IV 0.1–0.5 mg/kg IV 0.1–0.5 mg/kg IM or IV 0.1–2 mg/kg SC, IM or IV 0.05–0.1 mg/kg SC or IM 0.01–0.02 mg/kg SC, IM, 0.005–0.02 mg/kg SC or or IV IM 0.2–0.8 mg/kg SC, IM or IV 0.2–0.4 mg/kg SC, IM, or IV 0.4 + 0.05 SC or IM
Ferret 0.05 mg/kg SC or IM 0.2 mg/kg SC or IM 1–2 mg/kg IM 1 mg/kg IM
0.1 + 0.01 mg/kg SC, IM or IV 0.2–1.1 mg/kg SC, IM or IV 0.01–0.04 mg/kg SC, 0.04–0.08 mg/kg SC, IM€or€IV IM€or€IV
1 mg/kg IM or SC 0.1 mg/kg SC
Sources: Data from Lukasik, V. M. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA; and Harvey, R. C. et al. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
Premedication with carprofen in dogs has no effect on the concentration of isoflurane needed. Butorphanol alone or in combination with carprofen reduces the isoflurane concentration needed by approximately 20% (Ko, Lang, et al. 2000). Alpha-2 agonists (e.g., medetomidine) reduce the dose of subsequent inhalation and injection anesthetics by 50% (Lukasik 1999). Adding butorphanol or ketamine to medetomidine in dogs enhances sedation, but blood gas levels are significantly poorer than when medetomidine is given alone (Ko, Fox, and Mandsager 2000). Inhalation Anesthesia Inhalation anesthesia is the first choice in terms of safety, especially for long-duration anesthesia. All volatiles are safe to use in cats, dogs, and ferrets. Mask induction is possible, but induction by injection is preferred, followed by endotracheal intubation and inhalation anesthesia. Ferrets may be induced in a chamber and maintained on a mask or via intubation. By adding 50% N2O, the required volatile concentration is reduced by approximately 20%, which improves cardiovascular function, especially when volatile agents are the sole anesthetics used. Nitrous oxide also speeds up induction, which is beneficial for mask induction. When N2O is used as part of the anesthetic regimen, it is important to let the animal breathe pure oxygen for 5–10 min after discontinuation of N2O to prevent diffusion hypoxia. Injection Anesthesia Thiopentone can be used as an IV induction agent, given to effect. The recommended dose should be reduced by 50–75% after premedication with an alpha-2 agonist such as medetomidine, and by up to 50% after other premedication. Methohexital may be used in the same way and is suitable, but not preferred, for maintenance of anesthesia by continuous infusion. Alphaxalone + alphadolone (Saffan) has a wide margin of safety and is useful for IV induction and maintenance of anesthesia in the cat. Ketamine + midazolam (cat, dog), ketamine + diazepam (dog, ferret), ketamine + xylazine (ferret), and ketamine + medetomidine (cat, dog, ferret) are all useful combinations for short surgical procedures.
84550.indb 521
11/11/10 6:28:14 PM
522
Handbook of Laboratory Animal Science
Xylazine + butorphanol + ketamine has been described as an effective anesthetic combination in ferrets; however, hypoxemia and ventricular arrhythmias were observed (Ko et al. 1998). Propofol is useful for induction and maintenance by IV infusion in cats and dogs. In contrast to thiopental and volatile agents, propofol causes little reduction in renal blood flow and is especially useful in geriatric animals. Premedication with medetomidine reduces the propofol dose by approximately 75%. Heart rate is significantly lower, but recovery is smoother with medetomidine + propofol compared to medetomidine + ketamine in dogs (Hellebrekers et al. 1998). Sedation with medetomidine followed by intravenous fentanyl only is unsuitable for maintenance of surgical anesthesia in spontaneously breathing dogs due to severe respiratory depression (Hellebrekers et al. 1997). The combination of medetomidine with propofol or ketamine is not a respiratory depressive and may be used without mechanical ventilation. Analgesia and Anesthesia in Primates When methods for anesthesia are being selected, the diversity of nonhuman primates used for research must be considered (Popilskis and Kohn 1997). The wide range in body size plays a role when anesthetics and delivery method are chosen. Considerable species differences in the reaction to certain drugs make simple extrapolation of dosages dangerous. Analgesia Acetylsalicylic acid may be used to treat mild pain. Carprofen and ketoprofen are useful for postoperative pain, as are buprenorphine, oxymorphone, and butorphanol. The benefit of using an NSAID is lack of sedation and cardiopulmonary depression. Anesthesia Halothane, isoflurane, and enflurane are all safe to use in primates (Popilskis and Kohn 1997). Induction of anesthesia is accomplished with a short-acting IV injectable agent, like thiopental or propofol. Some species (e.g., chimpanzees and macaques) can be trained to present an arm or a leg Table€17.16╅Dose Rates for Injectable Anesthesia in Dogs, Cats, and Ferrets Drug Thiopentone Methohexital
Alphaxalone + alphadolone Ketamine + medetomidine Ketamine + diazepam Ketamine + midazolam Propofol
Dog
Cat
10–15 mg/kg IV 5 mg/kg IV + 0.3 mg/kg/ min IV infusion of 1–2.5% solution 6 mg/kg IV + 0.24 mg/kg/ min IV infusion 5.0–7.5 mg/kg + 0.04 mg/ 2.5–7.5 mg/kg + 0.08 mg/ kg IM kg IM or 1.25 mg/kg + 0.04 mg/kg IV 5 mg/kg + 0.25 mg/kg IV
5–6 mg/kg IV + 0.2–0.5 mg/kg/min IV infusion
Ferret — —
8–12 mg/kg IV 8 mg/kg + 0.1 mg/kg IM
25 mg/kg + 2 mg/kg IM
10 + 0.2 IM or 5 + 0.2 IV
—
6–7 mg/kg IV + 0.2–0.5 mg/kg/min IV infusion
—
Sources: Data from Reid, J., and A. M. Nolan. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA; and Harvey, R. C. et al. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
84550.indb 522
11/11/10 6:28:14 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
523
Table€17.17╅Dose Rates for Analgesics in Nonhuman Primates Drug Acetylsalicylic acid Carprofen Ketoprofen Oxymorphone Buprenorphine Butorphanol
Dose 20 mg/kg orally; 6–8 hourly 3–4 mg/kg SC daily for up to 3 days 2 mg/kg SC; daily 0.075–0.15 mg/kg IM; 4–6 hourly 0.005–0.010 mg/kg IM, SC, or IV; 6–12 hourly 0.01 mg/kg IV; 3–4 hourly
Sources: Data from Dobromylskyj, P. et al. 2000. In Pain Management in Animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders; and Popilskis, S. J., and D. F. Kohn. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
for venous access. Others need to be restrained chemically for intravenous access. Ketamine IM is used for this purpose and as part of an anesthetic regime. It has a wide margin of safety, but can produce psychotomimetic emergence reactions; therefore, a sedative (e.g., a benzodiazepine) should be administered as well. Xylazine, medetomidine, and midazolam are commonly used with ketamine, providing muscular relaxation and analgesia sufficient for minor surgical procedures. Tiletamine + zolazepam (Telazol or Zoletil) have the same properties. Alphaxalone + alphadolone (Saffan) can be used for induction by intramuscular injection, followed by intravenous infusion for maintenance of surgical anesthesia. Propofol can be used for IV induction and maintenance. During surgery, analgesia is provided or enhanced by addition of an opioid or a local anesthetic. Intravenous infusion of fentanyl is used for this purpose. Analgesia and Anesthesia—Birds, Fish, Reptiles, and Amphibia Analgesia Birds possess nociceptors and all the opioid receptors, as well as alpha-2-receptors, in the brain. Analgesic treatment after surgery in birds speeds up return to normal food consumption, which is important due to their high rate of metabolism (Forbes 1999). Carprofen, ketoprofen, and Table€17.18â•…Dose Rates for Anesthetics Used in Primates Drug Ketamine Ketamine + xylazine Ketamine + medetomidine Ketamine + midazolam Tiletamine + zolazepam (Telazol) Propofol Thiopental Alphaxalone + alphadolone Fentanyl
Dosea 5–20 mg/kg IM 7–20 mg/kg + 0.25–2 mg/kg IM 2–6 mg/kg + 0.03–0.1 mg/kg IM 15 mg/kg IM+ 0.05–0.09 mg/kg IV 1.5–10 mg/kg IM 1–5 mg/kg IV + 0.13–0.6 mg/kg/min IV infusion 10–17 mg/kg IV 11.5–19 mg/kg IM + 6–12 mg/kg IV 5–10 µg/kg IV + 10–25 µg/kg/h IV + low-dose isoflurane
Source: Data from Popilskis, S. J., and D. F. Kohn. 1997. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. a Doses required vary widely between NHP species.
84550.indb 523
11/11/10 6:28:14 PM
524
Handbook of Laboratory Animal Science
Table€17.19╅Dose Rates for Analgesics in Birds and Reptiles Drug
Birds
Reptiles
Carprofen
1–4 mg/kg IM
Ketoprofen Buprenorphine Butorphanol
2–4 mg/kg IM 0.01–0.05 mg/kg IM 1–4 mg/kg IM; 2–4 hourly
2–4 mg/kg SC, IV, IM, or orally once, then 1–2 mg/kg 24–72 hourly 2 mg/kg SC or IM; every 1–2 days 0.01 mg/kg IM 25 mg/kg IM (tortoises)
Sources: Data from Dobromylskyj, P. et al. 2000. In Pain Management in Animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders; and Malley, D. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA.
butorphanol are safe to use in birds. Fish, reptiles, and amphibia show avoidance reactions to noxious stimuli. Amphibians possess endogenous opioids and show an increased nociceptive threshold after opioid treatment. Review articles on pain and analgesia in birds and other nonmammalian vertebrates are recommended for further reading (Rodger 1999; Bowser 2001). Studies of clinical pain are scarce, but there are dose recommendations for analgesics in birds and reptiles based on experience. Anesthesia Birds Birds show great variation between species in their reactions to analgesics and anesthetics (Forbes 1999). The avian respiratory system is markedly different from that of mammals, with a gas exchange 10 times more efficient. Endotracheal intubation is recommended in birds over 100 g in all but the shortest procedures, to provide intermittent positive pressure ventilation (IPPV). Intubation also prevents aspiration of crop contents. Isoflurane is safe to use and preferred over halothane. Sevoflurane may also be used. Injectable anesthetics are administered by the IM or IV route. For sedation, ketamine or diazepam may be used. Ketamine can be used in combination with diazepam, midazolam, or medetomidine to provide surgical anesthesia for 20–30 min. Fish Administration of anesthetics to fish is usually via a bath solution (Rodger 1999). For maintenance of anesthesia, the gills must be irrigated with anesthetic in water. Tricaine (MS222) was developed for fish anesthesia and is the most common anesthetic. It is available as a powder, which Table€17.20â•…Dose Rates for Anesthetics in Birds Drug Ketamine Diazepam Ketamine + diazepam Ketamine + midazolam Ketamine + medetomidine
Dose 5–30 mg/kg IV or IM 0.5 mg/kg IM 5–20 mg/kg + 1.0–1.5 mg/kg IV or IM 5–20 mg/kg + 0.2 mg/kg IV or IM 3–6 mg/kg + 0.15–0.35 mg/kg IM or IV
Source: Data from Forbes, N. A. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA.
84550.indb 524
11/11/10 6:28:15 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
525
Table€17.21╅Dose Rates for Anesthetics in Reptiles Drug Acepromazine Midazolam Propofol Alphaxalone + alphadolone Ketamine + midazolam Ketamine + medetomidine
Dose and Route 0.1–0.5 mg/kg, IM 2 mg/kg, IM or SC 10 (snakes)–13 (lizards)–15 (cheloniansa) mg/kg, IV 6 (lizards)–9 (chelonians,a snakes) mg/kg, IV 20–40 mg/kg + 2 mg/kg, IM or SC (turtles) 10 mg/kg + 100 (snakes)–150 (cheloniansa) µg/kg, IM
Source: Data from Malley, D. 1999. In Manual of Small Animal Anesthesia and Analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. a Chelonians: tortoises, turtles, and terrapins.
gives an acidic solution in water. Buffering with sodium bicarbonate (200–250 mg/100 mg tricaine) or Tris buffer is necessary, except when seawater is used. Tricaine is more toxic in young fish and in soft warm water, and dosage should be adjusted accordingly. Sedation is achieved with a dose of 15–20 mg/mL and anesthesia with 50–100 mg/mL (Rodger 1999; Bowser 2001). It is important that the fish be kept moist during the whole procedure. For recovery, the fish is placed in a separate tank with water only, before it is returned to its regular tank. Reptiles Premedication with a sedative or an analgesic is useful in reptiles (Malley 1999). Acepromazine, midazolam, carprofen, and ketoprofen may all be used. Analgesics are beneficial if surgery involves the skin, coelomic cavity (chest and abdominal cavity in one), or skeletal system. Because respiratory muscles are relaxed during general anesthesia, artificial ventilation is necessary. It can be achieved by manually compressing the rebreathing bag every 20–30 sec so that the cranial two-fifths of the coelomic cavity is seen to rise. Artificial ventilation must be continued until spontaneous respiration has returned. Propofol or alphaxalone + alphadolone (Saffan) may be used for induction, followed by intubation and administration of oxygen or air by intermittent positive pressure ventilation (IPPV). Ketamine + medetomidine IM may also be used for induction and minor surgery (Greer, Jenne, and Diggs 2001). After completion of surgery, the effects of medetomidine can be reversed with atipamezole. Isoflurane is the most commonly used volatile in reptiles. Mask induction, followed by intubation and artificial ventilation, is useful in lizards. Snakes, chelonians (tortoises, turtles, and terrapins), and large lizards may be intubated while conscious, and anesthesia can be induced by artificial ventilation. Chamber induction is of limited use in reptiles. If 10% of the inspired gas mixture consists of carbon dioxide, the speed of recovery is increased (Malley 1999). Amphibians Most adult amphibians develop and breathe with lungs. Some salamanders, however, are lungless and breathe through the skin and mouth cavity. The skin is permeable and allows amphibians Table€17.22â•…Dose Rates for MS222 in Amphibians Tadpole; Newt 0.2–0.5 g/L
Frog; Salamander
Toad
0.5 g/L
1–3 g/L
Source: Data from Schaeffer, D. O. 1997. Anesthesia and analgesia in nontraditional laboratory animal species. In Anesthesia and Analgesia in Laboratory Animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press.
84550.indb 525
11/11/10 6:28:15 PM
526
Handbook of Laboratory Animal Science
to be anesthetized by immersion methods; tricaine methane sulfonate (MS222) in deionized water is the anesthetic of choice (Schaeffer 1997). When it is used in concentrations over 500 mg/L, buffering with sodium bicarbonate (450–1000 mg/L) or 0.5 M dibasic sodium phosphate (Na2HPO4) is necessary. Using high-concentration unbuffered solutions is stressful due to acidity and prolonged induction time. After the animal is removed from the induction bath, anesthesia is maintained for 5–20 min. Recovery can be accelerated by rinsing the animal with clean deionized water. Inhalation anesthesia in terrestrial amphibians may be induced in a chamber (4–5% halothane or isoflurane) and maintained on lower concentrations and assisted ventilation after intubation. Isoflurane may be bubbled through the water to anesthetize aquatic species. Hypothermia should not be used for invasive procedures because it will not induce adequate anesthesia. Studies indicate that poikilotherm sensory input remains intact at almost freezing temperatures (Schaeffer 1997).
Euthanasia Goals and Definition of Terms in Euthanasia As a consequence of the correct use of terminology, euthanasia refers to the process of inducing death with minimal or no pain or distress in the animal involved. For such humane euthanasia, criteria include a rapid loss of consciousness, a minimum of distress, minimal restraint, and an irreversible and reliable CNS depression, resulting in the death of the animal. Relevant operator-related criteria are operator safety and esthetic aspects of euthanasia. Finally, aspects such as costs and level of complexity of the procedure should be taken into account. As an extra requirement for animals that are part of a research protocol, the potential consequences of the euthanasia technique for the subsequent (histo)pathological or pharmacological analysis should be taken into consideration. Under all circumstances, death must be confirmed on the basis of circulatory and respiratory failure, a lowering of body temperature, and overall body stiffness (rigor mortis). The aims and goals for euthanasia of animals are defined in documents from the veterinary (AVMA 2001) and the laboratory animal (Working party reports 1996, 1997) communities. Methods of Euthanasia The different techniques used for euthanizing animals can be divided into two groups: euthanasia by pharmacological–chemical methods and euthanasia by mechanical–physical methods. Whereas the former usually involves the administration of an overdose of an anesthetic agent and is relatively fast and esthetically acceptable, the latter provides a method that eliminates pharmacological “contamination.” Based upon combined arguments such as species, age and body weight, available technique and expertise, and specific consideration of the further analyses foreseen after euthanasia, the choice between (and within) the different techniques is made. Pharmacological–Chemical Methods With few exceptions, this group involves an overdose of an injectable (e.g., pentobarbital) or an inhalant anesthetic, resulting in loss of consciousness and in the cessation of respiration. Consequently, the heart will stop and death will ensue. Administration of injectable agents in larger species is preferably by the intravenous route. In smaller species, intraperitoneal injection or, after prior sedation or anesthesia, intracardiac injection is employed. Frequently used agents include the barbiturates, of which pentobarbital (100–150 mg/kg) is the most common. Loss of consciousness
84550.indb 526
11/11/10 6:28:15 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
527
following intravenous administration is rapid, quickly followed by respiratory depression and failure. A typical phenomenon seen with the use of barbiturates is enlargement of the spleen, which may interfere with the postmortem evaluation. In larger animals, terminal gasping may be seen following loss of consciousness. This is not encountered with the use of T61®, a specific euthanasia preparation that combines a muscle relaxant with a hypnotic agent and local anesthetic. Although research supports that this preparation induces loss of consciousness before muscle relaxation sets in (Hellebrekers et al. 1990), there are no specific other advantages over pentobarbital except for the esthetic aspect. Although any injectable anesthetic agent administered at a high dose may in principle be used for euthanasia, the only other agent specifically used for euthanasia is potassium chloride (KCl). Because administration is painful, it is only acceptable when performed under general anesthesia. KCl is cardiotoxic, and intravenous or intracardiac administration will result in cardiac failure. As with barbiturates, enlargement of internal organs such as the spleen may be seen in conjunction with KCl usage. Volatile anesthetic agents can be used for inducing euthanasia as well when they are administered at sufficiently high concentrations under ventilatory support. The use of ether is discouraged because it induces excitation and irritation of mucous membranes upon initial administration. Halothane, enflurane, and isoflurane can be used for euthanasia when administered under circumstances of adequate ventilation. Next to the volatiles, the use of carbon dioxide (CO2) has been advocated for euthanasia. Obvious advantages, such as relative safety for the user, limited costs, and availability in compressed gas cylinders, are countered by the discussion on the preferred technique of administering CO2. Lower concentrations (<80% CO2) may result in a relatively slow induction of unconsciousness (Raj and Gregory 1990) and be accompanied by histologic lesions in the pulmonary system. High concentrations of CO2 (i.e., hypoxic mixtures) have been reported to cause distress, whereas establishing normoxia prevented this distress (Coenen et al. 1995). In contrast to these findings, it has been reported that signs of distress could not be documented if CO2 administration takes place in the animal’s home cage (Hackbarth, Küppers, and Bohnet 2000). The relative resistance of neonates and fetuses to high concentrations of CO2 renders this method less suitable for these age groups. Physical Methods Physical methods are indicated when histologic or pharmacological analysis is foreseen following euthanasia; they are primarily employed in small laboratory animal species. Methods include decapitation and cervical dislocation, which can be used in mice, rats, and small rabbits. For decapitation, a special apparatus (guillotine) is available, although, with certain limitations to the animal’s weight, sharp scissors with adequately long blades may be used. Cervical dislocation involves the severing of the spinal cord, whereby the connection between the brain and the vital organs is disrupted. Following cervical dislocation, it is advisable to exsanguinate the animal in order to confirm death. Although both of these methods may be esthetically displeasing, they are acceptable when applied properly and are without the “contamination” involved with pharmacological methods. Other physical methods that may be used under specific conditions and following necessary precautions include stunning, maceration, freezing in liquid nitrogen, and microwave irradiation (Working party report 1996). Special Considerations with Regard to Scientific Purpose When the most optimal method of euthanasia is being determined, the specific method-related changes need to be taken into consideration. Barbiturate-induced splenic enlargement and
84550.indb 527
11/11/10 6:28:15 PM
528
Handbook of Laboratory Animal Science
hypoxemia-induced histological changes due to hyperextension of the lungs (high CO2 concentrations) are potentially disrupting to proper histological analysis. Inflammatory changes in the peritoneum following injection of irritating substances or pharmacochemical interaction with postmortem analysis need to be taken into consideration, as does the extent of normal hepatic metabolism of injectable as well as volatile (halothane > enflurane > isoflurane) agents. Euthanasia of Different Species Euthanasia of Rodents and Rabbits For the euthanasia of rodents and rabbits, most of the physical and pharmacological techniques described earlier are applicable. Body weight limitations apply to physical methods such as decapitation and cervical dislocation, and specific expertise is needed in certain species, such as the hamster and the guinea pig, where the short neck and heavy muscles make it more difficult to assert a proper technique. Cervical dislocation is commonly used in small rodents (<200 g) and rabbits, resulting in a quick death due to physical damage to the brain stem. Depending on the expertise and strength of the operator, the use of this method should be limited to animals weighing less than 1 kg. For a full description of the technique, the reader is referred to the appropriate literature. Decapitation can easily be performed in rats and mice, but should be limited to the smaller representatives of species such as rats and rabbits (<1 kg). Although special equipment is available, sharp scissors with suitably long blades may be used in smaller animals. Larger animals may need sedation to facilitate handling. Other physical methods (concussion, captive bolt in rabbits) are only conditionally acceptable depending on the specific circumstances and expertise of the operator. Euthanasia by injection can be achieved with the use of pentobarbital or T61, whereby the use of the latter is limited to intravenous administration. Although intraperitoneal injection is more feasible in small rodents, intravenous administration is preferred in rabbits. The use of KCl is only acceptable when applied under full anesthesia. Inhalant anesthetics such as halothane, enflurane, and isoflurane can be used to induce and deepen anesthesia in rodents. Rabbits are distressed from induction with inhalants. (See “Analgesia and Anesthesia in Rabbits” section.) The use of ether is discouraged because it induces excitation and irritation of the mucous membranes. Whereas the use of CO is not advised based upon the health hazard for the operator, CO2 can be used to anesthetize and subsequently euthanize rodents but not rabbits. Despite numerous efforts and ongoing discussion, there is no clear consensus as to the preferred method for applying CO2 for euthanasia. The discussion focuses on the (initially applied) concentration of CO2 to which the animals are exposed. Exposure to high concentrations (>80%) induces fast onset of loss of consciousness, with some authors reporting undue (hypoxemiarelated) stress. Only CO2 obtained from compressed gas cylinders should be used for this purpose. Euthanasia of Small Ruminants and Pigs The choice of the most preferred method of euthanasia depends on the size of the animal. Whereas physical methods, such as concussion, may be applied under strict conditions to render young animals unconscious, adequate efficacy cannot be assured in large animals. Electrical stunning can be used as a first step toward euthanasia, when it is executed properly. The electrodes should be placed properly and, when applicable, the skin should be shaven to ensure good electrical contact. Under all circumstances, the operator must ensure adequate voltage and current to be applied to induce immediate unconsciousness. Following stunning, the animal must be exsanguinated immediately to ensure death.
84550.indb 528
11/11/10 6:28:15 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
529
Other physical methods, such as penetrating captive bolt or even shooting, should be considered only when other, more acceptable methods have been impossible to use. Euthanasia by injection can be achieved swiftly and reliably by intravenous administration of pentobarbital, but in larger animals, the volume of injection may limit its use. More concentrated formulations of the drug are available in some countries, but with their use, the aspect of operator safety should be taken into consideration. Other injectable agents used for euthanasia such as T61 also have to be administered intravenously. With the use of KCl, a state of general anesthesia is mandatory. Euthanasia by inhalation can be performed with the aid of any of the common volatile anesthetics. Alternatively, CO2 at concentrations of >70% can be employed to induce unconsciousness, as it is done in the process of slaughtering. After the animal is rendered unconscious, it must be exsanguinated immediately to ensure death. Euthanasia of Cats, Dogs, and Ferrets While physical methods have been described for use in these species, their application in general is not recommended. The use of the captive bolt technique or electrocution has been described in dogs, but, in general, prior sedation will be mandatory. Thereby, the potential advantage of eliminating pharmacological contamination is eliminated, making chemical euthanasia methods a better choice. Intravenous injection of pentobarbital and T61 are viable alternatives. Intracardiac administration of pentobarbital may only be performed under full anesthesia. Acceptable methods of euthanasia by inhalation include the use of an overdose of the commonly used volatile anesthetics (halothane, enflurane, and isoflurane). Controversy exists as to the ethical acceptability of using CO2 or CO for euthanasia of carnivores. Euthanasia of Primates The use and consequently the euthanasia of nonhuman primates will only be allowed under the strictest of regulations in most countries. Literature data on primate euthanasia are scarce. Most importantly, measures need to be taken to minimize stress and anxiety prior to euthanasia. This can adequately be achieved by allowing the intervention to take place in the animal’s own surroundings and out of sight or hearing of other primates. Sedation should be administered to facilitate handling and the further administration of drugs (see section on anesthesia and analgesia of primates). As a general rule, an overdose of a general anesthetic or hypnotic, such as pentobarbital, is considered the only acceptable method of euthanasia in primates (Working party report 1997). Euthanasia of Birds, Fish, Reptiles, and Amphibians Physical methods such as concussion, decapitation, and cervical dislocation in birds of <3 kg and maceration of young chicks (<72 h) have been described and are considered to be ethically acceptable under proper conditions. Alternatively, inhalation of commonly used volatile anesthetics, CO2 or CO, can be employed, whereby the latter alternative involves clear operator health hazards. Intraperitoneal injection of pentobarbital results in rapid death; alternatively, T61 may be injected intramuscularly in the pectoral muscle of small birds. Fish and amphibians can be euthanized by a blow to the head followed by immediate destruction of the brain. Cervical dislocation of small fish may be performed but must also be followed by destruction of the brain. Chemical means of euthanizing fish or amphibians include the use of the
84550.indb 529
11/11/10 6:28:15 PM
530
Handbook of Laboratory Animal Science
commonly used volatile anesthetics, CO2 and pH-corrected tricaine methane sulfonate (MS-222), or benzocaine dissolved in water in which the animal is placed. When single animals are to be euthanized, injections by the intravenous or intraperitoneal routes (fish) or in the dorsal lymph sac (frog) with pentobarbital or T61 may be preferred. The major concern in euthanizing reptiles involves the resistance of their brain to hypoxic conditions. It has been shown that the severed head can respond to stimuli for some time following decapitation. Therefore, proper techniques for euthanasia other than by chemical methods should involve destruction of the brain, as with the captive bolt or shooting. Chemical methods include the administration of pentobarbital intravenously or intraperitoneally.
References ACLAM Task Force. 2007. Guidelines for the assessment and management of pain in rodents and rabbits. Laboratory Animal Science 46:97–108. Albrecht, D., and H. Davidowa, H. 1989. Action of urethane on dorsal lateral geniculate neurons. Brain Research Bulletin 22:923. Andersen, H. E., R. T. Fosse, K. K. Kuiper, J. E. Nordrehaug, and R. J. Pettersen. 1998. Ketorolac (Toradol) as an analgesic in swine following transluminal coronary angioplasty. Laboratory Animals 32:307. AVMA. 2001. Report of the AVMA Panel on euthanasia. Journal of AVMA 218:669. Baamonde, A., M. Alvarez-Vega, A. Hidalgo, and L. Menendez. 2000. Effects of intraplantar morphine in the mouse formalin test. Japan Journal of Pharmacology 83:154. Balaban, R. S., and V. A. Hampshire. 2001. Challenges in small animal noninvasive imaging. ILAR Journal 42:248. Berul, C. I., M. J. Aronovitz, P. J. Wang, and M. E. Mendelsohn. 1996. In vivo cardiac electrophysiology studies in the mouse. Circulation 94:2641. Beyers, T. M., J. A. Richardson, and M. D. Prince. 1991. Axonal degeneration and self-mutilation as a complication of the intramuscular use of ketamine and xylazine in rabbits. Laboratory Animal Science 41:519. Boschert, K., P. A. Flecknell, R. T. Fosse, T. Framstad, M. Ganter, U. Sjostrand, J. Stevens, and J. Thurman. 1994. Ketamine and its use in the pig. Recommendations of the Consensus Meeting on Ketamine Anesthesia in Pigs, Bergen. Bowser, P. R. 2001. Anesthetic options for fish. In Recent advances in veterinary anesthesia and analgesia: Companion animals, ed. R. D. Gleed. and J. W. Ludders. International Veterinary Information Service (www.ivis.org). Brammer, A., C. D. West, and S. L. Allen. 1993. A comparison of propofol with other injectable anesthetics in a rat model for measuring cardiovascular parameters. Lab Anim 27:250. Brunson, D. B. 1997. Pharmacology of inhalation anesthetics. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Chaves, A. A., D. M. Weinstein, and J. A. Bauer. 2001. Noninvasive echocardiographic studies in mice: Influence of anesthetic regimen. Life Sciences 69:213. Claassen, V. 1994. Neglected factors in pharmacology and neuroscience research, vol. 12, in Techniques in the behavioral and neural sciences, ed. J. P. Huston. New York: Elsevier. Clark, J. A., Jr., P. H. Myers, M. F. Goelz, J. E. Thigpen, and D. B. Forsythe. 1997. Pica behavior associated with buprenorphine administration in the rat. Laboratory Animal Science 47:300. Clutton, R. E. 1999. Anesthetic equipment. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Gloucestershire, UK: BSAVA. Coenen, A. M. L., W. H. I. M. Drinkenburg, R. Hoenderken, and E. L. J. M. van Luijtelaar. 1995. Carbon dioxide euthanasia in rats: Oxygen supplementation minimizes signs of agitation and asphyxia. Lab Anim 29:262. Cruz, J. I., J. M. Loste, and O. H. Burzaco. 1998. Observations on the use of medetomidine/ketamine and its reversal with atipamezole for chemical restraint in the mouse. Laboratory Animals 32:18. Dalkara, T., K. Irikura, Z. Huang, N. Panahian, and M. A. Moskowitz. 1995. Cerebrovascular responses under controlled and monitored physiological conditions in the anesthetized mouse. Journal of Cerebral Blood Flow Metabolism 15:631.
84550.indb 530
11/11/10 6:28:16 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
531
Dobromylskyj, P., P. A. Flecknell, B. D. Lascelles, P. J. Pascoe, P. Taylor, and A. Waterman-Pearson. 2000. Management of postoperative and other acute pain. In Pain management in animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders. Dunlop, C. I., and R. F. Hoyt. 1997. Anesthesia and analgesia in ruminants. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Eger, R. P., and B. A. MacLeod. 1995. Anesthesia by intravenous emulsified isoflurane in mice. Canadian Journal of Anesthesia 42:173. Flecknell, P. A. 1999. Anesthesia and analgesia for exotic species—Rabbits, rodents and ferrets. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Flecknell, P., J. V. Roughan, and P. Hedenqvist. 1999. Induction of anesthesia with sevoflurane and isoflurane in the rabbit. Lab Anim 33:41. Flecknell, P. A., J. V. Roughan, and R. Stewart. 1999. Use of oral buprenorphine (“buprenorphine jello”) for postoperative analgesia in rats—A clinical trial. Lab Anim 33:169. Forbes, N. A. 1999. Anesthesia and analgesia for exotic species—Birds. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Gades, N. M., P. J. Danneman, S. K. Wixson, and E. A. Tolley. 2000. The magnitude and duration of the analgesic effect of morphine, butorphanol, and buprenorphine in rats and mice. Contemporary Topics in Laboratory Animal Science 39:8. Glen, J. B. 1980. Animal studies of the anesthetic activity of ICI 35 868. British Journal of Anesthesia 52:731. Goldkuhl, R., J. Hau, and K. S. Abelson. 2010. effects of voluntarily ingested buprenorphine on plasma corticosterone levels, body weight, water intake, and behavior in permanently catheterized rats. In Vivo 24:31–35. Grant, C., G. E. Summersides, and T. R. Kuchel. 2001. A xylazine infusion regimen to provide analgesia in sheep. Lab Anim 35:277. Green, C. J., J. Knight, S. Precious, and S. Simpkin. 1981. Metomidate, etomidate and fentanyl as injectable anesthetic agents in mice. Lab Anim 15:171. Greer, L. L., K. J. Jenne, and H. E. Diggs. 2001. Medetomidine-ketamine anesthesia in red-eared slider turtles (Trachemys scripta elegans). Contemporary Topics in Laboratory Animal Science 40:9. Grint, N. J., and P. J. Murison. 2008. A comparison of ketamine-midazolam and ketamine-medetomidine combinations for induction of anesthesia in rabbits. Veterinary Anesthesia and Analgesia 35:113. Hackbarth, H., N. Küppers, and W. Bohnet. 2000. Euthanasia of rats with carbon dioxide—Animal welfare aspects. Lab Anim 34:91. Hanna, R. M., R. E. Borchard, and S. L. Schmidt. 1988. Pharmacokinetics of ketamine HCL and metabolite I in the cat: A comparison of IV, IM, and rectal administration. Journal of Veterinary Pharmacology and Therapeutics 11:84. Harvey, R. C., R. R. Paddleford, S. J. Popilskis, and S. K. Wixson. 1997. Anesthesia and analgesia in dogs, cats, and ferrets. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Hayes, K. E., J. A. Raucci, N. M. Gades, and L. A. Toth. 2000. An evaluation of analgesic regimens for abdominal surgery in mice. Contemporary Topics in Laboratory Animal Science 39:18. Heavner, J. E. 1997. Pharmacology of analgesics. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Hedenqvist, P., J. V. Roughan, L. Antunes, H. Orr, and P. A. Flecknell. 2001. Induction of anesthesia with desflurane and isoflurane in the rabbit. Laboratory Animals 35:172. Hedenqvist, P., J. V. Roughan, and P. A. Flecknell. 2000a. Effects of repeated anesthesia with ketamine/ medetomidine and of preanesthetic administration of buprenorphine in rats. Lab Anim 34:207. ———. 2000b. Sufentanil and medetomidine anesthesia in the rat and its reversal with atipamezole and butorphanol. Lab Anim 34:244. Hedenqvist, P., J. V. Roughan, H. E. Orr, and L. M. Antunes. 2001. Assessment of ketamine/medetomidine anesthesia in the New Zealand white rabbit. Veterinary Anesthesia and Analgesia 28:18. Hellebrekers, L. J. 2000. Pathophysiology of pain in animals and its consequences for analgesic therapy. In Animal pain, a practice-oriented approach to an effective pain control in animals, ed. L. J. Hellebrekers. Utrecht, the Netherlands: Van der Wees Publishers.
84550.indb 531
11/11/10 6:28:16 PM
532
Handbook of Laboratory Animal Science
Hellebrekers, L. J., V. Baumans, A. P. Bertens, and W. Hartman. 1990. On the use of T61 for euthanasia of domestic and laboratory animals, an ethical evaluation. Lab Anim 24:200. Hellebrekers, L. J., E. J. de Boer, M. A. van Zuylen, and H. Vosmeer. 1997. A comparison between medetomidine-ketamine and medetomidine-propofol anaesthesia in rabbits. Lab Anim 31:58. Hellebrekers, L. J., and R. Sap. 1997. Medetomidine as a premedicant for ketamine, propofol or fentanyl anesthesia in dogs. Veterinary Record 140:545. Hellebrekers, L. J., H. van Herpen, J. F. Hird, C. U. Rosenhagen, R. Sap, and O. Vainio. 1998. Clinical efficacy and safety of propofol or ketamine anesthesia in dogs premedicated with medetomidine. Veterinary Record 142:31. Henke, J., S. Astner, T. Brill, B. Eissner, R. Busch, and W. Erhardt. 2005. Comparative study of three intramuscular anesthetic combinations in rabbits.Veterinary Anesthesia and Analgesia 32:61. Hildebrand, S. V. 1997. Paralytic agents. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Hu, C., P. A. Flecknell, and J. H. Liles. 1992. Fentanyl and medetomidine anaesthesia in the rat and its reversal using atipamazole and either nalbuphine or butorphanol. Lab Anim 26:15. Husby, P., J. K. Heltne, M. E. Koller, S. Birkeland, J. Westby, R. Fosse, and T. Lund. 1998. Midazolamfentanyl-isoflurane anaesthesia is suitable for haemodynamic and fluid balance studies in pigs. Lab Anim 32:316. Jablonski, P., B. O. Howden, and K. Baxter, K. 2001. Influence of buprenorphine analgesia on postoperative recovery in two strains of rats. Lab Anim 35:213. Jacobson, C. 2000. Adverse effects on growth rates in rats caused by buprenorphine administration. Lab Anim 34:202. Jones, R. S. 1999. Neuromuscular blockade. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Kaiser, G. M., N. R. Frühauf, H. Zhang, S. Westermann, I. Bolle, K. J. Oldhafer, and C. E. Broelsch. 2003. Intravenous infusion anesthesia with propofol-midazolam-fentanyl for experimental surgery in swine. Journal of Investive Surgery 16 (6): 353–357. Kalliokoski, O., K. R. Jacobsen, J. Hau, and K. S. Abelson. In press. Serum concentrations of buprenorphine after oral and parenteral administration in male mice. Veterinary Journal. Ko, J. C., S. M. Fox, and R. E. Mandsager. 2000. Sedative and cardiorespiratory effects of medetomidine, medetomidine-butorphanol, and medetomidine-ketamine in dogs. Journal of the American Veterinary Medical Association 216:1578. Ko, J. C., D. N. Lange, R. E. Mandsager, M. E. Payton, C. Bowen, A. Kamata, and W. C. Kuo. 2000. Effects of butorphanol and carprofen on the minimal alveolar concentration of isoflurane in dogs. Journal of the American Veterinary Medical Association 217:025. Ko, J. C., T. A. Smith, W. C. Kuo, and C. F. Nicklin. 1998. Comparison of anesthetic and cardiorespiratory effects of diazepam-butorphanol-ketamine, acepromazine-butorphanol-ketamine, and xylazine-butorphanolketamine in ferrets. Journal of the American Hospital Association 34:40798. Kolesnikov, Y., I. Chereshnev, and G. W. Pasternak. 2000. Analgesic synergy between topical lidocaine and topical opioids. Journal of Pharmacology and Experimental Therapeutics 295:546. Lascelles, B. D., P. J. Cripps, A. Jones, and A. E. Waterman. 1997. Postoperative central hypersensitivity and pain: The preemptive value of pethidine for ovariohysterectomy. Pain (73) 3: 461–471. Lascelles, B. D., A. E. Waterman, P. J. Cripps, A. Livingston, and G. Henderson. 1995. Central sensitization as a result of surgical pain: Investigation of the preemptive value of pethidine for ovariohysterectomy in the rat. Pain 62:201. Lascelles, B. D. X. 2000. Clinical pharmacology of analgesic agents. In Animal pain, a practice-oriented approach to an effective pain control in animals, ed. L. J. Hellebrekers. Utrecht, the Netherlands: Van der Wees Publishers. Liles, J. H., and P. A. Flecknell. 1992. The effects of buprenorphine, nalbuphine and butorphanol alone or following halothane anaesthesia on food and water consumption and locomotor movement in rats. Lab Anim 26:180. ———. 1993. The influence of buprenorphine or bupivacaine on the postoperative effects of laparotomy and bile-duct ligation in rats. Lab Anim 27:374. ———. 1994. A comparison of the effects of buprenorphine, carprofen and flunixin following laparotomy in rats. Journal of Veterinary Pharmacology and Therapeutics 17:284.
84550.indb 532
11/11/10 6:28:16 PM
Laboratory Animal Analgesia, Anesthesia, and Euthanasia
533
Lipman, N. S., R. P. Marini, and P. A. Flecknell. 1997. Anesthesia and analgesia in rabbits. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Ludders, J. W. 1999. Inhalant anesthetics. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Lukasik, V. M. 1999. Premedication and sedation. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Malley, D. 1999. Anesthesia and analgesia for exotic species—Reptiles. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Martinez, M. A., P. J. Murison, and E. Love. 2009. Induction of anesthesia with either midazolam or propofol in rabbits premedicated with fentanyl/fluanisone. Veterinary Record 164:803. Morton, D. B., M. Jennings, A. Buckwell, R. Ewbank, C. Godfrey, B. Holgate, I. Inglis, et al. 2001. Refining procedures for the administration of substances. Lab Anim 35:1. Nolan, A. M. 2000. Pharmacology of analgesic drugs. In Pain management in animals, ed. P. Flecknell and A. Waterman-Pearson. Philadelphia, PA: W. B. Saunders. Olson, M. E., D. Vizzutti, D. W. Morck, and A. K. Cox. 1994. The parasympatholytic effects of atropine sulfate and glycopyrrolate in rats and rabbits. Canadian Journal of Veterinary Research 58 (4): 254. Papaioannou, V. E., and J. G. Fox. 1993. Efficacy of tribromoethanol anesthesia in mice. Laboratory Animal Science 43:189. Popilskis, S. J., and D. F. Kohn. 1997. Anesthesia and analgesia in nonhuman primates. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Raj, A. B. M., and N. G. Gregory. 1990. Effect of rate of induction of carbon dioxide anesthesia on the time of onset of unconsciousness and convulsions. Research in Veterinary Science 49:360. Rao, S., and A. S. Verkman. 2000. Analysis of organ physiology in transgenic mice. American Journal of Physiology—Cell Physiology 279:C1. Reichert, J. A., R. S. Daughters, R. Rivard, and D. A. Simone. 2001. Peripheral and preemptive opioid antinociception in a mouse visceral pain model. Pain 89:221. Reid, J., and A. M. Nolan. 1999. Intravenous anesthetics. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Robertson, S. A., P. M. Taylor, M. Bloomfield, and J. W. Sear. 2001. Buprenorphine disposition after buccal administration in cats: preliminary observations. Proceedings, AVA, Spring 2001. Rodger, H. 1999. Anesthesia and analgesia for exotic species—Fish. In Manual of small animal anesthesia and analgesia, ed. C. Seymour and R. Gleed. Quedgeley, UK: BSAVA. Roughan, J. V., and P. A. Flecknell. 2001. Behavioral effects of laparotomy and analgesic effects of ketoprofen and carprofen in rats. Pain 90:65. Roughan, J. V., O. B. Ojeda, and P. A. Flecknell. 1999. The influence of preanesthetic administration of buprenorphine on the anesthetic effects of ketamine/medetomidine and pentobarbitone in rats and the consequences of repeated anesthesia. Lab Anim 33:234. Schaeffer, D. O. 1997. Anesthesia and analgesia in nontraditional laboratory animal species. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. SchÖffmann, G., P. Winter, R. Palme, A. Pollak, G. Trittenwein, and J. Golej. 2009. Hemodynamic changes and stress response of piglets to surgery. Lab Anim 43:243. Schou, H., L. Kongstad, V. Perez de Sa, O. Werner, and A. Larsson. 1998. Uncompensated blood loss is not tolerated during acute normovolemic hemodilution in anesthetized pigs. Anesthesia and Analgesia 87:786. Schreiber, S., M. M. Backer, J. Yanai, and C. G. Pick. 1996. The antinociceptive effect of fluvoxamine. European Neuropsychopharmacology 6:281. Silverman, J., and W. W. Muir, III. 1993. A review of laboratory animal anesthesia with chloral hydrate and chloralose. Laboratory Animal Science 43:210. Simpson, D. P. 1997. Prolonged (12 hours) intravenous anesthesia in the rat. Laboratory Animal Science 47:519. Skarda, R. T. 1996. Local and regional anesthetic and analgesic techniques: Dogs. In Lumb & Jones’ veterinary anesthesia, 3rd ed., ed. J. C. Thurmon, W. J. Tranquilli, and G. J. Benson. Baltimore, MD: Williams & Wilkins. Skolleborg, K. C., J. E. Gronbech, K. Grong, F. E. Abyholm, and J. Lekven. 1990. Distribution of cardiac output during pentobarbital versus midazolam/fentanyl/fluanisone anesthesia in the rat. Lab Anim 24:221.
84550.indb 533
11/11/10 6:28:16 PM
534
Handbook of Laboratory Animal Science
Smiler, K. L., S. Stein, K. L. Hrapkiewicz, and J. R. Hiben. 1990. Tissue response to intramuscular and intraperitoneal injections of ketamine and xylazine in rats. Laboratory Animal Science 40:60. Steffey, E. P. 1996. Inhalation anesthetics. In Lumb & Jones’ veterinary anesthesia, 3rd ed., ed. J. C. Thurmon, W. J. Tranquilli, and G. J. Benson. Baltimore, MD: Williams & Wilkins. Stenqvist, O. 2000. Inhalationsanestesi. In Anestesi, liber, ed. M. A. B. Halldin and S. G. E. Lindahl. Swindle, M. 1998. Surgery, anesthesia & experimental techniques in swine. Ames: Iowa State University Press. Tanaka, N., N. Dalton, L. Mao, H. A. Rockman, K. L. Peterson, K. R. Gottshall, and J. J. Hunter. 1996. Transthoracic echocardiography in models of cardiac disease in the mouse. Circulation 94:1109. Van Ham, L. M., J. Nijs, D. R. Mattheeuws, and G. G. Vanderstraeten. 1996. Sufentanil and nitrous oxide anesthesia for the recording of transcranial magnetic motor evoked potentials in dogs. Veterinary Record 138:642. Van Woerkens, E. C., A. Trouwborst, D. J. Duncker, M. M. Koning, F. Boomsma, and P. D. Verdouw. 1992. Catecholamines and regional hemodynamics during isovolemic hemodilution in anesthetized pigs. Journal of Applied Physiology 72:760. Wixson, S. K., and K. L. Smiler. 1997. Anesthesia and analgesia in rodents. In Anesthesia and analgesia in laboratory animals, ed. D. F. Kohn, S. K. Wixson, W. J. White, and G. J. Benson. New York: ACLAM and Academic Press. Woolf, C. J., and M. S. Chong. 1993. Preemptive analgesia—Treating postoperative pain by preventing the establishment of central sensitization. Anesthesia and Analgesia 77:372. Working party report. 1996. Recommendations for euthanasia of experimental animals (part 1). Lab Anim 30:293. ———. 1997. Recommendations for euthanasia of experimental animals (part 2). Lab Anim 31:1.
84550.indb 534
11/11/10 6:28:16 PM
Chapter 18
Welfare Assessment and Humane End Points David B. Morton and Jann Hau
Contents Introduction..................................................................................................................................... 535 Recognition and Assessment of Animal Well-Being: Theory........................................................ 537 Animal Suffering............................................................................................................................ 538 Pain ............................................................................................................................................... 538 “Dystress”....................................................................................................................................... 541 Fear ............................................................................................................................................... 542 Mental Distress............................................................................................................................... 542 Recognition and Assessment of Animal Well-Being: Practical Application................................. 543 Husbandry and the Five Freedoms................................................................................................. 548 Freedom from Thirst, Hunger, and Malnutrition....................................................................... 548 Freedom from Discomfort......................................................................................................... 549 Freedom from Pain, Injury, and Disease................................................................................... 549 Freedom to Express Normal Behavior....................................................................................... 550 Freedom from Fear and Distress................................................................................................ 551 Routine Procedures Used in Husbandry and Research.................................................................. 552 Experimental Procedures................................................................................................................ 554 Score Sheets.................................................................................................................................... 555 Establishing a Score Sheet......................................................................................................... 555 Interpreting a Score Sheet.......................................................................................................... 559 Humane End Points........................................................................................................................ 559 Transgenic Animals........................................................................................................................ 561 Humane Killing of Animals........................................................................................................... 563 Concluding Remarks.......................................................................................................................564 References....................................................................................................................................... 565 Relevant Reference Sites on the World Wide Web......................................................................... 572 Introduction There are ethical and legal obligations to treat animals in a humane manner and to ensure high standards with respect to their husbandry and care, resulting in maximizing their well-being under the constraints of their use in science. The use of animals by humans is always associated with responsibility 535
84550.indb 535
11/11/10 6:28:16 PM
536
Handbook of Laboratory Animal Science
for the welfare of the animals concerned. However, when animals are used for research and subjected to procedures that may cause them pain, distress, suffering, and lasting harm, there are additional obligations to refine those techniques so as to cause only the minimum of pain and distress. It is important to separate the pain and distress that may be an unavoidable component of the research from that which is avoidable through, for example, better experimental technique or husbandry. As long ago as 1959, Russell and Burch referred to them as direct (necessary) and contingent (avoidable) inhumanities. They considered contingent inhumanity as being incidental or inadvertent (which may not be quite the same as avoidable), but it is generally agreed that any animal suffering not required to achieve the scientific objective can only confound the experiment being conducted. Many scientists are aware that severely stressed animals are not suitable for valid biomedical research or, as Russell and Burch wrote: For some time it has been obvious to experimental biologists that severe pain, fear or in general distress in experimental animals is an unmitigated nuisance in nearly all kinds of investigationâ•›…â•›the effect of the treatment under study is utterly confused by the effects of the distress, and the animal ceases to be of the slightest use as a model of normal physiological function. While this is not always true (e.g., gaining manual skills under a badly administered anesthetic), maintaining high standards of animals’ well-being is still an ethical, legal, and scientific obligation. More recently, Wurbel and others (Wurbel 2001; Garner 2005, Wurbel and Garner 2007) have criticized the assumption that husbandry and management procedures can be standardized to the point where they can be ignored; it now appears that they are important contributors to data variance and that enrichment may in fact reduce it (Hess et al. 2008; Richter, Garner, and Wurbel 2009).
Welfare can be defined in different ways (Broom and Johnson 1993; Appleby and Hughes 1997; Bekoff and Meaney 1997). Some definitions focus on an animal’s emotional perception of its environment (i.e., the stimuli that emanate from that environment) (Fraser 1995; Duncan 1996). Other definitions focus on how animals cope with their environment and how they have to compensate for adverse changes in their environment by behavioral and physiological adjustments. This latter definition has the advantage in that coping mechanisms may be identified and measured, allowing practical advantageous changes to an animal’s environment to be made, thus improving animal well-being (Broom 1997; Fraser 1995; Fraser and Broom 1997). A third approach is to compare an animal in its captive environment with its counterpart in the wild or in a more natural environment (Dawkins 1992). The terms “well-being” and “welfare” are often used synonymously, but well-being may also be seen as how an animal is faring in the short term (DeGrazia 1996), with welfare being a longer-term average (whole life). “Quality of life” is another concept that is a useful measure under standardized husbandry and management (farming and laboratory animal caging) systems. Retrospective judgments can be made on the overall balance of the negative and positive experiences for an animal even to the point of making a value judgment of whether an animal had a life “not worth living,” a life “worth living,” or even a “good” life (Morton 2007; FAWC 2009). This approach allows for short-term negative experiences to be outweighed by longer term benefits. For example, it may be necessary to reduce an animal’s well-being temporarily while it is being restrained and vaccinated, but its welfare is increased because of a reduced risk of serious disease, so the animal has a longer life with better quality. Note also that animal health can be seen as a subset of welfare and not the other way around. Animals can be perfectly healthy but still have a poor quality of life (e.g., confinement in too small a cage). However, if one looks at welfare as a mental state, then it follows that poor mental health is a welfare problem. Whichever way one sees it, the quality of an animal’s life is determined by a mixture of psychological, physical, and physiological states. However, it can be generally said that an animal’s welfare is compromised when its physiological or physical health or its psychological well-being, in relation to its cognitive capacity, is affected negatively.
84550.indb 536
11/11/10 6:28:17 PM
Welfare Assessment and Humane End Points
537
Some people take the view that causing animals to suffer is a breach of the rights of the animal or a failure of humans to meet their obligations to care for vulnerable or less fortunate beings. Those who do not believe that animals have rights will still require some justification for imposing poor welfare. This justification has to be balanced by the perceived benefits from this use of animals and be in proportion to the harms done. This is why ethical debate is so important as it tries to help rationalize and justify our uses of animals (see Chapter 2, this volume). In many countries, the use of animals in research has to be seriously ethically justified before their use is allowed, and a trivial gain in scientific knowledge may not be enough. Notwithstanding that, any use of animals should cause only the minimum of suffering necessary to achieve a scientific objective; ethical frameworks help in this analysis as they start to evaluate key points in the debate. In this chapter, we deal with the recognition of suffering in animals and how that can be recognized, assessed, and mitigated.
Recognition and Assessment of Animal Well-Being: Theory Unlike humans, who can say when they are in pain or distress or feel uncomfortable, animals cannot. The same applies to human babies, deaf mutes, and others who are conscious but unable to communicate. However, it would be foolish to deny that they cannot experience pain and distress. The observer has therefore to recognize certain signs that indicate when such humans and animals are suffering. This places an obligation on scientists and animal technical staff that they should learn to do this because only when they can recognize the signs of pain and distress in an animal can something be done about it, such as providing pain relief, withdrawing the animal from the study, or humanely killing it. Fortunately, given the variety of animals used in research, we can make certain predictions about whether animals are likely to experience adverse states that hold good for all sentient beings (i.e., all animals that are able to experience pain, pleasure, etc.). A good start is to ask whether a human in that condition would be experiencing pain or distress, but we need to be careful and apply what has been termed a “critical anthropomorphism” (Morton, Burghardt, and Smith 1990; NRC 1996; OECD 2001). Another approach is to use “zoomorphism”—extrapolating between different species of animals (Bock and Webster 2009). This means that one has to take into account the biological characteristics of the species of animal and also, if possible, to have some idea of what an individual animal has experienced in its life. For example, signs of pain and distress will vary according to the species insofar as anatomy differs. Animals that have been raised in the wild or always been in a captive environment may respond differently. Animals that have already been subjected to or conditioned to experimental procedures, such as injections or blood removal—with or without food reward—may behave differently than those that are naïve or unconditioned. Moreover, the experience an animal has had of a particular human being (e.g., the one doing the injection) will undoubtedly influence its behavior and physiological homeostasis. Publications examining the interactions between humans and experimental animals have clearly shown major effects (Davies and Balfour 1992; Hemsworth and Coleman 1998), including increased weight gain in gentled rats compared with rats unaccustomed to humans. These become important issues when an individual animal is handled; of increasing importance, however, is how we can measure animal pain and distress in other situations (e.g., after carrying out a scientific procedure) and whether this can be done reliably. In addition to there being an ethical mandate to avoid causing animal suffering and discomfort during an experiment, it is likely to produce better science (Chance 1956; Riley 1981; Rose 1994; Claassen 1994), as so convincingly presented by the many examples listed by Chance as early as 1957 (Chance 1957). These included the incidence of cancer in mice being related to numbers in a cage, blood eosinophil levels in mice
84550.indb 537
11/11/10 6:28:17 PM
538
Handbook of Laboratory Animal Science
being altered by sounds, crowding having an impact on the susceptibility of rats to tuberculosis, and stress influencing neoplastic events and immunocompetency. Animal Suffering Suffering can be defined as a negative emotional state that derives from adverse physical, physiological, and psychological circumstances, in accordance with the cognitive capacity of the species and of the individual being (Morton, personal communication). Suffering is an emotive word and some believe only humans can truly suffer. In older literature, the expressions “ill at ease” and “uncomfortable” were used to describe what we today might call suffering. An example of this is the situation for dogs kept in cages slightly too small for them, resulting in a disturbed metabolism rendering them useless for “biochemical” studies (Moberg and Mench 2000). Suffering incorporates notions of self-awareness in relation to others and over time. It is still the subject of some debate about whether this is a uniquely human attribute or if animals have this capacity and, if so, to what degree. Increasingly, scientific evidence is accumulating that some of the higher nonhuman primates, such as chimpanzees, gorillas, and baboons, are self-aware, and there is plenty of anecdotal evidence for other species including parrots and dolphins (Dawkins 1993; Denton 1993) and, more recently, the magpie (Prior, Schwarz, and Gunturkün 2008). Self-awareness and self-consciousness add another dimension to the “simple” physical experience of pain and mental distress, such as anticipating adverse outcomes and frustrations of future pleasures such as life plans in humans. In this chapter, we use “suffering” to cover all those feelings an animal might experience that we as humans would normally find unpleasant or aversive in some way, such as pain, mental distress, fear, and boredom. Such suffering can vary in intensity and duration and is difficult to measure and quantify. In many countries, it is intensity that is the primary concern and that is reported in annual statistics, rather than some mathematical approximation of total suffering such as intensity × duration (often referred to as “severity”). The impact of these adverse states on an animal is probably best assessed using clinical signs and behavior and determining how far an animal has deviated from normality, rather then trying to quantify specific physiological responses, such as the level of endorphins or corticosteroids. The latter measurements are useful in a research setting into animal welfare, but when animals are being used in medical research and the like, a holistic approach looking at the impact of an experiment on animals is the most appropriate and most practical. Notwithstanding this holistic approach, it is scientifically worthwhile understanding some of the physiological ways in which animals respond to adverse conditions because they might well affect the data collected. In addition to the immune system, at least four integrated body systems come into play to varying degrees according to the stimulus and what is experienced by an animal (Figure€18.1). As they read this chapter, readers should always have at the backs of their minds what they might feel themselves and how they might respond, and acknowledge that animal research into these adverse mental states has resulted in the development of successful treatments for humans and animals such as analgesic, antidepressive, and anxiolytic drugs. Pain Pain is a highly conserved mechanism that evolved to protect animals from serious injury. The pain-perceiving, or nociceptive, system is common for mammals, probably for birds, and possibly throughout the vertebrates (Table€ 18.1). Potential tissue damage is detected by free nerve endings called nociceptors that respond to excessive heat, pressure, and chemical stimulation. The nociceptors are linked to the dorsal horn of the spinal cord by myelinated A-delta and nonmyelinated C fibers. The
84550.indb 538
11/11/10 6:28:17 PM
Welfare Assessment and Humane End Points
HypothalamicPituitary Adrenal cortex
539
Nociception pain
dystress
Sympathetic Adrenal Medulla
Neurochemical Receptors Mental distress: depression boredom frustration grief happiness
flight fright fight
Figure 18.1â•…Venn diagram showing the various physiological response mechanisms in an animal and how they interact.
Table€18.1╅Comparison between Animals of Some Anatomical Structures and Biological Mechanisms of Importance to Pain Perception Invertebrates Earthworm Nociceptive system Brain structure analogous to human cerebral cortex Nociceptors connected to higher brain structure Opioid-type receptors Response modified by analgesia Response to noxious stimuli persists Associates neural with noxious stimuli
Vertebrates
Insects
Octopus
Fish
Amphibian
Reptiles
Birds
Mammals
?
—
?
?
?
?
+
+
—
—
?
+
+
+
+
+
—
—
?
?
?
?
?
+
+
+
?
+
+
+
+
+
?
?
?
?
?
?
+
+
—
—
+/–
?
?
?
+
+
—
+
+
+
+
+
+
+
Source: Adapted from Smith, J. A., and K. M. Boyd, eds. 1991. Lives in the Balance: The Ethics of Using Animals in Biomedical Research, the Report of a Working Party of the Institute of Medical Ethics. Oxford, England: Oxford University Press. Notes: (+) present, (–) absent, (?) not known, (+/–) some weak evidence.
84550.indb 539
11/11/10 6:28:18 PM
540
Handbook of Laboratory Animal Science
nociceptive information is subsequently transferred to supraspinal nuclei in the thalamus by ascending neurons that cross over to the opposite side of the cord in the dorsal horn. These ascending neurons also synapse with motor efferent nerves in the dorsal horn and, together, they form the spinal reflex. This rapid response (milliseconds) draws the affected body part away from the adverse threat, thus helping to protect the animal from injury. The release of nerve transmitters from the neurons ensures that the message proceeds to the brain and up to the cerebral cortex when the animal perceives this release as the feeling of “pain.” Various neuronal links with other parts of the brain collate this feeling with memory, and the animal then has choices of how to respond, so-called retroduction (Morton 2000a). Thus, the animal decides what course of action to take (e.g., to flee, to fight, to vocalize, etc.). At this point in time, several seconds after the stimulus, the animal is likely to be aware of what it is doing and not responding by reflex or even instinctively. The pain perception is accompanied and is mitigated by the release of endorphins in the brain and spinal cord as they raise the threshold for pain perception. In addition to this release of endorphins in the central nervous system (CNS), there are descending inhibitory fibers that pass down the spinal cord from the brain and synapse with neurons in the dorsal horn; these also raise the threshold for upward transmission to the brain. These descending fibers do not appear to be present at birth in many species (e.g., rodents and humans), even though the ascending nerves are present and active; therefore, young, immature animals of some species may well feel more pain than older animals, in which the nervous system is fully formed and functional (Fitzgerald 1994). In the mammalian fetus, neuroanatomical maturation of the nociceptive system and the final critical corticothalamic connections (for pain perception) take place normally later on in the second half of gestation. However, to experience pain emotionally, an individual must be cognitively aware of the stimulus, and it is most likely that the fetus is actively maintained asleep and unconscious throughout gestation and cannot be woken by nociceptive stimuli (Mellor et al. 2005). In contrast with this view, several states in the United States have considered legislation requiring physicians to inform women seeking abortion that the fetus feels pain and to offer fetal anesthesia (Lee et al. 2005). When dealing with animal experiments involving fetal manipulation, it is not always appropriate to give pain relief because pain treatment may produce greater adverse outcomes than the effects of a transitory nociceptive stimulus; however, more research is clearly needed in this area. Table€18.2 indicates the likelihood of animals’ feeling as measured by the development of the various physiological response mechanisms (Diesch et al. 2007). Table€18.2â•…Stage of Development and Likely Sentience Adverse effect Pain Dystress Mental distress Fear
Zygote
Embryo
Fetus (early)
Fetus (late)a
Perinatal
Preweaning
Prepubertal
Adult
— — —
— — —
— — —
+ ++ +
–/+++ ++ +
+++ +++ +
++ +++ ++
++ +++ +++
—
—
—
+
+
+
++
+++
Notes: Notional scale of absence (—) or presence (+/++/+++) of sense-associated structures. NB for pain, the complete presence in a fully adult sentient form (++) is likely to mean that less pain is felt than in an immature form, when the nervous system’s pain inhibitory pathways are not yet fully developed (hence, +++ cf ++). More recently, data looking at EEG recordings for some species suggest that neonatal animals may not be able to respond to noxious stimuli for several days after birth; however, this is subject to some uncertainty (Diesch, T. et al. 2007.) For dystress, in some species, the system matures with age (++ to +++). For mental distress and fear, an animal will learn from experience as it gets older (++ to +++) and as its system develops (+ to ++). a Depends on whether it has breathed.115
84550.indb 540
11/11/10 6:28:18 PM
Welfare Assessment and Humane End Points
541
There are also nociceptors in the organs of the body that link to the central nervous system by afferent neurons of the autonomic nervous system. These are sensitive to stretch (e.g., the gut) and also ischemia (heart). Visceral pain is less easily diagnosed and controlled than the somatic pain described before. In most experimental work, pain is an unwanted side effect, so it is necessary to relieve it. This is best achieved by exercising what is known as “preemptive analgesia” (see Chapter 17). If pain is not relieved, then neurons adjacent to the activated one in the dorsal horn of the spinal cord will also be sensitized and the painful area will extend laterally; thus, the animal will feel more pain (hyperalgesia). Moreover, stimuli that are normally not painful become painful (allodynia). Pain probably cannot be measured from endorphin levels or nerve activity, but parametric observation of behavior, for instance, may be helpful. Behaviors exhibited will largely depend on the site affected (e.g., immobility, posture when abdominal muscles have been traumatized, lameness after surgery or arthritis, depth of respiration after thoracotomy, vocalizing when the site is pressed) as well as other clinical signs that can be measured, such as eating and drinking, body weight, and temperature. Oral voluntary ingestion ad libitum is an efficacious, convenient, and noninvasive way of administering perioperative buprenorphine to rats, as judged by concentrations of the drug in the circulation, corticosteroid response, and effects on body weight and water consumption (Goldkuhl et al. 2008; Kalliokoski et al. 2009). At the centralized laboratory animal facilities of the University of Copenhagen, all rats subjected to invasive procedures receive buprenorphine mixed in a commercially available nut paste before and after the procedure as the default system (Hau et al. 2009).
“Dystress” When animals are exposed to environmental stressors such as extremes of temperatures or when they perceive a threat by virtue of their senses (e.g., sound, touch, and sight), neurons in the CNS are stimulated; this results in activation of the hypothalamic pituitary adrenal (HPA) axis leading to, among other things, the release of adrenocorticotrophic hormone (ACTH). This in turn will stimulate the cortex of the adrenal gland to synthesize and release corticosteroids that are predominantly either cortisol or corticosterone, depending on the species. Thus, if high levels of corticosteroids are present in the circulation and they persist for long periods of time, despite their short half-lives, it would not be unreasonable to conclude that the animal was not coping and was being subjected to a “bad stress”—hence, the word “dystress” (compare with mental distress discussed later). If the animal does not adapt, then other activities of the hypothalamic axis are affected, and this has an impact on nearly all the homeostatic mechanisms of the body—hence, the potential impact on the scientific objective. Such animals show poor growth rates, even body weight loss, as well as disturbances of the endocrine system to the point where the animal is no longer fit in a biological sense. It may not be able to breed and reproduce and may show considerable pathology; in general, the animal becomes more susceptible to infectious and noninfectious diseases due to suppression of the immune system (Moberg and Mench 2000; Yeates and Main 2007). By and large, dystress is an unconscious phenomenon and can be measured by looking at the level of ACTH and corticosteroids in the circulation and in excretions such as urine and feces, the response of the adrenal gland to exogenous ACTH, and signs of pathology, as well as the functioning of whole body systems such as the endocrine system, the immune system, and the reproductive system.
84550.indb 541
11/11/10 6:28:18 PM
542
Handbook of Laboratory Animal Science
Fear Fear that stimulates an animal to flee or fight is a short-term stress that again involves the adrenal gland, but this time the medulla, where catecholamines such as adrenalin, noradrenalin, and dopamine are released. It helps prepare the animal physiologically for whatever action it decides to take. There is a redistribution of blood to vital organs such as muscle, increased heart rate and blood pressure, increased acuity of the sense organs for hearing and sight, and, in some species, piloerection (cats and dogs) or a tonic immobility (rabbits, chickens). Fear can be measured through measuring end organ functions such as heart rate and muscle tone, as well as circulating catecholamines and behavior.
Mental Distress Mental distress is the product of an animal sensing and responding to an adverse environment and involves a combination of memory of early experiences and diverse feelings. As such, it reflects the psychological well-being of an animal and feelings such as boredom, frustration, anxiety, and anticipating adverse events. The other side of this coin is positive well-being and feelings such as contentment (cat purring), happiness (dog wagging tail), and so on (Yeates and Main 2007). These feelings are recognizable mainly from the behavior of an animal, although research in neuroscience is uncovering some of the underlying mechanisms such as receptor molecules for benzodiazepines, endorphins, serotonin, and noradrenaline. Animals may show stereotyped behavior when they are kept in environmental conditions that are inadequate in some way or another. It is not always possible to decide what an environment should provide; possibly, one that bears some relationship to that in which a species has evolved would be a good starting point (without the predators!). After that, one can realistically only compare types of husbandry systems (e.g., caging versus pen housing). Measurements of the level of diversity of behaviors such as through compiling an ethogram (during the day and the night because many laboratory animals, such as rodents, are crepuscular or nocturnal), what environments animals will choose to inhabit, how hard animals will work to reach such an environment, and the percentage of time they spend carrying out stereotypies will all provide important clues as to how well the environment is meeting an animal’s needs (Lawrence and Rushen 1993). A stereotyped behavior is defined as one that is sometimes abnormal (i.e., not part of an animal’s normal behavioral repertoire), is frequently repeated, is unvarying in its form, and appears to serve no useful purpose within that animal’s current environment (e.g., pacing, chewing bars, digging, and even prolonged periods of inactivity). These behaviors are thought to reflect inner feelings such as frustration (digging) and boredom (inactivity, pacing, chewing) (Mason 1993; Wemelsfelder 1993). To be able to carry out some behaviors seems important to some animals at some times, and they will work hard to gain access to environments that enable them to carry out such activities (e.g., nest building in pregnant sows a few days before birth). These behaviors may be seen as physiological needs for keeping certain essential variables within a more or less narrow range necessary for maintaining life (FAWC 1993), as opposed to being simply part of an animal’s normal repertoire. It is obviously important to be able to identify these needs in order not to stress an animal inadvertently. Interestingly, by providing a more natural or diversified environment, some basic behaviors are carried out in more natural ways. For example, rodents with access to the same food in a hopper and on the floor in the substrate will preferentially forage in the substrate rather than take it from the hopper (known as contrafreeloading). Not meeting these needs may have an adverse impact on the development of the brains of animals as well as on the science (Ashby 1952; van Praag, Kempermann, and Gage 2000). Recent research demonstrated that the CNS neuroreceptor density is changed in animals showing certain
84550.indb 542
11/11/10 6:28:18 PM
Welfare Assessment and Humane End Points
543
stereotypic behavior patterns. Some studies claim to have reversed these types of behaviors by the use of antidepressive drugs such as serotonin reuptake inhibitors. In conclusion, it is not possible to recognize and measure all these adverse states on the basis of a single simple measure. No level of hormone or simple behavior, even in humans, correlates with what human patients report as their experience. All vertebrate animals have integrated genetically programmed bodies, conditioned by experience, with adaptability for survival. One cannot look at such complicated organisms as a single system but rather as one that responds to adverse states according to its intrinsic biology and experience. Take, for example, a dog in pain. She may well be frightened of what will happen next; when the scientist who operated on her comes in through the door, she may change her behavior and become aggressive as a result of what she is feeling and what she sees as potential outcomes. All of these responses are a result of how that animal feels at that time, in those circumstances; its memory; and its well-being. Recognition and Assessment of Animal Well-Being: Practical Application Recognition and assessment of adverse states in animals will vary according to the species, the system of husbandry, and, of course, the individual response of an animal (biological variation is such that animals will inevitably differ even though what is being done to them appears to be the same—for example, lethal dose tests where, at the LD50 dose, 50% of the animals die and 50% survive). Nevertheless, systems for recognizing and scoring animal suffering in a controlled laboratory environment in the context of good husbandry systems that impose a minimum of stress and develop reliable scientific and humane end points can be made. These are described in the following. Evaluation of clinical signs such as whether an animal is eating and drinking, its body weight, its appearance, its posture, and its behavior becomes crucial for underpinning the science and the well-being of that individual animal (Kempermann, Kuhn, and Gage 1997; Morton and Griffiths 1985; Spinelli and Merkowitz 1985; Soma 1987; Morton 2000b; Roughan and Flecknell 2001). Some scientists see assessment of clinical signs as not being objective and as a matter of subjective interpretation; nothing could be further from the truth. What is true is that some clinical signs are metric and can be quantified (like body weight, body temperature, respiration and heat rates, behavior), whereas others are parametric and cannot. These would include signs like quality of respiration (deep, shallow, labored), posture, appearance such as closed eyes, disturbed pelage as with ruffled fur or feathers, diarrhea, coughing, and convulsions (Table€18.3). Such parametric signs tend to be present or not present, though it is often possible to grade them in some way or another. For example, pain in the foot such as a broken bone causing an animal (or human) to limp is a repeatable clinical observation, backed up in this case from verbal evidence in humans that the limping is painful because it hurts to bear weight on the foot. Moreover, an animal may be hopping lame or bear some weight and be limping. Both of these signs are objective observations in exactly the same way that one can observe the stars or that oil floats on water. The cause of the lameness may indeed vary, but the impact on the animal is what matters in this context, and if that pain affects the animal’s appetite and it loses body weight or its body temperature is raised, then we start to get an idea of the overall impact of the experimental or clinical condition on that animal and its ability to cope. The following section is based on the Organization for Economic Cooperation and Development’s (OECD) common conditions and clinical signs that may be indicative that an animal is experiencing pain and distress (OECD 2001). By and large, the list is based on observations in rats and mice, but many of the signs also apply to other mammals used in toxicity testing. When one or more signs or conditions are observed, these should be documented in a written record with the dates of initial and
84550.indb 543
11/11/10 6:28:18 PM
544
Handbook of Laboratory Animal Science
Table€18.3╅ Clinical Signs and Conditions Indicating the Need for Closer Observation, Treatment, or Humane Killing Abortion Agalactia
Analgesia Anemia
Anuria Apathy Ataxia/ uncoordination/ staggering/ unbalanced Bleeding from any orifice Blepharospasm
Blood in feces or urine Blood around nose and eyes Boarded abdomen
Body temperature, abnormal Body weight loss, body condition, or emaciation
Breathing difficulties (dyspnea) Cachexia Chewing, persistent Chromodachryorrhea Circling
Comatose
84550.indb 544
May be detected by fetal remains on bedding, blood on bedding, decrease in abdominal size. May be observed by no milk in stomachs of nursing rodents or failure to express milk from the mammary gland. Young will die and, if not cross fostered or provided with supplemental nutrition or milk, should be humanely killed. See reflexes. Indicates a loss of blood (through feces, urine, reproductive tract) or poor red blood cell replenishment, to the extent that it produces clinical signs of labored or decelerated breathing (also discernible as pale membranes, pale ears and feet, dyspnea, hyperventilation). No urine flow (anuria) due to renal failure (it may be reduced) but worth checking for urine retention. See immobile/inactive. This can be due to neuromuscular coordination, weakness (check body weight), or postseizure recovery period. Observe carefully and continue to check body weight.
See anemia. Some internal hemorrhaging may be detectable as blood escapes from natural orifices. The seriousness will depend on the amount and frequency of the bleeds (q.v. anemia). See eyelid closure. The cause is usually some damage to the eye, and this should be investigated further. If incidental to the study (particularly when only one eye is affected), then veterinary advice should be sought, and the animal may be treated or withdrawn from the study. See anemia. In rodents, it is necessary to differentiate between blood and porphyrin secretion. It is often a stress-related condition in rodents, the secretions not being removed by grooming. If it is blood, the presence in only one nostril may be due to physical injury. This may be detected by holding a small animal up to the ear and squeezing abdomen gently. If breathing stops, then this is indicative of abdominal pain. Causes may be peritonitis due to leakage of gut contents into the abdomen or an inflamed abdominal organ. It is extremely painful. Any alteration in body temperature could be accompanied by a lowered activity level. Hypothermia of more than 10% from normal body temperature may be associated with impending death. Hyperthermia may indicate infection and/or pain. This is particularly of concern when body weight has decreased by more than 20% compared with control animals, or body weight has decreased by more than 25% over a period of 7 days or more. It is usually accompanied by reduced or absence of food intake. Body condition should be scored as well as in chronic conditions (e.g., tumor growth) because body weight may stay the same or even increase, but loss of muscle and subcutaneous fat leads to a marked loss of body condition. This is detectable through feeling the pelvis and backbone, and one may see a square tail because muscle atrophy reveals the square shape of the underlying vertebrae. This can be presented in a variety of signs such as panting, hyperventilation, labored breathing, seesaw or abdominal thoracic breathing, or grunting with each breath (this may also be indicative of abdominal pain). See body weight loss. See compulsive behavior. See blood around nose and eyes. See also ataxia. This is characterized by an animal going repeatedly around the cage, making a track; it may be accompanied by body weight loss. This may indicate damage to the brain or to the inner ear. It may be caused by a concurrent infection, but could also be caused by test substances. See also recumbency. The animal may be unarousable due to extreme lassitude, sedation, etc. or toxic effects of a substance.
11/11/10 6:28:18 PM
Welfare Assessment and Humane End Points
545
Table€18.3╅ Clinical Signs and Conditions Indicating the Need for Closer Observation, Treatment, or Humane Killing (Continued) Compulsive behavior Constipation
Convulsions Corneal ulceration
Coughing/sneezing Cyanosis Dehydration
Diarrhea
Discharge
Dyspnea (difficulty in breathing)€ Edema Emaciation Epistaxis (nasal bleeding) Excitable
Eyelid closure Eyes fixed/sunken Fractured bone Gasping Grooming—failure to do so Hunched/stiff posture Hyperreflexia Immobile/inactive
Such behaviors may include gnawing, biting at the substrate, or even biting at parts of its own body (e.g., feet). This may be indicated by lack of feces in the cage, but must be differentiated from decreased feces due to anorexia. If prolonged, the animal will become lethargic and die. See seizures. This may be accompanied by blepharospasm, watery eyes, and ocular and nasal discharge. It can be particularly painful (early stages) and may be incidental to the study or caused by a test substance, such as drug-induced decreased tear production. Seek veterinary advice, and if recovery is sought, treat under veterinary supervision. If persistent, this may be an intercurrent infection, and veterinary advice should be sought. This is characterized by blue or dark red extremities, such as pinna, feet, and mucous membranes of eye and mouth. This can be assessed by lifting and twisting the skin and observing how quickly it returns to its normal “flat” position. It usually occurs as a result of reduced water intake or inadequate water intake in the case of intestinal (diarrhea), kidney, or endocrine disease (polyuria). Diarrhea can present in a variety of forms from frank watery or bloody feces (dysentery) to soft stools. Increased frequency of defecation can indicate greater severity. Humane criteria listed for body weight should be considered. Normally, animals keep themselves very clean. Discharge may be from any external orifice. Veterinary advice should be sought to differentiate between infectious etiologies and effects of test substances. See breathing difficulties. This can be a cause of severe distress. Characterized by swelling in areas such as extremities (e.g., hock) below the mandible. May be indicative of insufficient heart function or low protein levels in the blood See body weight loss. See anemia and chromodachryorrhea. See seizures. In this case, an animal may be difficult to restrain or catch, and it may throw itself around a cage in a type of fit, causing injuries. This may be due to excessive fear or to neuronal change altering an animal’s behavior. See blepharospasm and corneal ulceration. Eyelids may be fully or partially closed. This is usually observed in the presence of severe body weight loss and dehydration. It indicates that an animal is close to death and should be treated or humanely killed. This may be indicated by a swollen limb or lameness. See dyspnea. In rodents, this may lead to porphyrin accumulations near the eyes and nose, and there may be soiling in the anogenital region. The animal is definitely ill and may be in severe pain and discomfort. See boarded abdomen. This is often seen in sick animals and may be due to abdominal discomfort or just a general sign of illness. See excitable. This is an exaggerated response to a stimulus such as noise or touch. This includes inactivity, lassitude, listlessness, and reluctance to move. The animal is ill and may be close to death if accompanied by body weight loss, dehydration, or sunken or fixed eyes. Carry out the red light response test by turning out the normal white lights and observing the animal in the dark or under a red light, when it will carry out its nocturnal patterns of behavior. (This is normally characterized by an increase in activities such as investigation, climbing, and play within 5 min). (continued)
84550.indb 545
11/11/10 6:28:19 PM
546
Handbook of Laboratory Animal Science
Table€18.3╅ Clinical Signs and Conditions Indicating the Need for Closer Observation, Treatment, or Humane Killing (Continued) Jaundice (icterus)
Joints swollen
Kyphosis
Limping/lameness Locomotory behavior Lordosis
Loss of condition, body muscle Mammary gland abnormalities Moribund Motor excitation Not eating/drinking Pale mucous membranes
Paralysis Paresis
Piloerection Pinna reflex Prostrate Pruritis Pupillary constriction/ dilation
Rales pulmonary
84550.indb 546
This is typically observed by the presence of yellowish-colored ears, feet, and membranes. Serum clinical chemistry (bilirubin) can assist in determining the cause, such as hemolysis (prehepatic icterus), liver damage (hepatic icterus), bile tract blockage (posthepatic icterus), or infection. It may also be accompanied by inactivity when painful condition exists. This painful condition may be indicated when accompanied by a strong withdrawal and vocalization response, an inability to move around freely, relative inactivity compared with controls, or if animal (rodent) remains inactive during the red light response behavior test. (This red light response behavior test is carried out by turning out the normal white lights and observing the animal in the dark or under a red light, when it will carry out its nocturnal patterns of behavior. This is normally characterized by an increase in activities such as investigation, climbing, and play within 5 min). This is characterized by fixed convex/outward curvature of the spine. This may be due to spasm of the flexor muscle of the vertebral column and, if so, would be painful, and the animal should be humanely killed. If intermittent, then it may be a form of seizure (see seizures). The animal may be unable to bear weight fully on that limb due to pain in the foot, leg, or one of the joints. Fractures should be considered as a possible cause. This may be reduced (see immobile) or abnormal in some way. Fixed concave/inward curvature of the spine is lordosis. This may be due to spasm of the extensor muscle of the vertebral column and, if so, would be painful, and the animal should be humanely killed. If intermittent, then it may be a form of seizure (see seizures). See body weight loss. A painful condition may be present if one or more mammary glands are swollen, discolored, or discharging pus or blood, or the animal is extremely sensitive to touching of the gland (vocalization, withdrawal, and overreaction). See comatose. This is an exaggerated movement or limb response to a touch; see hyperrefl↜exia. See body weight loss. See also anemia, cyanosis, and dyspnea. This may be indicative of anemia or circulatory insufficiency (e.g., cardiac or pulmonary insufficiency, or shock). If accompanied by labored or accelerated breathing, it may be indicative of a severe or irreversible condition. A hematocrit can be conducted to quantify the severity of suspected anemia. This may occur because of action of substance on the CNS or spinal cord. Any animal dragging its limbs should be humanely killed. This may occur because of action of substance on the CNS or spinal cord or musculature or neuromuscular junction. Any animal showing obvious muscle weakness that may affect its ability to eat or drink or breathe should be humanely killed. This is noticeable when hairs of an animal’s fur look harsh or starey as they are partially erect. This is a sign of not grooming and general ill health. See refl↜exes. Pinch the ear flap and normally an animal will shake its head. See recumbent. This usually characterizes an animal that has lost its righting reflex and has been in that condition for a few hours. See self-mutilation. An animal may scratch or bite at itself. This may lead to superficial injury, but can progress to deeper lesions and infection. A light responsiveness test should be carried out to determine if the condition is fixed or if there is a pupillary response. Dilatation of the pupil together with inactivity may indicate that an animal is close to death, especially with a sluggish pupil response time. Otherwise, dilation or constriction may well be a substance effect. See dyspnea. This is detected by stethoscope and may indicate pulmonary secretions as a result of intercurrent infection (pneumonia) or a test substance. Substances inducing bronchial and bronchiolar secretions may predispose to infection.
11/11/10 6:28:19 PM
Welfare Assessment and Humane End Points
547
Table€18.3╅ Clinical Signs and Conditions Indicating the Need for Closer Observation, Treatment, or Humane Killing (Continued) Rectal prolapse
Recumbency
Red eye(s)/nose Reflexes
Retention of feces Righting reflex Salivation
Seizures
Self-mutilation
Skin bruising/color/ crepitus Spasm Staggering Sunken flanks
Suppuration Swellings
Tenesmus Tetany Tremor Urine retention Vaginal prolapse
See tenesmus and diarrhea. This occurs when part of the rectum protrudes from the anal sphincter. The animal will have to be humanely killed because the prolapse may become infected or the animal may self-mutilate. See prostrate. The animal may be lateral (on its side) or abdominal, and if it has lost its righting reflex, that is more serious. It may be temporary or prolonged, though, if for more than a few hours, the animal is likely to be close to death if it is not in any form of seizure. See epistaxis and chromodachryorrhea. This is indicative of an animal failing to groom. It may also have a soiled anogenital region. Sluggish responses or loss of reflexes such as corneal, pupillary, pedal, righting (ability to correct to normal posture when gently pushed or overbalanced), or responses to noise may be due to unconsciousness or extreme lassitude. Used to assess anesthesia and analgesia. See constipation. See reflexes. This is indicative of a failure to swallow or hypersalivation in response to the test substance. If the animals is unable to swallow, a clinical examination is required to determine the etiology because this may well affect an its ability to eat (see body weight). The animal may lie on its side and tremor. The muscles may be rigid or flaccid. It may last only for a few seconds or may be longer. It may be brought on by interaction with the observer. If the seizure lasts for more than 1 min and is repeated more than five times a day without being induced, then the animal should be humanely killed, especially if due to the substance being tested. If seizures are induced and further time for study is needed, then animals should be moved to a quiet area and handled minimally. See pruritis. This is licking, scratching, or gnawing at an area, which, if persistent, may result in ulcerative dermatitis. If the extent of the self-mutilation is greater than 2 cm2 or whole phalanges have been removed from the digits, consider humane killing or other appropriate action. This may be due to a subcutaneous bleed or air under the skin (if over the thorax, consider lung puncture and humane killing). If due to gas-forming organisms, treat or humanely kill. See seizures. See ataxia. The abdominal walls of an animal may be suddenly drawn in (writhing, squirming) and can indicate abdominal pain (as in a colicky pain) or it may also be through emaciation (see body weight and dehydration). This is indicative of infection. See discharge, although suppuration may come from sources other than natural orifices. See joint swelling. Note the position and extent of swelling. This may indicate edema (q.v.), hernias of the inguinal or femoral rings, abscess, growth of some sort, bruising, pregnancy, etc. This is constant straining to pass feces. It is usually associated with diarrhea (q.v.) and rectal prolapse. See seizures. The animal may show muscular twitching or rapid skin movements. See also seizures and convulsions If suspected, palpate hardened and distended bladder through the abdominal wall. It is often painful. This can be confused with renal failure. See anuria. Part of the vagina protrudes from the vulva in this condition. The animal will have to be humanely killed because the prolapse may become infected or the animal may self-mutilate. (continued)
84550.indb 547
11/11/10 6:28:19 PM
548
Handbook of Laboratory Animal Science
Table€18.3╅ Clinical Signs and Conditions Indicating the Need for Closer Observation, Treatment, or Humane Killing (Continued) Vocalization
Vomiting
This may be unprovoked, result from handling, or be associated with an animal being fearful of being touched. If abnormal or persistent, it may be indicative of a painful or distressful condition. This is rare in rodents because they lack the physiological reflex and are anatomically unable to do so because of the arrangement of the diaphragmatic musculature. In other animals, check on frequency and volume lost (see body weight and check for fluid loss; see dehydration). If allowed to persist, the animal will die through dehydration and electrolyte imbalance.
subsequent observations and all treatments. If the animal is not humanely killed, a more detailed examination of the animal should be performed, the frequency of observation should be appropriately increased, and the cage or pen should be clearly marked. This list is not all encompassing for every possibility that may occur, and animal care facilities should add other clinical signs and conditions that may be appropriate for specific experimental studies being carried out. Husbandry and the Five Freedoms Laboratory animals are kept in confinement for 100% of their life span, but they may be in an experiment for only a fraction of that time. That is why we have to consider the burdens we place on them for the whole of their lives and not just the impact of the scientific procedure. The five freedoms framework, first put forward in the 1960s (FAWC 1993; Blom et al. 1993), concerns the welfare of all domesticated animals but is derived from its use in farming. It is generally applicable but may not always be appropriate if the scientific objective will be frustrated (e.g., keeping animals in metabolic cages). The five freedoms can be listed in the following manner and will be dealt with sequentially in the following sections:
1. Freedom from thirst, hunger, and malnutrition 2. Freedom from discomfort 3. Freedom from pain, injury, and disease 4. Freedom to express normal behavior 5. Freedom from fear and distress
Freedom from Thirst, Hunger, and Malnutrition Access to food and water and a diet to maintain full health and vigor are important. Laboratory animals normally have free access to clean water, but one aspect of malnutrition—that of obesity— may be a common neglected problem when animals such as rats are kept for lifetime studies. Although we know more about nutrition of the rat probably than that of any other species, including man, overfeeding is a common problem. The general practice of ad libitum feeding of rats leads to their being overweight, with a shortened life span due to accelerated senescence and increase in the frequency of diseases such as nephropathy and various forms of cancer. It has been found that restricted feeding, administering perhaps 75% of normal ad lib food intake, results in leaner and healthier rats with fewer diseases and longer life expectancy (see also Chapter 12). However, restricted feeding is not compatible with social housing, and systems to combine restricted feeding with social housing need to be developed. Nevalainen and colleagues (Kasanen et al. 2009; Kemppinen et al. 2009) have addressed this problem and developed feeding systems where the feed pellets are placed in holes drilled in wood, which the rats have to gnaw through in order to gain access to the food, thereby reducing their feed intake.
84550.indb 548
11/11/10 6:28:19 PM
Welfare Assessment and Humane End Points
549
On occasion, animals may be deprived of food and water as part of an experiment or prior to surgery; however, in some species, such as rats and mice, starving animals can be prejudicial for good health. In these small rodents, the consequences of removing food and water, even for just a few hours, can lead to dehydration and physiological disturbances such as a serious depletion in liver glycogen. In any event, starvation is not necessary for anesthesia because they are unable to vomit due to the prominent cardiac sphincter of their diaphragmatic muscle. Freedom from Discomfort This can be helped by providing a suitable environment including shelter and a comfortable resting area. Chronic discomfort may be inevitably associated with certain experimental protocols and disease models. However, housing animals on wire, rather than solid floors, for prolonged periods may cause some discomfort, and rodents have indicated a preference for solid-bottom cages as well as specific substrates such as sawdust and paper bedding for making nests (Blom et al. 1993, 1995, 1996; van de Weerd et al. 1994, 1997a, 1997b, 1998a). Moreover, overcrowding or mixing unfamiliar animals with each other, mating incompatible pairs (as in the mating of juvenile female mice with heavier mature males for embryo production in transgenesis), or disturbing neonatal mice in the nest, resulting in their not being fed, can compromise animal well-being (van de Weerd 1998b). Freedom from Pain, Injury, and Disease Prevention or rapid diagnosis and treatment can help provide freedom. Freedom from pain is normally one area where therapeutic intervention can easily be employed, such as after surgery (see Chapter 17), where postoperative pain is not a desired scientific outcome but an unwanted side effect; its elimination may well reduce the variance in the data. Other sources of pain may not be possible to relieve because the administration of pain-relieving drugs might interfere with achieving the scientific objectives, such as the use of morphine in neurotransmitter research in the central nervous system. However, this has to be justified so that it is quite certain that such harms are unavoidable (i.e., are really necessary to the science and cannot be minimized in any way). Under normal circumstances, injuries are not frequent in colonies of laboratory animals apart from occasional damage to a claw or when teeth overgrow and have to be clipped. However, the increasing practice of group housing has been associated with an increase in the frequency of bite wounds, and this usually happens when groups become incompatible or when there is some environmental disturbance that disrupts an established hierarchy. Some strains of male mice, for instance, are difficult to maintain in groups after puberty (JWGR 1998, 2003b; van Loo et al. 2000, 2001), as are some of the more common primate species. Nonhuman primates show severely disturbed behavior when raised in isolation, and allogrooming must be seen as an essential need for these animals. If prevented from grooming, the male and female primates have been observed to “groom” the ground, which is a redirection in which a thwarted activity is performed on a substitute object (Bastock, Morris, and Moynihan 1953). Rhesus macaques can exhibit significant aggression and can be difficult to maintain in harmonious groups (Augustsson and Hau 1999), although some laboratories have used group-housing strategies successfully (Bernstein 1991; Guhad, Augustsson, and Hau 1998). Pair housing is obviously much better than single housing and has been used successfully for rhesus macaques (Reinhardt 1987, 1989, 1994), but group housing may still be the best system. A problem with pair housing occurs when one of the animals goes into experiment or is killed before the other. It can then be difficult for the remaining animal to adjust to a new partner, even if one is indeed available within the project. A similar problem occurs when a group is more or less rapidly diminishing over time because of use of animals in projects. This will eventually disturb the
84550.indb 549
11/11/10 6:28:19 PM
550
Handbook of Laboratory Animal Science
established hierarchy in the group, which will be accompanied by the problems experienced when a new group hierarchy is formed (van Loo et al. 2000, 2001). Infectious diseases are normal dealt with by killing the animals, sterilizing the room or rooms (even the building), and introducing a new colony of healthy animals, particularly with rodents and rabbits. It should not be forgotten that disease per se can be a welfare problem due to temporary ill health or even prolonged ill health for some chronic diseases such as pneumonia. Larger animals and now, to an increasing extent, genetically modified mice are often treated as individual patients when suffering from disease. Another problem is when animals suffer from an induced or spontaneous disease when they are being used as a model for a human disorder. These animals obviously need attention, and the alleviation of pain and discomfort is no different from that for any other patient suffering from a similar disease unless, again, such an omission is well justified and based on real validated concerns. Freedom to Express Normal Behavior By providing sufficient space, proper facilities, and company of an animal’s own kind unless there is a veterinary or scientific reason for single housing, this freedom can be addressed. Surprisingly, even subdominant mice that have been injured choose to be with more dominant mice rather than be on their own (van Loo et al. 2002). The benefit for animals of being housed in an environment allowing expression of important species-specific behavior (needs) has been increasingly acknowledged during the past decade. Too often, laboratory animals have been kept in barren environments that lead them to respond by trying to cope with this deficiency through carrying out stereotypic behaviors. Small birds like quail and zebra finches are often kept in barren cages in the laboratory, and modest enrichment has been demonstrated to result in increased locomotor activity, singing, and vocalization (Jacobs et al. 1995; Hawkins et al. 2001). Rabbits particularly have been kept confined in tactilely barren conditions, and the small amount of space for them to exercise has been shown to have pathological effects, such as osteoporosis and a predisposition to fracture (JWGR 1993). Animals may substitute some activities for others; thus, being provided with a plastic piece of gutter may substitute for building a proper nest, even though the animals will show some preferences for nesting materials (Blom et al. 1993, 1995). Rabbits show evidence of boredom such as overgrooming and thus are prone to develop hairballs, which may be lethal to the animals (JWGR 1993). Primates can do the same and may selfmutilate to the point of serious damage, such as autotomy and digit removal; mice show stereotypic escape behaviors; dogs may continually jump; and so on. In particular, for nonhuman primates, those responsible for enclosure designs may learn from the recent literature on housing of these species in zoos and wildlife parks (Estevez and Christman 2006). Novel approaches include the use of electivity indices to analyze the degree of preference for features in the animals’ environments (Ross et al. 2009). A body of evidence specifically looking at common laboratory animals such as rodents is now building up. Experiments on choice and effort for reward are becoming more commonplace; hopefully, we will learn more about animals’ behavior and their needs in the future. In line with the preceding comments, the revised Appendix A to the Council of Europe Convention (ETS 123), as a general rule, says that laboratory animals should be kept in an enriched (enhanced in some way to be more than simply a barren space) environment and that they should be housed in harmonious social groups whenever possible. Being housed with another member or other members of the same species is recognized as the single most important enrichment factor to a singly housed animal. There is thus every good reason to group house animals when possible. However, this is not without its problems, especially when intact males of certain species are housed. As a consequence, there seems to be a trend toward avoiding the use of male rabbits and male primates. The practice of castrating male animals to allow group housing may seem convenient
84550.indb 550
11/11/10 6:28:19 PM
Welfare Assessment and Humane End Points
551
and it certainly seems to work, but this is a practice that is not readily accepted in all countries where castrating animals for convenience is seen as an unacceptable infringement on their integrity. However, this has to be balanced against the quality of the lives of the animals concerned and the fact that if males are not used at all, 50% of all animals born will have to be killed.
Freedom from Fear and Distress By ensuring conditions that avoid mental suffering, freedom from fear and distress can be addressed. This freedom, along with freedom from pain, is particularly important and certainly one of the freedoms where all staff in contact with laboratory animals have the possibility to improve conditions by actively implementing refinement initiatives. There is no doubt that severely stressed animals attempting to cope with a stressor by dramatic physiological and sometimes also behavioral adaptation strategies are unsuited as laboratory animals, because their coping strategies may interfere with the usefulness of the data generated. The extra variance may well even affect the validity of the data and their interpretation. There are therefore scientific as well as humanitarian reasons for eliminating significant stressors for animals. Fearfulness, as in anticipating an intervention or simply being removed from a familiar environment to a strange one, may end up causing more pain to an animal. Thus, trivial procedures such as handling, sexing, and weighing an animal or giving it a subcutaneous injection may be associated with fear of the unknown and even a perception of pain in a timid animal not used to human contact. In contrast, animals confident with their surroundings and staff are less affected by simple procedures and will often welcome the contact with the staff member in spite of being dosed or receiving an injection. It is thus imperative that all staff that will eventually handle and restrain an animal use the necessary time (often just a few minutes every day over a week or two) to handle the animal gently and thus habituate or condition it to the circumstance so that it can adjust and gain confidence. While mice seem to habituate less well to handling than many other species (Irwin et al. 1986; Balcombe, Barnard, and Sandusky 2004), rats habituate very well to people, leading to diminishing stress responses to handling. Larger animals such as primates and dogs may be given a reward after dosing, but for some reason this is not common for rodents. However, it has been practiced successfully with rats in combination with saliva sampling (Guhad and Hau 1996). Animals such as nonhuman primates are easy to train to collaborate in husbandry, veterinary, and experimental procedures. This significantly reduces fear and stress in the animals (Lambeth et al. 2006) and eliminates the need to capture, restrain, and anesthetize (nonhuman primates) the animals prior to even trivial procedures. Reinhardt (2002), Schapiro, Bloomsmith, and Laule (2003), and Laule, Bloomsmith, and Schapiro (2003) provide a number of general examples of positive reinforcement training of nonhuman primates to cooperate for husbandry, veterinary, and research procedures. For a detailed review on this, see Hau and Schapiro (2004). Finally, other routine husbandry procedures such as weaning and identification methods may be seen as routine for the humans, but they are one-off events for the animals. An animal may therefore come to experience (and expect) that every time it is handled, something unpleasant is going to happen; this will inevitably cause fearfulness and some mental distress. Practical examples of this are easy to see when one visits a zoo or other animal facility in which veterinary staff wear a different color coat than other animal care staff. The reaction of the animals is often more fearful toward the veterinarian, who gives injections, than it is to the care staff in “friendly” clothes, who bring food and entertainment. Other useful “animal need indices” (ANI)—such as, for example, the Austrian “Tiergerechtheitsindex,” developed primarily at the farm level as an instrument for assessing and grading livestock housing with respect to the well-being of the animals (Bartusek 1999)—may
84550.indb 551
11/11/10 6:28:20 PM
552
Handbook of Laboratory Animal Science
conveniently be adapted to use in laboratory animal facilities. The approach pursued in Austria considers the five husbandry conditions: (1) freedom of movement, (2) social contact, (3) condition of flooring, (4) stable environment, and (5) stockpersons’ care levels. A scoring system leads to a sum of points, and the ANI values have been divided into different grades of good to poor welfare. In principle, this is a similar practical approach to optimization like the five freedoms, but has the advantage of including an assessment of staff competence and performance, which is relevant also to assessment of laboratory animal facilities. However, it should be noted that it is assumed that these so-called input measures translate into good welfare outcomes and that is by no means certain (Main et al. 2003). In farm animal welfare assessment, greater emphasis is now being given to welfare outcome measures, as well as resource input. In theory, one should inform the other, although contingency planning for emergencies requires a different set of inputs (e.g., spare generators for a power failure, flood defenses). Poole (1991, 1992, 1998) has also written several articles on evaluating an animal’s environment and makes some interesting contrasts with the wild counterparts of a species.
Routine Procedures Used in Husbandry and Research Animals in captivity are completely dependent on human staff with regard to their well-being. To ensure a high state of animal well-being, animal facilities must be constructed and managed with the aim to provide the best possible environment for the species in question, taking into account their physiological and mental (behavioral) needs (T. Poole 1997). The holding rooms for animals should be constructed and ventilated in such a way that animals of different species can be separated not only physically but also with no sight, sound, or smell of other species. This is probably most important when housing predatory species near prey species in the facility. In this context, it should be remembered that mice normally react with anxiety to rats as if they were predators, although this was not confirmed in a recent very large study (Pritchett-Corning, Chang, and Festing 2009). There are also anecdotes of rats and mice living together just as dogs and cats may do. A good hygienic status is imperative to high-quality animal husbandry, and it is a prerequisite to minimize the introduction and spread of infectious diseases. The behavior of rats is affected by cleaning their cage (Saibaba et al. 1996), and when clean bedding is provided to animals, it is now considered important not to disrupt the olfactory environment through leaving a minor proportion of the dirty bedding in the cage, although in mice this has been disputed (Bayne et al. 2002; Ambrose and Morton 1997, 2000). As with all animal husbandry, it is important that routines be adhered to quite rigidly (Willner, Muscat, and Papp 1992). Changes from daily husbandry routines upset the animals, and consistency and habit are important to allow the animals to feel confident and thus thrive and reproduce. Movement and transportation must be considered stressful events for laboratory animals and should be reduced to a minimum. Consequently, facilities for clinical examination and minor procedures like blood sampling and administration of substances should be available in a nearby procedure room. When animals are transported within an institution (e.g., from holding room to laboratory), the transport should be chosen to minimize the stress associated with it (Tuli, Smith, and Morton 1995). Telemetry studies have shown that heart rate, body temperature, and animal activity changes significantly when animals are transported by truck, and that a 10- to 12-day period is required for guinea pigs to return to pretransport levels (Tuli et al. 1995; Stemkens-Sevens et al. 2009). When animals are transported over longer distances, a careful choice of transport caging, transport means, and route should be made in order to make the journey as short as possible with minimum stress to the animals. Nonhuman primates are often transported, in single crates, long distances from breeders in Asia to Europe and the United States; this has been documented to be of
84550.indb 552
11/11/10 6:28:20 PM
Welfare Assessment and Humane End Points
553
significant stress to the animals (Honess, Johnson, and Wolfensohn 2004; Fernstrom et al. 2008). Guidelines exist for transportation of animals, such as the provisions of the European Convention on the Protection of Animals during International Transport (ETS 65). Cages, pens, and other enclosures used for housing laboratory animals should allow sufficient space for them to express a diverse behavioral repertoire, and the environment should provide a reasonable complexity, rendering species-important behavior possible (e.g., dust bath for chickens). It would also seem reasonable to provide sufficient space on the basis of what an animal needs to use as opposed to a simple relationship with size. Thus, young animals that like to play and carry out certain space-occupying behaviors should be given more space than lethargic older and heavier animals (JWGR 1993). However, this is not normally the case, and cage size is almost exclusively based on body weight rather than age. Recommended minimum cage sizes and ranges for acceptable environmental parameters like ambient temperature may be found in Appendix A to the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (ETS 123), 2008). Larger species seem to be more unfortunate with respect to recommended cage sizes in various guidelines. This may be due to tradition and the way these species are customarily housed when they are used as pets (e.g., rabbits and guinea pigs) or farm animals (e.g., horses) in enclosures allowing very little locomotor behavior. It is now generally agreed and advocated in European and North American guidelines that laboratory animals—with the exception of animals that are naturally solitary—should be housed in pairs or groups with members of their own species. Most animals used in research are social or gregarious animals, and housing them singly restricts their natural behavior unnecessarily. Rat pups born to mothers in isolation show significantly lower weight gain and slower locomotor development than pups born in social groups (Young et al. 1996), and when roosters housed in pairs are separated, they respond by a transient increased excretion of corticosteroids, indicating a stress activation of the hypothalamic–pituitary–adrenal axis (Carlsson et al. 2008). Only scientific or veterinary reasons should justify a practice of housing animals singly (e.g., when they are placed in metabolism cages) and then for the shortest period possible. Metabolic cages with grid floors are often used in biomedical research and it has been debated whether housing in these cages is more stressful than single housing in standard rodent cages, as well as how long rodents need to acclimatize to metabolic cages prior to a study. Housing rats on grid floor, as compared with housing in cages with bedding, is in itself associated with significantly higher corticosterone levels (Heidbreder et al. 2000) and an increase in blood pressure and heart rate (Krohn, Hansen, and Dragsted 2003). Gil and colleagues (1999) found an age difference in the catecholamine responses of rats to metabolic cage housing for 1 week. In young rats, urinary noradrenaline excretion decreased during the period, whereas in older rats the levels remained constant during the time in metabolic cages. Urinary adrenaline excretion was similar in both age groups and did not differ significantly during the period when housing was in the metabolic cage. Gomez-Sanchez and Gomez-Sanchez (1991) reported that transfer of rats to metabolic cages was associated with an increase in corticosterone synthesis and advised that scientists allow rats to acclimatize to the metabolism cages prior to experiment. Later studies of excretion of corticosteroids from rats (Eriksson et al. 2004), pigs (Royo et al. 2005), and cynomolgus monkeys (Paramastri et al. 2007) indicated that metabolism cage housing is associated with at least some stress for the animals. Using two animals as the experimental unit in each metabolism cage may be an attractive solution and has been practiced successfully by the authors. Housing animals in groups, however, increases the need for observant and competent staff and may increase the time they have to spend with the animals. On the other hand, that time has to be set against the time and expense saved by reducing the number of cages to be changed and cleaned. It is time consuming to monitor newly formed groups to ensure that aggression injuries are reduced to an acceptable level, but even after a stable hierarchy has been established, fighting can break out
84550.indb 553
11/11/10 6:28:20 PM
554
Handbook of Laboratory Animal Science
if a group is disturbed (e.g., by removing an animal for experimental purposes). Similarly, when animals are returned to the group after having been away to the laboratory, a hierarchy has to be reestablished. Consequently, animals like rabbits and Old World primates are often housed in pairs because this allows social contact between animals and significantly reduces many of the problems associated with housing animals in larger groups (Reinhardt 1987, 1989, 1994). Cages and pens should always be provided with appropriate bedding materials and shelters in which the animals can hide or escape. The bedding serves a number of functions. It allows the animal to remain dry and clean and to create and control its own microenvironment with regard to temperature and light if the bedding is present in quantities generous enough to allow burrowing. Bedding may be used to allow foraging behavior (which is important in primate husbandry, for example), and many species create a nest for sleep and rest from the bedding material provided. The provision of shelters, however, reduces the ability to monitor the well-being of animals, although it meets their biological needs. Many facilities use transparent shelters and the animals seem to use these as much as they use opaque ones, thus solving the problem (Sherwin 2002). Acclimatization of animals to a new room or cage takes some days, and it is important that this period of acclimatization be permitted; otherwise, it may affect the scientific data. For example, Damon and co-workers (1986) found a 60-fold difference in the LD50 dose between animals that had been acclimated to a metabolic cage and those that had not. Nonhuman primates make take much longer to acclimatize to a new environment; in particular, wild-caught animals may take 4–8 months to habituate to a captive environment (Kagira et al. 2008). This is yet one more argument against using wild-caught animals for biomedical research. Using restraint for the administration of substances and the removal of body fluids, if not done well, can also be aversive to an animal, particularly when the procedure is repeated over several days or weeks (Tuli 1993; Wadham 1996). In fact, animals may never habituate to the procedure and show an adrenal response from both the medulla and the cortex. Experimental Procedures The ethical framework of the three Rs of Russell and Burch (1959)—replacement, reduction, and refinement—can be modified slightly from the original concepts and turned into a series of questions when the use of animals in a research project is being considered. Are there nonsentient alternatives available that could replace the use of animals? Are the numbers of animals appropriate for statistical acceptability? And, finally, can the amount of suffering be reduced (e.g., provision of postoperative analgesia) or can the well-being of the animals be improved in any way (e.g., an enriched husbandry system)? Refinement has been called the Cinderella of the three Rs (Hau and Carver 1994; Van Zutphen 1998) and should be considered to be much broader than simply relating it to the suffering of an individual animal during an experiment. It may even be possible to improve the experimental design so that all the animals used in the project suffer less or fewer animals are used by staging the work so that key experiments are carried out first. Other approaches to improving the design might incorporate pilot studies, a grading of stimuli (e.g., in developing a novel analgesic, to give a minor nociceptive stimulus before a major one or, in cancer research, to evaluate treatment on small tumors before progressing to larger ones), and that in vitro work precedes in vivo work when appropriate (e.g., cell toxicity of novel chemicals) (Morton 1998b). The rapid development in minimally invasive surgery and imaging technology such as x-ray, magnetic resonance imaging, positron emission technology, bioluminescence, and fluorescent markers (Hudson 2005; Hagelschuer et al. 2009) may represent major advances with respect to refinement and reduction in biomedical research. Imaging techniques generally allow serial studies of individual animals and reduce the overall number of animals needed to achieve statistical
84550.indb 554
11/11/10 6:28:20 PM
Welfare Assessment and Humane End Points
555
significance. A greater amount of data can be obtained from smaller numbers of animals and there is often no invasive surgery involved. Imaging techniques may introduce earlier humane end points and replace invasive procedures. It should be noted, however, that using fewer animals by subjecting the animals used to multiple procedures (injections, anesthesia events, and scans) is not necessarily progress in terms of animal welfare. Telemetry technology—radiotelemetry—is associated with similar potential positive and negative consequences with respect to animal welfare; these have been discussed at length in a report of the JWGR (2003a). In the report, recommendations are made on key issues, including selecting a device; external attachment or implantation; surgery; anesthesia; postoperative care, including pain management; monitoring animal well-being; housing and care of animals on telemetry studies; conducting field studies using wild animals; removing devices; rehoming or releasing animals; and disseminating good practice in publications. The report also includes guidance aimed specifically at ethics or animal care and use committees to help them assess projects involving telemetry and decide whether the technique is justified in each case. The advantages of radiotelemetric devices include remote monitoring of scientific outcome measures (such as heart rate, blood pressure, and even levels of chemical ions such as sodium, potassium, and oxygen levels as well as neurotransmitter levels) without disturbing the animal through capture and restraint, which are both stressful for the animal (especially for wild animals used in field studies). If it is possible to have an indwelling probe to detect something, then it is theoretically possible for signals to be transmitted to a receiver to record and collate the data. In other circumstances where the animal may not be within range of a receiver or be on the move (as in bird migration or fish swimming to their breeding grounds), data can be stored and a data logger recovered sometime later. Animals can be reused without any further intervention (except perhaps removal of a data logger unless it is programmed to transmit and reboot). However, the disadvantages are that it normally requires an anesthetic and surgery to implant devices, unlike small microchip transponders that can easily be injected subcutaneously. These surgical procedures carry the risk of infection at a later date (possibly when the animals are miles away), interference with normal functioning such as locomotion, swimming, or flying, and a further operation to remove the device to reuse it or to retrieve the data. On the other hand, not all radiotelemeters need to implanted and many can be mounted on the external surface of an animal (e.g., on a collar). However, they may come off and be less stable in an animal’s environment because they can catch on cage or pen surfaces or the floor as animals carry out normal behaviors such as play, grooming, and locomotion. In order to minimize such risks, animals may also be singly housed, which may be necessary if more than one telemetered animal is being used at the same time, and there is the additional stress of social isolation. Next, we will examine how to determine when animals are suffering and to what degree and how this information can be used to determine and implement humane end points as well as to gain valuable clinical information in medical research involving animal models of human disease, including transgenic animals. Score Sheets Establishing a Score Sheet A score sheet is simply a list of those clinical signs likely to be seen in a particular scientific procedure for an individual species or even for a particular strain or breed of that species (see Tables€18.1 and 18.4) (OECD 2001; Morton 2000b; Van Zutphen 1998). It is not possible to prepare a generalized list of clinical signs that will occur in all experiments (compare the predictable side effects for a failed skin graft with a failed liver or heterotopic heart graft), so score sheets have to be drawn up specifically
84550.indb 555
11/11/10 6:28:21 PM
84550.indb 556
19/11/01
DATE OF OPERATION
– –
37.4 – – +/– NT – –
– ? 10 – 3 3 – – – –
Not inquisitive and alert Not eating jelly mash No. of pellets given to rat No. of pellets left No. of grapes given to rat No. of grapes left Grain (sprinkling) Not drinking No feces passed Body weight (g) Change from start (%) Body temp (°C) Crusty red eyes/nose Sunken eyes Eyes half closed No red light response (not tested = NT) Coat soiling Pale eyes and ears
– – – 80 N –
20/11 1 7.45
– – 4 4 3 0 Y – – 339 4.3%+ 37.5 – – – NT
– – –/+ 80 N ?
19/11 0 16.45
– –
38.7 – – – NT
3 2 Y – –
– – 13
– – – 90 N –
1 16.35
PRE-OP WEIGHT: 325g
ISSUE No.
Inactive Hunched posture Starey coat Rate of breathing (per min) Type of breathing (*) Not grooming
DATE DAY TIME
SPSHR4
RAT No.
2 16.15
– –
8 4 2 1.5 Y –/+ – 312 –4.30% 36.2 – – – NT
–
On Handling
– – – 80 N –
– –
37.1 – – – NT
–/+ –
–
– – – 80 N –
From a Distance
21/11 2 8.35
Table€18.4╅Score Sheet from an Experiment on Renal Nerve Physiology
– –
+/– –/+ 8 5.5 3 2 Y + +/– 296 –9.02% 37.5 – – – NT
+/– – + 80 N +/–
22/11 3 8.25
– –
35.7 – – + NT
+ –/+ 10 5.5 2 2 Y + 3
+ +/– + 84 N +
3 16.05
+ –
37.5 + +/– +/– NT
+ ? 10 10 2 2 Y + 3
+ – +/– 80 N +
3 22.05
– +/–
+/– + 5 10 3 2 Y + + 282 –13.20% 36.9 + + + +
+/– – + 90 N +/–
23/11 4 7.25
– +
35.9 + + + +
5 3 3 Y + +
+ +
+ – + 120 R +
4 12.05
– +
5 3 3 Y + + 275 –15.40% 34.7 + + + +
+ +
+ – + 120 R +
4 4.25
556 Handbook of Laboratory Animal Science
11/11/10 6:28:21 PM
84550.indb 557
– – – – ++ – – + – Y Y Y Y Y
– – – – +R – – + + Y Y Y Y Y
Y Y Y Y Y
– – – – ++ – – + – Y Y Y Y Y
+/– – – – ++ BLUE + + – Y Y Y Y Y
+/– – – – ++ BLUE + + – Y Y Y Y Y
+/– – – – ++ BLUE + + – Y Y Y Y Y
– – – – ++ BLUE + + – Y Y Y Y Y
– – – – ++ BLUE + + – Y Y Y Y Y
+ – – – ++ BLUE +/– + – Y Y Y Y Y
+ – – – ++ BLUE +/– + – Y Y Y Y Y
+ – – – ++ BLUE +/– + –
Special husbandry requirements: Animals should be kept on Vetbed 24 h postop in a warm environment. Animals will be used in a rodent access cage. Scoring details: * Dragging leg: score as + if can move leg forward and ++ if cannot * Breathing: R = rapid; S = shallow; L = labored; N = normal Humane end points and actions: If bleeding occurs at the cannula exit site, this will be treated by the veterinarian. Animals will be checked at least twice daily and daily clinical monitoring of animals (body weight, respiration, behavior, eating and drinking, bowel habits, etc.) will be carried out to detect any adverse effects. Blood pressure and renal nerve activity will be measured regularly and any marked deviation from normal values and patterns will be evaluated together with the clinical signs. If there is a clear indication of progressive deterioration, the animal will be killed. Any animal showing abnormal motor or sensory effects will not be allowed to proceed into the study without further discussion with veterinarian and animal care staff. Any animal not returning to its starting body weight within 7 days will be killed. Any animal losing greater than 20% body weight in a 2-week period will be carefully examined with respect to the other clinical signs, and its growth rate may be an important factor. Scientific measures: Blood sample to be taken prior to killing but after loss of consciousness.
Dehydration Diarrhea Hyperactive Circling *Dragging leg L R Cold limb L R Toes curled Lameness Blue extremities, more than one foot Head wound OK Neck wound OK Right side jugular wound OK Right side femoral wound OK Flank left Drug for baroroflex curve NAD (nothing abnormal detected) Other Vocalizing on handling or approach SIGNATURE:
Welfare Assessment and Humane End Points 557
11/11/10 6:28:21 PM
558
Handbook of Laboratory Animal Science
for each scientific procedure. They are therefore usually specific for each of those characteristics— experiment, species, and strain. The use of pilot studies as recommended by the CCAC (CCAC 1998; van den Heuvel et al. 1990) helps determine what the relevant and important clinical signs are likely to be on the first few animals; in this regard, the opinion of the animal care staff will be invaluable. The laboratory animal veterinarian should also be skilled in identifying objective clinical signs from his or her knowledge of the biology of the species, including the range of an animal’s relevant behavioral and physiological responses (Kuijpers and Walvoort 1991; Morton and Townsend 1995; Schlede et al. 1992; Schlede, Diener, and Gerner 1999; Beynen et al. 1987). Regularly observing animals throughout a pilot study will help identify critical periods during the experiment when animal well-being is particularly at risk (e.g., in the immediate postoperative period; or, in a study on infection, after the incubation period; or at the predicted time of tumor growth or organ graft failure). The score sheet first lists signs that can be observed from a distance in a relatively undisturbed animal, which should be showing its natural unprovoked behavior (e.g., appearance, behavior, posture, respiratory rate and pattern). The animal is then observed at closer range, followed by close examination of the animal (body weight, body condition and temperature, heart rate, dehydration). The signs are scored as being present (+) or absent (–); if the observer is unsure, then that too can be indicated (+/–). Clinical signs are reduced to an observation that can be scored in this binary way (i.e., is present or absent); otherwise misinterpretation and subjective evaluation (i.e., observer error) may creep in. This is one of the strengths of the scheme because it leaves little room for observer error; such errors, when they have occurred, have usually been at the lower levels of intensity (Soothill, Morton, and Ahmad 1992). The convention is that negative signs indicate normality (i.e., within the normal range) and that positive signs indicate the animal is outside the normal range. However, this results in some contorted descriptors because an animal showing convulsions would be a sign of poor wellbeing and be scored a (+), but if an animal were eating, then this is a good sign but could be scored also as a (+). In this latter case, the descriptor is changed to “not eating” so that if an animal is not eating it is scored as a (+), and the convention is maintained. Using this convention, it is possible to scan a score sheet to gain an overall impression of animal well-being: The more plusses that appear, the more an animal has deviated from normality, with the inference that the scientific procedure is having more impact than before. It is not unreasonable to assume—and this would be borne out by human experience—that the greater the deviation from normality has been, the greater the impact has been. Thus, an animal may lose body weight because it is in pain or feeling unwell because of an infection and thus does not eat. This weight loss reflects one impact of the experiment on that animal. If this can be reversed through the provision of analgesics or other treatment, then the severity can be reduced and the well-being of the animal, as well as the science, improved. Other important data can be recorded on the sheets in addition to the clinical signs. In this way, they become a more complete record of the experiment and may prompt correlating factors that are not immediately obvious. At the bottom of the sheet, there are guidance notes on what should be provided for the animals in terms of husbandry and care. There are also guidelines on how to record qualitative clinical signs (such as lameness, diarrhea, respiration), as well as criteria concerning when to implement humane end points (see later discussion). Finally, if an animal has to be killed, there are instructions about what other actions should be taken, such as a blood sample before killing, so that the maximum amount of information is always obtained from a study. While these sheets take time to fill in, it is relatively easy for an experienced person to see if an animal is unwell, so the NAD (nothing abnormal diagnosed) box is simply checked. However, if an animal is not normal, it takes time to score it and to make judgments over what actions should be taken; that is the price for practicing humane science.
84550.indb 558
11/11/10 6:28:21 PM
Welfare Assessment and Humane End Points
559
Interpreting a Score Sheet It can be seen from Table€18.4 that, as time goes on, there are more plusses to the right than to the left. Several other points should be noted. As the rat started to show clinical signs, it was scored more frequently. On Day 0, some clinical signs were seen as the animal had just had a 7 h operation; the following day, it seemed to be recovering well: It had gained weight and was eating, although it was lame in the right leg due to the placement of a femoral cannula. Over the next 3 days, it lost body weight and its body temperature gradually decreased. Its activity appeared to vary, but eventually it did not respond to the red light test (this test is carried out by turning out the normal white daytime lights and observing rats in the dark or under a red light, when it will carry out its nocturnal patterns of behavior characterized by an increase in activities such as investigation, climbing, and play within 5 min). By Day 4 the coat had become starey (ruffled), the body temperature had dropped significantly, and the breathing had become more rapid and labored. Furthermore, there was a significant body weight loss (15.4%) accompanied by a progressive dehydration—a strong indication that the animal had not eaten or drunk much, if anything, during the previous 3 days. In addition, the blood supply to the limb was obviously impaired; this was a surprising observation because, in the control strains of rat, this lameness is normally reversed fairly quickly. Thus, there seemed to be a strain difference with the SPSHRs (stroke-prone spontaneously hypertensive rats). The rapid weight loss and dehydration, labored breathing, inactivity, lack of a red light response, etc. confirmed that the animal was becoming severely physiologically compromised and was not going to yield valid results in relation to the scientific objective. Even more significantly, its temperature was now 34.7°C—a very poor sign. In our experience from following such animals through to death in earlier studies, this animal would have died over the next day or so; consequently, it was decided to kill the animal on humane as well as on scientific grounds before the end of the experiment. Even if the animal might not have died, the level of suffering was agreed to be a sufficient reason to kill the animal on humane grounds alone. In the United Kingdom, where an ethical balance is struck between the anticipated benefits of a research project and the degree of animal suffering (as indicated by the humane end points), the severity limit of “moderate” had been exceeded. Score sheets are important tools for identifying clinical signs indicative of treatment effect. These signs may assist in determination of subsets of characteristics for which deviations may be used to develop and introduce earlier end points of animal experiments.
Humane End Points As soon as an animal ceases to be scientifically useful in an experiment (e.g., due to physiological or psychological perturbations) or the objective of the experiment has been achieved, it will normally be humanely killed. Humane end points can also be implemented when they reflect suffering that cannot be justified or when particular clinical signs affirm a specific outcome such as pending death (Tables€18.5 and 18.6) (OECD 2001; CCAC 1998; van den Heuvel et al. 1990). For example, decreased body temperature has been associated with pending death in mice with bacterial infections (Cussler, Morton, and Hendriksen 1999) and slow circling movements in mice with neurotropic viruses such as rabies (Hendriksen and Morton 1999). Many other approaches can be used; the reader is referred to the report of the first meeting on humane end points (Johannes et al. 1999) in which papers in various fields of research give not only clinical signs but also chemical and hematological markers of end points. This approach of using surrogate signs for scientific outcome measures such as death has to be validated by allowing some animals to die and carefully
84550.indb 559
11/11/10 6:28:21 PM
84550.indb 560
Clinical biochemistry Hematology Hematocrit PCV low Pale feet, tongue, membranes, mucosae Count breaths/min Type of breathing: Rapid, slow Deep, shallow Regular, irregular Color of membranes and extremities Beats per minute (difficult in small animals) May indicate animal in pain
Clinical thermometer Thermistors (rapid) Radiotelemeter (expensive) Transponder
Measure circumference Measure weight bearing Assess pain response
Clinical biochemistry and hematology
Anemia Blood in urine or feces
Body temperature: Hyperthermia Hypothermia
Joints swollen
Mammary glands: swollen Pus and blood in milk
a
VD = veterinary direction.
Heart rate
Breathing rate
Food and water intake as a percent of normal intake Body weight loss Body condition score
How Measured
Anorexia—see also: Appetite poor Body weight loss Body condition poor (cachexia, emaciation)
Analogue Objective Signs
If unexpected, is animal still scientifically useful? Is there a dependent litter? cross foster Treat under VDa
If unexpected, can antiinflammatory drugs be given? Infection?
Failed remedial action and if more than 4°C away from normal for more than 12 h More critical in smaller mammals (e.g., young mice) If unexpected and in severe pain and associated with body weight loss (see Anorexia) If persistent despite treatment
If prolonged or very high (50% increase)
Assess cause—if raised for prolonged periods, may lead to heart failure over several weeks Assess cause Remedial action (e.g., cooling, warming)
If unexpected and in pain
If more than 4°C away from normal and remedial action failed
If raised by more than 25% for more than a few days
If animal is becoming distressed
If animal is becoming distressed, in shock (very pale), or cyanotic
Assess cause (e.g., increased rates can indicate infection, pain, low oxygen, high CO2, anemia) If unexpected, will it affect results?
If animal likely to become irreversibly dehydrated If body condition score is low
End Point Consideration
If animal is becoming unexpectedly anemic
If amounts to starvation conditions for more than 2 or 3 days (20% body weight loss) or 25% loss over 7 days and is likely to persist
Indicator for Euthanasia
If animal is becoming severely clinically anemic: labored or increased breathing rate, edema
Assess cause—if diet intake is reduced and body weight loss is more than 10% over 2 days (NB ascites and tumor growths may mask loss of body weight, so body condition scoring becomes important) Assess cause—if unexpected, will anemia affect results?
Indicator for Intervention
Table€18.5╅ Examples of Clinical Signs and Actions That Could Be Taken
560 Handbook of Laboratory Animal Science
11/11/10 6:28:22 PM
Welfare Assessment and Humane End Points
561
recording the clinical signs throughout the animals’ lives so that a retrospective analysis can be done. It has been used successfully, and refinements to some of the protocols required for safety testing of vaccines have been proposed (Krug and Cussler 1999). It is easier to apply humane end points to animals that are used in areas of research such as safety assessment, where routine procedures are used and the only variable is, for example, a new chemical entity or a batch of vaccine, because comparisons can be made with previous tests and experience. However, many promising attempts are being made in other research disciplines (Krug and Cussler 1999; Johannes et al. 1999). If the scientific outcome is going to be frustrated by the suffering that the animal is undergoing (psychological as well as physiological), then it is appropriate to kill those animals humanely because keeping them alive will not generate valid, reproducible, and reliable scientific data. Transgenic Animals The growth in the development, generation, and use of transgenic animals (mainly mice) raises considerable animal welfare concerns (Hubrecht 1995; Poole 1995; Mepham et al. 1998; Van der Meer et al. 2001; Robinson et al. 2003; FELASA 2007). The numbers generated per year have increased exponentially over the past two decades and the trend looks likely to continue. Many physiological functions are under the control of more than one gene (i.e., they are polygenic), so some considerable research will be needed to identify the important and appropriate combinations. The insertion of DNA into the genome of animals or the deletion of specific genes gives rise to unpredictable outcomes in terms of animal well-being (as well as scientific results) in the first instance, but, thereafter, transgenic lines can be selected and bred or cloned to avoid or select for a specific genotype. It is not an accurate science, although the methodology is constantly improving to avoid unwanted effects such as the expression of a lethal gene and to yield more accurate targeting of gene insertion. Embryo manipulation procedures do not appear to affect the welfare of offspring in the mouse (van der Meer and Van Zutphen 1997), and the large offspring syndrome observed in farm ruminants, such as cattle and goats (EFSA 2008), has not been reported in rodents. It is important to remember, though, that absence of evidence is not evidence of absence, and time will tell whether this syndrome may occur also in rodents and other species. The potential welfare risk factors associated with gene modification in animals include possible loss of gene function, harmful effects of transgene-derived protein, unexpected physiological and behavioral results of manipulation, and, of course, the pain and discomfort for the animals in which the expected disease model develops satisfactorily. Gene insertion through microinjection into the pronucleus of a fertilized egg may disrupt other genes and may be inappropriately expressed in some tissues, and the overall effect may be dependent on copy number. Increased fetal and juvenile morbidity and mortality is a frequent finding in genetically modified animals, and there is therefore a need for detailed monitoring of neonatal, juvenile, and adult animals for their lifetimes to identify any adverse effects. Score sheets can help standardize the signs to look for, and there have been many such publications (Broom 1997; JWGR 2003b; van der Meer and Van Zutphen 1997; van der Meer, Baumans, and Van Zutphen 1997; Costa 1997; Jenkins and Francis 1999; Dennis 2000; Robinson et al. 2003). There are also good scientific reasons for detailed monitoring and description of phenotype and behavior in order to study the effect of the gene under study, as well as the unexpected side effects of the manipulation. It is noteworthy that in the production of transgenic farm animals, the scientist often tests the system in mice first and that the welfare problems seen in the transgenic farm animals were often present in the test mice, but were not spotted because of suboptimal monitoring of the mice.
84550.indb 561
11/11/10 6:28:22 PM
562
Handbook of Laboratory Animal Science
Table€18.6╅ Questions to Determine Whether Earliest Possible End Points Have Been Sought What are the scientific justifications for using the proposed end point? Have all existing relevant data been evaluated? What is the expected time course for the animals from the initial treatment to first signs of pain and distress, to the death of the animal? When are the effects to the animal expected to be the most severe? If the course of adverse effects cannot be determined prior to the start of the study, could they be developed through the conduct of a pilot study with appropriate observations by the animal care and veterinary staff? Has a list of observations on which the end point will be based been developed? Who will monitor the animal and maintain records of observations? Has a chain for reporting observation findings been established? What will be the frequency of observations during the course of the study and during those times predicted to be critical for the animals? Do the investigators, veterinary care, and animal care staffs have the training and experience necessary to perform the observations necessary to monitor the animals effectively and efficiently? What steps have been implemented to attend to animals that demonstrate severe signs and symptoms? Source: CCAC (Canadian Council on Animal Care). 1998. Guideline on choosing an appropriate endpoint in experiments using animals for research.
The generation of transgenic animals raises concerns for the animals involved in the production as well as when they are being bred to maintain the line; some of these are detailed in Tables€18.6 and 18.8. Some of these concerns also apply to natural mutations or those induced by the use of mutagens such as ethyl-nitroso-urea. There is one difference in that the generation of transgenic animals involves more technical procedures (Table€18.7), but the maintenance of a line raises similar
Table€18.7╅Assessment of Adverse Effects and Technique or Event Technique or Event Handling Anesthesia Vasectomy
Acclimation, frequency, time constraints (1000s animals?) Dystress Scrotal laparotomy
Superovulation Mating
Injection Normal disparity
Cross-breeding of genetically modified strains Embryo transfer
Knockouts Knockins “Natural” mutants Vaginal Laparoscope Laparotomy Early fetal Late fetal/neonatal Normal Dystocia/Caesarean Physical/anesthetic gas Inhalation of CO2 Peritoneal injection of sodium pentobarbitone
Nuclear transfer Premature death Birth Euthanasia
a
84550.indb 562
Examples
Estimated Severity Mild Mild Mild + Moderatea Mild + Mild Mild to moderate Mild to substantial
Mild Mild + Moderatea Mild (dam?) Mild + Moderate Moderate + substantiala Mild Moderate Mild to moderate
When analgesics are not given.
11/11/10 6:28:22 PM
Welfare Assessment and Humane End Points
563
Table€18.8╅Some Adverse Effects Associated with Husbandry Practices and Some Genetic Variables Technique Husbandry Individual identification
Genetic identification
Time of identification Construct characters
Event Social isolation Grouping aggressive strains Ear notching Ear tagging Microchip Hair dye Tail tipping Bleeding Body cells and tissues (e.g., blood, salivary swab, hair) As soon as possible (e.g., before weaning) Number of copies Mode of expression Promoter Gene character
Estimated Severity Mild/moderate Mild Mild Mild to moderate Mild Mild Moderate Mild Mild Depends on enrichment strategies Neutral to substantial
issues for genetic mutants as well as genetically modified animals (Table€18.8). Many of the adverse effects may be seen in the background strain, so it is important that the effect of the mutation or modification be distinguished from that. Moreover, the effects may be subtle and may only be noticed after extensive testing (e.g., using the SHIRPA testing protocol; JWGR 2003b; Rogers et al. 1997). Clinical observations by skilled animal care staff may produce invaluable results in relation to the phenotypic effect of the genetic modification. The animal care staff are far more likely to observe significant behavior and developmental differences than the scientists trying to understand the gene function; nevertheless, together, they will make a strong team. Again, score sheets are particularly useful when the genetic effects are known and when animals may have to be humanely killed before their suffering is too great. As in vaccine testing protocols, the repeated observation of the same scientific event enables clinical signs to be predictable and humane end points to be reliably established. For each transgenic line, a passport should be drawn up to accompany animals to other laboratories giving details of the clinical signs, their time of onset, incidence, and methods of alleviation and humane end points. Humane Killing of Animals Nearly all research animals have to be killed at the end of a scientific procedure so that tissues can be retrieved for analysis and to comply with legislation such as the EU directive 86/609/EEC (see Chapter 3, this volume). On some occasions, this may not be necessary. In those cases, it may be possible to release animals back to the wild (e.g., wild-caught animals, animals bred in captivity such as early free-feeding amphibia forms such as tadpoles, birds hatched from eggs taken from the wild), as pets, or back to the farm or a retirement home (e.g., chimpanzees) or even to return them to stock for reuse. However, this must be carefully controlled to avoid inadvertent reuse of animals. Several publications deal with humane methods of killing (see also Chapter 17, this volume), especially involving farm animals, where billions of animals will be killed each year. In research laboratories, there is sometimes some concern that the method of killing may affect the scientific data. If this is the case and an inferior method in regard to the welfare of the animals is being used,
84550.indb 563
11/11/10 6:28:22 PM
564
Handbook of Laboratory Animal Science
then that method should be validated and the unwanted side effect clearly shown, rather than assuming that it does. While an overdose of nearly all anesthetics is acceptable for euthanasia, the use of carbon dioxide has recently been brought into question by several authors (Leach et al. 2002a, 2002b; EFSA 2005; Kirkden, Neil, and Weary 2005a, 2005b; Niel and Weary 2006, 2007; Niel, Stewart, and Weary 2007a,b).
Concluding Remarks We realize that many other trends and developments in refinement and optimization of animal welfare deserve recognition—for instance, the welcome introduction of imaging techniques and telemetry technology allowing observations of physiological parameters from a distance and thus making many invasive techniques obsolete. The use of detailed behavioral analyses and search for immunochemical, affymetrix gene arrays and neurological markers of stress and reduced welfare—as well as the use of classical stress markers, quantified in urine, feces, saliva, and body hair, and pathology in the assessment of stress, health, and welfare—are active and interesting research fields. However, we had to draw the line somewhere, and we hope to have given readers an update on developments in the humane use of animals in research. The importance of the need for enrichment of animal cages and pens—particularly, provision of a suitable social environment—is gaining general recognition in the biomedical community (Lancet 1993); it is our hope that this chapter will give food for thought and encourage scientists to include welfare assessment routinely as an integral component of their research projects. That the highest possible welfare state of animals goes hand in hand with the generation of the sound scientific results has been known since the 1950s (Russell and Burch 1959; Chance 1956, 1957), but assessment of animals’ well-being is not a simple business. The welfare state of an animal is a highly complex concept, and we have chosen a holistic and practical approach to assess and improve animal welfare. We agree with Webster (1998), who was concerned “by the misuse of scientific method by those who seek to obtain so-called objective measurement of something which they preconceive to be a stress.” As we see it, there are no easy measures of welfare, but behavioral and clinical measures should be components of most, if not all, proper welfare assessments. This requires that all technical, scientific, and veterinary staff engaged in laboratory animal husbandry and experimentation must have a proper knowledge of normal biology and behavior of the species with which they work. Only then will it be possible to identify abnormal behavior and clinical changes associated with a scientific procedure or system of husbandry. Knowledge and competence obtained through education and training of all staff categories are essential for continuous refinement of animal experimentation (see also Chapter 5, this volume). In addition to competence, the way forward also requires commitment and collaboration (Hau 1999) from scientific and technical staff to improve the welfare of animals in experiments and to search actively for ways and means to terminate experiments as early as possible when animal welfare is being compromised. This includes actively searching for behavioral, clinical, and physiological parameters that may be used as indices of effect of treatment rather than adverse welfare states, especially those that involve death or a moribund state. However, it must always be remembered that no matter what is measured objectively, whether numerically or by observation, the next step of interpretation will always be subjective. Collaboration through good communication between laboratory animal technicians, scientists, laboratory animal veterinarians, institutional certificate holders, local and regional ethics committees, and the regulatory and inspection authorities are all vital to progress in this area.
84550.indb 564
11/11/10 6:28:22 PM
Welfare Assessment and Humane End Points
565
References Ambrose, N., and D. B. Morton. 1997. The effects of environmental enrichment on cage-cleaning aggression in male laboratory mice. BandK Sci. Now 6 (1): 1–3. ———. 2000. The use of cage enrichment to reduce male mouse aggression. Journal of Applied Animal Welfare Science 3:117–126. Appleby, M. C., and B. O. Hughes. 1997. Animal welfare. Wallingford, Oxon, UK: CAB International, 316 pp. Ashby, W. R. 1952. Design for a brain. London: Chapman & Hall. Augustsson, H., and J. A. Hau. 1999. Simple ethological monitoring system to assess social stress in grouphoused laboratory rhesus macaques. Journal of Medical Primatology 28 (2): 84–90. Balcombe, J. P., N. D. Barnard, and C. Sandusky. 2004. Laboratory routines cause animal stress. Contemporary Topics in Laboratory Animal Science 43:42. Bartusek, H. 1999. A review of the animal needs index (ANI) for the assessment of animals’ well-being in the housing systems for Austrian proprietary products and legislation. Livestock Production Science 61:2–3, 179–192. Bastock M., D. Morris, and M. Moynihan. 1953. Some comments on conflict and thwarting in animals. Behaviour 6:66–84. Bayne, K. A. L., J. A. Mench, B. V. Beaver, and D. B. Morton. 2002. Laboratory animal behavior. In Laboratory animal medicine, 2nd ed., chap. 32. New York: Academic Press, 2002. Bekoff, M., and C. A. Meaney. 1997. Encyclopedia of animal rights and animal welfare. Westport, CT: Greenwood Publishing Group. Bernstein, I. S. 1991. Social housing of monkeys and apes: Group formations. Laboratory Animal Science 41:329–333. Beynen A. C., V. Baumans, A. P. Bertens, et al. 1987. Assessment of discomfort in gallstone-bearing mice: A practical example of the problems encountered in an attempt to recognize discomfort in laboratory animals. Lab Anim 21:35–42. Blom, H. J. M., V. Baumans, C. J. A. H. Van Vorstenbosch, L. F. M. Van Zutphen, and A. C. Beynen. 1993. Preference tests with rodents to assess housing conditions. Animal Welfare 2:1–87. Blom, H. J. M., G. Van Tintelen, V. Baumans, J. Van Den Broek, and A. C. Beynen. 1995. Development and application of a preference test system to evaluate housing conditions for laboratory rats. Applied Animal Behavior Science 43:279–290. Blom, H. J. M., G. Van Tintelen, C. J. A. H. Van Vorstenbosch, V. Baumans, and A. C. Beynen. 1996. Preferences of mice and rats for types of bedding material. Lab Anim 30:234–244. Broom, D. M. 1997. Assessing the welfare of transgenic animals. In Welfare aspects of transgenic animals, ed. L. F. M. van Zutphen and M. van der Meer, 58–67. Heidelberg: Springer-Verlag. Broom, D. M., and K. G. Johnson. 1993. Stress and animal welfare. London: Chapman & Hall. Bock, B., and J. Webster. 2009. Oral communications at Welfare Quality Meeting, Florence, March 2009 http://www.knowinganimals.org/word%20docs/Knowing%20Animals%20Book%20of%20Abstracts. pdf (accessed November 4, 2009). CCAC (Canadian Council on Animal Care). 1998. Guideline on choosing an appropriate endpoint in experiments using animals for research. Carlsson, H. E., F. Royo, S. Faheem, M. Tufvesson, and J. Hau. 2009. Separation of pair housed roosters is associated with transient increased corticosteroid excretion. Research in Veterinary Science 86 (1): 183–187. Chance, M. R. A. 1956. Environmental factors influencing gonadotrophin assay in the rat. Nature 177:228–229. ———. 1957. The contribution of environment to uniformity. Collected Papers—Laboratory Animals Bureau 6:59–73. Claassen, V. 1994. Neglected factors in pharmacology and neuroscience research: Biopharmaceutics, animal characteristics, maintenance, testing conditions. In Techniques in the behavioral and neural sciences, vol. 12, series ed. J. P. Huston. New York: Elsevier. Costa, P. 1997. Production of transgenic animals: Practical problems and welfare aspects. In Welfare aspects of transgenic animals, ed. L. F. M. van Zutphen and M. van der Meer, 68–77. Heidelberg: Springer–Verlag.
84550.indb 565
11/11/10 6:28:22 PM
566
Handbook of Laboratory Animal Science
Cussler, K., D. B. Morton, and C. F. M. Hendriksen. 1998. Humane endpoints in vaccine research and quality control. In Humane Endpoints in Animal Experiments for Biomedical Research, Proceedings of the International Conference, November 22–25, 1998, Zeist, the Netherlands, ed. C. F. M. Hendriksen and D. B. Morton. Royal Society of Medicine, London, 1999, pp. 95–101. Damon, E. G., A. F. Eidson, C. H. Hobbs, and F. F. Hahn. 1986. Effect of acclimation on nephrotoxic response of rats to uranium. Laboratory Animal Science 36:24–27. Davies, H., and D. Balfour. 1992. The inevitable bond: Examining scientist–animal interactions. London: Cambridge University Press. Dawkins, M. S. 1992. Animal suffering: The science of animal welfare, 2nd ed. London: Chapman & Hall. ———. 1993. Through our eyes only. The search for animal consciousness. San Francisco: W. H. Freeman. DeGrazia, D. 1996. Taking animals seriously: Mental life and moral status. London: Cambridge University Press. Dennis, M. B., Jr. 2000. Humane endpoints for genetically engineered animal models. ILAR Journal 41:2, 94–98. Denton, D. A. 1993. The pinnacle of life. Consciousness and self-awareness in humans and animals. New South Wales, Australia: Allen and Unwin Pty. Ltd. Diesch, T., D. J. Mellor, C. B. Johnson, and R. G. Lentle. 2007. Responsiveness to painful stimuli in anesthetised newborn and young animals of varying neurological maturity (wallaby joeys, rat pups and lambs). AATEX 14 (special issue) 549–552. Proceedings of the 6th World Congress on Alternatives & Animal Use in the Life Sciences, August 21–25, 2007, Tokyo, Japan. Duncan, I. J. H. 1996. Animal welfare defined in terms of feelings. Acta Agriculturae Scandinavica Section A, Animal Science (Suppl. 27): 29–35. EFSA. 2005. Aspects of the biology and welfare of animals used for experimental and other scientific purposes. http://www.efsa.europa.eu/en/science/ahaw/ahaw_opinions/1286.html. ———. 2008. Food safety, animal health and welfare and environmental impact of animals [1] derived from cloning by somatic cell nucleus transfer (SCNT) and their offspring and products obtained from those animals [2] http://www.efsa.europa.eu/EFSA/efsa_locale-1178620753812_1211902019540.htm. Eriksson, E., F. Royo, K. Lyberg, H. E. Carlsson, and J. Hau. 2004. Effect of metabolic cage housing on immunoglobulin A and corticosterone excretion in feces and urine of young male rats. Experimental Physiology 89 (4): 427–433. Estevez, I., and M. C. Christman. 2006. Analysis of the movement and use of space of animals in confinement: The effect of sampling effort. Applied Animal Behavior Science 97:221. European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (ETS 123). 2008. Appendix A. FAWC (Farm Animal Welfare Council). 1993. Second report on priorities for research and development in farm animal welfare. Ministry of Agriculture Fisheries and Food. London: Tolworth. ———. 2009. Farm animal welfare in Great Britain: Past, present and future. Farm Animal Welfare Council Area 5A, 9 Millbank, c/o Nobel House, 17 Smith Square, London SW1P 3JR, http://www.fawc.org.uk. FELASA. 2007. Guidelines for the production and nomenclature of transgenic rodents. Laboratory Animals 41:301– 311 (). Fernstrom, A. L., W. Sutian, F. Royo, K. Westlund, T. Nilsson, H. E. Carlsson, Y. Paramastri, J. Pamungkas, D. Sajuhti, S. J. Schapiro, and J. hau. 2008. Stress in cynomolgus monkeys (Macaca fascicularis) subjected to long-distance transport and simulated transport housing conditions. Stress 11 (6): 467–476. Fitzgerald, M. 1994. Neurobiology of fetal and neonatal pain. In Textbook of pain, 3rd ed., ed. P. Wall and R. Melzack, 153–163. London: Churchill Livingstone. Fraser, A. F., and D. M. Broom. 1997. Farm animal behavior and welfare, 3rd ed. Wallingford, Oxon, UK: CAB International. Fraser, D. M. 1995. Science, values and animal welfare: Explaining the inextricable connection. Animal Welfare 4:103–117. Garner, J. P. 2005. Stereotypies and other abnormal repetitive behaviors: Potential impact on validity, reliability, and replicability of scientific outcomes. ILAR Journal 46:2. Gil, M. C., J. A. Aguirre, A. P. Lemoine, E. T. Segura, M. Barontini, and I. Armando. 1999. Influence of age on stress responses to metabolic cage housing in rats. Cell Molecular Neurobiology 19:625–633. Goldkuhl. R., H. E. Carlsson, J. Hau, and K. Abelson. 2008. Effect of oral and subcutaneous buprenorphine treatment on postoperative serum corticosterone levels in male rats. European Surgical Research 41:272.
84550.indb 566
11/11/10 6:28:23 PM
Welfare Assessment and Humane End Points
567
Gomez-Sanchez, E. P., and C. E. Gomez-Sanchez. 1991. The effect of gonadectomy and aromatase inhibition on the excretion of 19-nordeoxycorticosterone in rats. Journal of Steroid Biochemistry and Molecular Biology 39:185–188. Guhad, F., H. Augustsson, and J. Hau. 1998. The Torneby Primate Facility: Optimization of housing conditions for rhesus macaques in Sweden. Scandinavian Journal of Laboratory Animal Science 25:173–176. Guhad, F. A., and J. Hau. 1996. Salivary IgA as a marker of social stress in rats. Neuroscience Letters 27:137–140. Hagelschuer, I., P. Hauff, A. Strube, S. M. Käkönen, and G. von Degenfeld. 2009. Luminescent imaging technology as an opportunity to reduce and refine animal experiments: Light at the end of the tunnel? ALTEX 26 (3): 177–185. Hau, J. 1999. Humane endpoints and the importance of training. In Humane Endpoints in Animal Experiments for Biomedical Research, Proceedings of International Conference, November 22–25, 1998, Zeist, the Netherlands, ed. C. F. M. Hendriksen and D. B. Morton, Royal Society of Medicine London, 1999, pp. 106–113. Hau, J., and J. F. A. Carver. 1994. Refinement in laboratory animal science: Is it a Cinderella subject, and is there conflict and imbalance within the 3Rs? Scandinavian Journal of Laboratory Animal Science 21:161–167. Hau, J., O. Kalliokoskim, G. Ostergaard, et al. 2009. Implementation of self-administered oral analgesic treatment in rats subjected to invasive procedures. Journal of American Association for Laboratory Animal Science 48 (5): 546. Hau, J., and S. J. Schapiro. 2004. The welfare of nonhuman primates. In The welfare of laboratory animals, ed. C. Phillips, vol. 2, ed. E. Kaliste. Dordrecht: Kluwer Academic Publishers. Hawkins, P., D. B. Morton, D. Cameron, et al. 2001. Laboratory birds: Refinements in husbandry and procedures, fifth report of the BVAAWF/FRAME/RSPCA/UFAW Joint Working Group on Refinement. Laboratory Animals 35 (Suppl. 1): 1–163. Heidbreder, C. A., I. C. Weiss, A. M. Domeney, C. Pryce, J. Homberg, G. Hedou, J. Feldon, M. C. Moran, and P. Nelson. 2000. Behavioral, neurochemical and endocrinological characterization of the early social isolation syndrome. Neuroscience 100:749–768. Hemsworth, P. H., and G. J. Coleman. 1998. Human–livestock interactions: The stockperson and the productivity and welfare of intensively farmed animals. Wallingford, Oxon, UK: CAB International. Hendriksen, C. F. M., and D. B. Morton, eds. 1999. Humane Endpoints in Animal Experiments for Biomedical Research, Proceedings of International Conference November 22–25, 1998, Zeist, the Netherlands, Royal Society of Medicine, London, 1999, p. 150. Hess, S. E., S. Rohr, B. D. Dufour, B. N. Gaskill, E. A. Pajor, and J. P. Garner. 2008. Home improvement: C57BL/6J mice given more naturalistic nesting materials build better nests. Journal of the American Association for Laboratory Animal Science 47 (6): 25–31. Honess, P. E., P. J. Johnson, and S. E. Wolfensohn. 2004. A study of behavioral responses of nonhuman primates to air transport and rehousing. Lab Anim 38 (2): 119–132. Hubrecht, R. C. 1995. Genetically modified animals, welfare and UK legislation. Animal Welfare 4:163–170. Hudson, M. 2005. The welfare and scientific advantages of noninvasive imaging of animals in biomedical research. Animal Welfare 14:303–319. HSUS. 2001. www.hsus.org. Irwin, J., P. Ahluwalia, R. M. Zacharko, and H. Anisman. 1986. Central norepinephrine and plasma corticosterone following acute and chronic stressors: influence of social isolation and handling. Pharmacology and Biochemical Behavior 24:1151. Jacobs, H., N. Smith, P. Smith, L. Smyth, P. Yew, P. Saibaba, and J. Hau. 1995. Zebra finch behavior and effect of modest enrichment of standard cages. Animal Welfare 4:3–9. Jenkins, E. S., and R. F. Francis. 1999. New guidance notes for projects generating or maintaining genetically modified animals. ATLA 27 (Suppl. 1): 883–902. Johannes, S., U. Rosskopf-Streicher, D. Hausleithner, H. Gyra, and K. Cussler. 1998. Use of clinical signs in efficacy testing of erysipelas vaccines. Humane Endpoints in Animal Experiments for Biomedical Research, Proceedings of International Conference, November 22–25, 1998, Zeist, the Netherlands, ed. C. F. M. Hendriksen and D. B. Morton. Royal Society of Medicine, London, 1999, 114–117. JWGR (Joint Working Group on Refinement). 1993. Second report. Refinements in rabbit husbandry. Lab Anim 27:301–329.
84550.indb 567
11/11/10 6:28:23 PM
568
Handbook of Laboratory Animal Science
———. 1998. Third report. Refining rodent husbandry: The mouse. Lab Anim 32 (3): 260–269. ———. 2003a. Seventh report. Refinements in telemetry procedures. Part A. Lab Anim 37:261–299. ———. 2003b. The welfare of genetically modified rodents: Refinements in their generation, management and care. Lab Anim 37:1–51. Kagira, J. M., M. Ngotho, J. K. Thuita, N. W. Maina, and J. Hau. 2007. Hematological changes in vervet monkeys (Chlorocebus aethiops) during eight months’ adaptation to captivity. American Journal of Primatology 69 (9):1053–1063. Kalliokoski, O., K. R. Jacobsen, J. Hau, and K. S. Abelson. 2009. Serum concentrations of buprenorphine after oral and parenteral administration in male mice. Veterinary Journal Dec. 30, 2009. Kasanen, I. H., K. J. Inhilä, O. M. Vainio, V. V. Kiviniemi, J. Hau, M. Scheinin, S. M. Mering, and T. O. Nevalainen. 2009. The diet board: Welfare impacts of a novel method of dietary restriction in laboratory rats. Lab Anim 43:215. Kempermann, G., H. G. Kuhn, and F. H. Gage. 1997. More hippocampal neurones in adult mice living in an enriched environment. Nature 386:493–495. Kemppinen, N., J. Hau, A. Meller, K. Mauranen, T. Kohila, and T. Nevalainen. 2009. Impact of aspen furniture and restricted feeding on activity, blood pressure, heart rate and fecal corticosterone and immunoglobulin A excretion in rats (Rattus norvegicus) housed in individually ventilated cages. Lab Anim Oct 23 [Epub ahead of print]. Kirkden, R. D., L. Neil, and D. M. Weary. 2005a. Aversion to carbon dioxide. Lab Anim 39:453–455. ———. 2005b.How aversive is gradual fill carbon dioxide euthanasia for rats? CALAS/ACSAL, 44th Annual Symposium, Vancouver, BC, June 25–28, p. 31. Krohn, T. C., A. K. Hansen, and N. Dragsted. 2003. Telemetry as a method for measuring the impact of housing conditions on rats’ welfare. Animal Welfare 12:53–62. Krug, M., and K. Cussler. 1999. Endotoxin in porcine vaccines: Clinical signs and safety aspects. Humane Endpoints in Animal Experiments for Biomedical Research, Proceedings of International Conference, November 22–25, 1998, Zeist, the Netherlands, ed. C. F. M. Hendriksen and D. B. Morton, Royal Society of Medicine, London, 1999, pp. 114–117. Kuijpers, M. H. M., and H. C. Walvoort. 1991. Discomfort and distress in rodents during chronic studies. In Animals in biomedical research, ed. C. F. M. Hendriksen and H. W. B. M. Koeter, chap. 18, 247–263, 281. New York: Elsevier. Lambeth, S. P., J. Hau, J. E. Perlman, M. A. Martino, and S. J. Schapiro. 2006. Positive reinforcement training affects hematologic and serum chemistry values in captive chimpanzees (Pan troglodytes). American Journal of Primatology 68 (3): 245–256. Lancet. 1993. Editorial: A predator’s compassion. Lancet 343:1311–1312. Laule, G. E., M. A. Bloomsmith, and S. J. Schapiro. 2003. The use pf positive reinforcement training techniques to enhance the care, management, and welfare of primates in the laboratory. Journal of Applied Animal Welfare Science 6:163–173. Lawrence, A. B., and J. Rushen, eds. 1993. Stereotypic animal behavior, fundamentals and applications to welfare, chap. 2, pp. 7–40. Wallingford, Oxon, UK: CAB International. Leach, M., V. Bowell, T. Allan, and D. B. Morton. 2002a. Aversion to gaseous euthanasia in rats and mice. Comparative Medicine 52 (3): 249–257. ———. 2002b. Aversion to various concentrations of different inhalational general anesthetics in rats and mice. Veterinary Record 150:808–815. Lee, S. J., H. J. Ralston, E. A. Drey, J. C. Partridge, and M. A. Rosen. Fetal pain: A systematic multidisciplinary review of the evidence. JAMA 294:947. Main, D. C. J., J. P. Kent, F. Wemelsfelder, E. Ofner, and F. A. M. Tuyttens. 2003. Applications for methods of on-farm welfare assessment. Animal Welfare 12 (4): 523–528. Mason, G. J. 1993. Forms of stereotypic behavior. Stereotypic animal behavior; fundamentals and applications to welfare, ed. A. B. Lawrence and J. Rushen, chap. 2, 7–40. Wallingford, Oxon, UK: CAB International. Mellor, D. J., T. J. Diesch, A. J. Gunn, and L. Bennet. 2005. The importance of “awareness” for understanding fetal pain. Brain Research Reviews 49:455. Mellor, D. J., and N. G. Gregory. 2003. Responsiveness, behavioral arousal and awareness in fetal and newborn lambs: Experimental, practical and therapeutic implications. New Zealand Veterinary Journal 51:2–13.
84550.indb 568
11/11/10 6:28:23 PM
Welfare Assessment and Humane End Points
569
Mepham, T. B., R. D. Coombes, M. Balls, et al. 1998. The use of transgenic animals in the European Union. The report and recommendations of ECVAM Workshop 28. ATLA 26:21–43. Moberg, G. P., and J. A. Mench. 2000. The biology of animal stress. Basic principles and implications for animal welfare. Wallingford, Oxon, UK: CABI. Morton, D. B. 1998a. The importance of nonstatistical design in refining animal experimentation. ANZCCART facts sheet. ANZCCART News 11:2. ———. 1998b. Humane end points in animal experimentation for biomedical research: Ethical, legal and practical aspects. In Humane Endpoints in Animal Experiments for Biomedical Research, Proceedings of International Conference, November 22–25, 1998, Zeist, the Netherlands, ed. C. F. M. Hendriksen and D. B. Morton, Royal Society of Medicine, London, 1999, pp. 5–12. ———. 2000a. Self-consciousness and animal suffering. Biologist 47:77–80. ———. 2000b. A systematic approach for establishing humane endpoints. ILAR Journal 41 (2): 80–86. ———. 2007. A hypothetical strategy for the objective evaluation of animal well-being and quality of life using a dog model. Animal Welfare 16(S):75–81. Morton, D. B., G. Burghardt, and J. A. Smith. 1990. Critical anthropomorphism, animal suffering and the ecological context. Hasting’s Center Report Spring Issue on Animals, Science and Ethics 20 (3): 13–19. NRC (National Research Council). 1996. Guide for the care and use of laboratory animals. Washington, D.C.: National Academy Press. Morton, D. B., and P. H. M. Griffiths. 1985. Guidelines on the recognition of pain, distress and discomfort in experimental animals and an hypothesis for assessment. Veterinary Record 116:431–436. Morton, D. B., and P. Townsend. 1985. Dealing with adverse effects and suffering during animal research. In Laboratory animals—An introduction for experimenters, rev. version, ed. A. A. Tuffery, 215–231. New York: John Wiley & Sons. Niel, L., S. A. Stewart, and D. M. Weary. 2007. Effect of flow rate on aversion to gradual-fill carbon dioxide exposure in rats. Applied Animal Behavior Science 109:77–84. Niel, L., and D. M. Weary. 2006. Behavioral responses of rats to gradual-fill carbon dioxide euthanasia and reduced oxygen concentrations. Applied Animal Behavior Science 100:295–308. ———. 2007a. Effects of novelty on rats’ responses to CO2 exposure. Applied Animal Behavior Science 111:183–194. ———. 2007b. Rats avoid exposure to carbon dioxide and argon. Applied Animal Behavior Science 107:100–109. OECD (Organization for Economic Cooperation and Development). 2001. Environmental health and safety publications series on testing and assessment no. 19 guidance document on the recognition, assessment, and use of clinical signs as humane endpoints for experimental animals used in safety evaluation, 14. Paramastri, Y., F. Royo, J. Eberova, H. E. Carlsson, D. Sajuthi, A. L. Fernstrom, J. Pamungkas, and J. Hau. 2007. Urinary and fecal immunoglobulin A, cortisol and 11-17 dioxoandrostanes, and serum cortisol in metabolic cage housed female cynomolgus monkeys (Macaca fascicularis). Journal of Medical Primatology 36 (6): 355–364. Poole, T. 1997. Identifying the behavioral needs of zoo mammals and providing appropriate captive environments. RATEL 24 (6): 200–211. Poole, T. B. 1991. Criteria for the provision of captive environments. In Primate responses to environmental change, ed. H. O. Box, 265. London: Chapman & Hall. ———. 1992. The nature and evolution of behavioral needs in mammals. Animal Welfare 1:203–220. ———. 1995. Welfare considerations with regard to transgenic animals. Animal Welfare 4:81–85. ———. 1997. Identifying the behavioral needs of zoo mammals and providing appropriate captive environments. RATEL 24 (6): 200–211. ———. 1998. Meeting a mammal’s psychological needs: Basic principles. In Second nature—Environmental enrichment for captive animals, ed. D. J. Shepherdson, J. D. Mellen, and M. Hutchins, 83–97. Washington, D.C.: Smithsonian Institution Press. Prior, H., A. Schwarz, and O. Güntürkün. 2008. Mirror-induced behavior in the magpie (Pica pica): Evidence of self-recognition. PLoS Biology 6 (8): e202. Pritchett-Corning, K. R., F. T. Chang, and M. F. Festing. 2009. Breeding and housing laboratory rats and mice in the same room does not affect the growth or reproduction of either species. Journal of American Association of Laboratory Animal Science 48:492.
84550.indb 569
11/11/10 6:28:23 PM
570
Handbook of Laboratory Animal Science
Reinhardt, V. 1987. Advantages of housing rhesus monkeys in compatible pairs. Scientists Center for Animal Welfare Newsletter 9 (3): 3–6. ———. 1989. Behavioral responses of unrelated adult male rhesus monkeys familiarized and paired for the purpose of environmental enrichment. American Journal of Primatology 17:243–248. ———. 1994. Pair-housing rather than single-housing for laboratory rhesus macaques. Journal of Medical Primatology 23:426–431. ———. 2002. Comfortable quarters for primates in research institutions. In Comfortable quarters for laboratory animals, 9th ed. Washington, D.C: Animal Welfare Institute. Richter, S. H., J. P. Garner, and H. Wurbel. 2009. Environmental standardization: Cure or cause of poor reproducibility in animal experiments? Nature Methods 6 (4): 257. Riley, V. 1981. Psychoneuroendocrine influences on immunocompetence and neoplasia. Science 212:1100–1109. Robinson, V., D. B. Morton, D. Anderson, J. F. A. Carver, R. J. Francis, R. Hubrecht, E. Jenkins, K. E. Mathers, R. Raymond, I. Rosewell, et al. 2003. Refinement and reduction in the production of genetically modified mice. Lab Anim 37 (Suppl. 1): 1–49. Rogers, D. C., E. M. Fischer, S. D. Brown, J. Peters, A. J. Hunter, and J. E. Martin. 1997. Behavioral and functional analysis of mouse phenotype: SHIRPA, a proposed protocol for comprehensive phenotype assessment. Mammalian Genome 10:711–713. Rose, M. A. 1994. Environmental factors likely to impact on an animal’s well-being—An overview. Improving the Well-Being of Animals in the Research Environment, Proceedings of the ANZCCART Conference, ed. R. M. Baker et al., 99–116. Ross, S. R., S. J. Schapiro, J. Hau, and K. E. Lukas. 2009. Space use as an indicator of enclosure appropriateness: A novel measure of captive animal welfare. Applied Animal Behavior Science 121:42. Roughan, J. V., and P. A. Flecknell. 2001. Behavioral effects of laparotomy and analgesic effects of ketoprofen and carprofen in rats. Pain 90:1–2, 65–74. Royo, F., K. Lyberg, K. Abelson, H. E. Carlsson, and J. Hau. 2005. Effect of repeated confined single housing of young pigs on fecal excretion of cortisol and IgA. Scandinavian Journal of Laboratory Animal Science 32 (1): 33–37. Russell, W. M. S. 1994. Enhancing animal comfort in the laboratory. Humane Innovations and Alternatives 8: 587–597. Russell, W. M. S., and R. L. Burch. 1959. The principles of humane experimental technique. London: Methuen. UFAW, special edition, 1992. Saibaba, P., G. D. Sales, G. Stodulski, and J. Hau. 1996. Behavior of rats in their home cage: Daytime variations and effects of routine husbandry procedures analyzed by time sampling techniques. Lab Anim 30:13–21. Schapiro, S. J., M. A. Bloomsmith, and G. E. Laule. 2003. Positive reinforcement training as a technique to alter nonhuman primate behavior: Quantitative assessment of effectiveness. Journal of Applied Animal Welfare Science 6:175–187. Schlede, E., W. Diener, and I. Gerner. 1999. Humane endpoints in toxicology. In Humane Endpoints in Animal Experimentation for Biomedical Research, Proceedings of International Conference, November 22–25, 1998, Zeist, the Netherlands, ed. C. F. M. Hendriksen and D. B. Morton, Royal Society of Medicine, London, 1999, pp. 75–78. Schlede, E., U. Mischke, R. Roll, and D. Kayser. 1992. A national validation study of the acute-toxic-class method—An alternative to the LD50 test. Archives of Toxicology 66:455–470. Sherwin, C. 2002. Comfortable quarters for mice in research institutions. In Comfortable quarters for laboratory animals, 9th ed., ed. V. Reinhardt and A. Reinhardt, 6–17. Washington, D.C.: Animal Welfare Institute (www.awionline.org/pubs/cq02/Cq-mice.html). Smith, J. A., and K. M. Boyd, eds. 1991. Lives in the balance: The ethics of using animals in biomedical research, the report of a working party of the Institute of Medical Ethics. Oxford, England: Oxford University Press. Soma, L. R. 1987. Assessment of animal pain in experimental animals. Laboratory Animal Science 37:71–74. Soothill, J. S., D. B. Morton, and A. Ahmad. 1992. The HID50 (hypothermia-inducing dose 50): An alternative to the LD50 for measurement of bacterial virulence. International Journal of Experimental Pathology 73:95–98.
84550.indb 570
11/11/10 6:28:23 PM
Welfare Assessment and Humane End Points
571
Spinelli, J. S., and H. Markowitz. 1985. Prevention of cage associated distress. Lab Anim 14:19–24. Stemkens-Sevens S., K. van Berkel, I. de Greeuw, B. Snoeijer, and K. Kramer. 2009. The use of radiotelemetry to assess the time needed to acclimatize guinea pigs following several hours of ground transport. Lab Anim 43 (1): 78–84. Tuli, J. 1993. Stress and parasitic infection in laboratory mice. PhD thesis, University of Birmingham, AL. Tuli, J., J. A. Smith, and D. B. Morton. 1995. Stress measurements in mice after transportation. Lab Anim 29:132–138. van den Heuvel, M. J., D. G. Clark, R. J. Fielder, P. P. Koundakjian, G. J. A. Oliver, D. Pelling, N. J. Tomlinson, and A. P. Walker. 1990. The international validation of a fixed dose procedure as an alternative to the classical LD50 test. Food and Chemical Toxicology 28:469–482. van der Meer, M., V. Baumans, B. Olivier, et al. 2001. Behavioral and physiological effects of biotechnology procedures used for gene targeting in mice. Physiology and Behavior 73 (5): 719–730. van der Meer, M., V. Baumans, and L. F. M. Van Zutphen. 1997. Measuring welfare aspects of transgenic animals. In Animal alternatives, welfare and ethics, L. F. M. Van Zutphen and M. Balls, 229–233. New York: Elsevier. van der Meer, M., and L. F. M. Van Zutphen. 1997. Use of transgenic animals and welfare implications. Welfare aspects of transgenic animals, ed. L. F. M. Van Zutphen and M. van der Meer, 78–89. Heidelberg: Springer–Verlag. van de Weerd, H., V. Baumans, J. Koolhaus, and L. Van Zutphen. 1994. Strain-specific behavioral response to environmental enrichment in the mouse. Journal of Experimental Animal Science 36:117–127. van de Weerd, H. A., F. A. R. van de Broek, and V. Baumans. 1996. Preference for different types of flooring in two rat strains. Applied Animal Behavior Science 46:3–4, 251–261. van de Weerd, H. A., P. L. P. Van Loo, L. F. M. Van Zutphen, J. M. Koolhaas, and V. Baumans. 1997a. Nesting material as environmental enrichment has no adverse effects on behavior and physiology of laboratory mice. Physiology and Behavior 62 (5): 1019–1028. ———. 1997b. Preferences for nesting material as environmental enrichment for laboratory mice. Lab Anim 31:133–143. ———. 1998a. Strength of preference for nesting material as environmental enrichment in laboratory mice. Applied Animal Behavior Science 55:369–382. ———. 1998b. Preferences for nest boxes as environmental enrichment for laboratory mice. Animal Welfare 7:11–25. van Loo, P. L. P., A. C. de Groot, L. F. M. Van Zutphen, and V. Baumans. 2002. Do male mice prefer or avoid each other’s company? Influence of hierarchy, kinship and familiarity. Journal of Applied Animal Welfare Science 4 (2): 91–103. van Loo, P. L. P., C. L. J. J. Kruitwagen, L. F. M. Van Zutphen, J. M. Koolhaas, and V. Baumans. 2000. Modulation of aggression in male mice: Influence of cage cleaning regime and scent marks. Animal Welfare 9:281–295. van Loo, P. L. P., J. A. Mol, J. M. Koolhaas, L. F. M. Van Zutphen, and V. Baumans. 2001. Modulation of aggression in male mice: Influence of group size and cage size. Physiology and Behavior 72:675–683. van Praag, H., G. Kempermann, and F. H. Gage. 2000. Neural consequences of environmental enrichment. Nature Reviews Neuroscience 1:191–198. Van Zutphen, L. F. M. 1998. Refinement, the Cinderella of the 3Rs. In The ethics of animal experimentation, London: European Biomedical Research Association. Wadham, J. J. B. 1996. Recognition and reduction of adverse effects in research on rodents. PhD thesis, University of Birmingham, AL. Webster, A. J. F. 1998. What use is science to animal welfare? Naturwissenschaften 85 (6): 262–269. Webster, J. 1995. Animal welfare. A cool eye towards Eden, 163. Oxford, England: Blackwell Scientific. Wemelsfelder, F. 1993. The concept of animal boredom and its relationship to stereotyped behavior. In Stereotypic animal behavior, fundamentals and applications to welfare, ed. A. B. Lawrence and J. Rushen, chap. 4, 65–95. Wallingford, Oxon, UK: CAB International. Willner, P., R. Muscat, and M. Papp. 1992. Chronic mild stress-induced anhedonia: A realistic animal model of depression. Neuroscience and Biobehavioral Reviews 16:525–534 Wurbel, H. 2001. Ideal homes? Housing effects on rodent brain and behavior. Trends in Neurosciences 24:207–211.
84550.indb 571
11/11/10 6:28:23 PM
572
Handbook of Laboratory Animal Science
Wurbel, H., and J. P. Garner. 2007. Refinement of rodent research through environmental enrichment and systematic randomization. NC3Rs 9:1–9. Yeates, J. W., and D. C. J. Main. 2008. Assessment of positive welfare: A review. Veterinary Journal 175:293–300. Young, L. A., G. Pavlovska-Teglia, G. Stodulski, and J. Hau. 1996. Effect of oral administration of corticosterone and group housing on body weight gain and locomotor development in the neonatal rat. Animal Welfare 5:167–176.
Relevant Reference Sites on the World Wide Web AWIC Animal Welfare Information Center http://www.nal.usda.gov/awic/ CCAC guidelines on HEPs and other issues http://www.ccac.ca/english/gdlines/endpts/appopen.htm Various papers on refinement including JWGR reports http://www.lal.org.uk/ OECD guidance http://www.oecd.org/ehs/test/mono19.pdf ALTWEB http://altweb.jhsph.edu/news/news.htm Center for Alternatives—good links for refinement http://caat.jhsph.edu/ Interniche for alternatives to the use of animals in teaching http://www.interniche.org/alt_loan.html NORINA for alternatives to the use of animals in teaching and other refinement papers http://oslovet.veths.no/teaching/materials.html Training programs http://dcminfo.wustl.edu/education/models.html Refinement and enrichment http://www.animalwelfare.com/lab_animals/rhesus/pho1–10.htm http://www.awionline.org/lab_animals/biblio/index.html SHIRPA protocol and other protocols http://www.mgu.har.mrc.ac.uk/MGU-welcome.html http://www.bzl.unizh.ch/de/database/formtransg/index2.html
84550.indb 572
11/11/10 6:28:23 PM
Chapter 19
Surgery Basic Principles and Procedures
M. Michael Swindle, Heather Elliott, and Alison C. Smith Contents Introduction..................................................................................................................................... 574 Planning for Surgery....................................................................................................................... 574 Preparing for Surgery..................................................................................................................... 575 Surgical Facilities and Equipment............................................................................................. 575 Large Animals....................................................................................................................... 575 Rodents.................................................................................................................................. 575 Surgical Instruments.................................................................................................................. 577 Large Animals....................................................................................................................... 577 Rodents.................................................................................................................................. 578 Care and Preparation of the Animal before Surgery...................................................................... 578 Large Animals........................................................................................................................... 578 Rodents....................................................................................................................................... 579 Care of the Animal during Surgery................................................................................................ 580 Preparation of the Surgical Team............................................................................................... 580 Aseptic Surgery.......................................................................................................................... 581 Handling of Tissues and Instruments........................................................................................ 581 Wound Closure and Suturing..................................................................................................... 581 Approaches to Surgery.................................................................................................................... 582 Laparotomy................................................................................................................................ 582 Thoracotomy.............................................................................................................................. 585 Catheterization/Cannulation...................................................................................................... 586 Craniotomy................................................................................................................................. 586 Postoperative Care.......................................................................................................................... 590 Large Animals........................................................................................................................... 590 The Recovery Environment.................................................................................................. 590 Monitoring and Nursing Care............................................................................................... 591 Rodents....................................................................................................................................... 592 Warmth and Comfort............................................................................................................ 592 Respiratory Depression......................................................................................................... 592 573
84550.indb 573
11/11/10 6:28:24 PM
574
Handbook of Laboratory Animal Science
Fluid Therapy........................................................................................................................ 592 General Care......................................................................................................................... 593 Postoperative Pain........................................................................................................................... 593 References....................................................................................................................................... 593 Introduction Surgery is a particularly invasive procedure and could possibly result in pain and distress if not managed properly. It is an absolute requirement of the surgeon or investigator to minimize any pain or infection by the careful use of analgesics or antibiotics in consultation with the attending veterinarian and to be considerate of the animal’s welfare at all times. To minimize adverse effects, it is important that the surgeon have access to a well-organized and well-equipped surgical facility and that he or she use proper surgical technique. This includes an understanding of the animal’s physiology and anatomy in addition to using excellent surgical technique. In particular, the use of asepsis in both large animals and rodents is important. Excellence in surgery is gained by studying the relevant literature, by judicious practice on cadavers or other appropriate models, and, importantly, by initially seeking the help of a colleague experienced in the technique. This chapter provides a brief outline of some of the important principles of surgery. The reader who requires more detailed information on these subjects should consult more comprehensive texts (De Boer, Archibald, and Downie 1975; Dougherty 1981; Hecker 1974; Lumley, Green, and Angell-James 1990; Knecht et al. 1981; Swindle and Adams 1988; Slatter 1985; Waynforth and Flecknell 1992; Waynforth 1995; Bradfield et al. 1992; British Laboratory Veterinary Association 1992; Cunliffe-Beamer 1983, 1989; Green and Simpkin 1990; Lambert 1982; Petty 1982; Van Dongen et al. 1990; Swindle 1983, 1988, 2007; Swindle, Smith, and Goodrich 1998). Planning for Surgery To help minimize adverse effects on the animal, it is important to ensure that all aspects of the surgery are thoroughly planned and understood by everyone who will be involved, however simple the procedure. It is useful to prepare a checklist of requirements, which will be useful not only to the surgeon, but also to ancillary personnel who may be assisting. Assistants will usually be involved in large animal surgery, but the value of an assistant for rodent surgery is often underestimated. Such a person will be able to help in maintaining a sterile surgical field by taking on all the “dirty” tasks, such as preparing the animal beforehand, and can also perform other useful jobs to support the surgeon. A typical checklist might include the following: • • • • • • • •
Choice and availability of the animal or animals Acclimatization and any preoperative habituation or training of the animal Preoperative evaluation of the animal’s health Provision of surgical and pre- and postoperative facilities Choice, provision, and sterilization of surgical instruments, equipment, and ancillary supplies Number of assistants required Preparation of the animal for surgery Management of the animal postoperatively and in an emergency and any ongoing long-term observations and care • Written protocol of the surgical and anesthetic procedures • Experience of the surgeon, to include adequate practice in handling instruments and tissues and in the surgical procedure • Record keeping
84550.indb 574
11/11/10 6:28:24 PM
Surgery
575
An important consideration is when to carry out the operation. Surgery done late in the day or at the end of the week may require that postoperative care continue well into the evening or weekend. This could be inconvenient and lead to insufficient care being provided if an adequate number of staff members is not available.
Preparing for Surgery Surgical Facilities and Equipment Large Animals At least three major components need to be supplied for facilities for aseptic survival surgery in large animals: an operating room, an animal preparation room, and a surgeon preparation room. These three areas need to be separate and to have controlled circulation. Other ancillary areas should include space for storage, animal recovery, and instrument preparation. The animal preparation room and the surgeon preparation room may be used for some of these functions. Floors, walls, and ceilings should be impervious to moisture and easily sanitized. Operating room lights should be suspended from the ceiling. It is preferable to have directly piped gases such as oxygen and a vacuum for evacuation and suction. At the least, equipment for gas anesthesia, electrocautery, body temperature, and ECG monitoring should be available. Other equipment useful for major procedures includes supplies for intravenous fluid administration, pulse oximetry, blood pressure monitoring, blood gas analysis, and monitoring for exhaled CO2. Intensive care units and cardiopulmonary emergency kits should also be available for major procedures. A floor plan of a suitable operating theater and animal facility is shown in Figure€19.1. Rodents Surgery on rodents should only be carried out in a dedicated area. Ideally, this should be a facility separate from other activities. For rodents, the most usual arrangement for survival surgery involves a single large room. A portion of the room is used for anesthetizing and preparing the animal and the surgeon and another portion is used for the surgery; a small part is reserved for closely monitoring the animal during the initial stages of recovery before returning it to its home cage. A room showing a surgical station and areas for animal preparation and recovery is shown in Figure€19.2. This single room arrangement is suitable when space is at a premium; however, ideally, as for large animals, each of the functions is best carried out in separate rooms (if available) so that microbiological contamination and exposure of conscious animals to potential stressors can be kept to a minimum. Consideration should also be given to providing an area for holding animals immediately before or after surgery close to, but not within, the surgical room. The surgical room or rooms must be kept clean by the judicious use of disinfectants and must be adequately ventilated with an appropriate system to scavenge waste anesthetic gases. A good source of surgical lighting is important, both in the room and as a locally directed cold light. The ability to sterilize instruments and other supplies, such as surgical gowns and drapes, is also necessary. Surgery is normally performed on a table or bench. The animal should be placed on a thermostatically controlled heat blanket with a rectal probe to maintain and monitor its body temperature. The blanket and bench are covered with a sterile drape to maintain the sterility of the instruments subsequently placed on them. Some surgeons prefer to place a cork or rubber board on the bench
84550.indb 575
11/11/10 6:28:24 PM
84550.indb 576
Storage
Non Survival OR OR
Eqpt. Rm. Surgeon Prep
Animal Room
Animal Prep
Locker Room
Locker Room
Animal Room
OR
Instrument Prep
X Ray
Schematic Floor Plan
Surgeon Prep
Animal Room
Storage
Imaging Room
Diet Prep
Animal Room
View Room
Equipment
Cath Lab
Eqpt
Procedure Room
Storage Procedure Room
Animal Room
Histology
Necropsy
Animal Room
Diagnostic Lab
Hospital Room
Animal Room
Figure 19.1â•…Floor plan of an operating theater and adjacent animal housing and support areas. (Courtesy of the Medical University of South Carolina.)
LAB
ICU
ICU
Office
Office
Animal Room
Animal Room
576 Handbook of Laboratory Animal Science
11/11/10 6:28:26 PM
Surgery
577
Figure 19.2â•…Rodent operating room. The assistant is removing the hair from the rat over the surgical site prior to the operation. In the far right-hand corner is an animal incubator for the animal to recover in.
underneath the drape; this allows pins and similar instruments to be stuck into it to secure ties and aid positioning. Surgical Instruments Large Animals The surgical instruments required will vary with the procedure being performed. A general surgery pack should include curved and straight hemostats of various sizes, a scalpel handle, Allis tissue forceps, scissors of various sizes, forceps, needle holders, stainless steel bowls, an instrument tray, and both hand-held and self-retaining retractors. The recommended number and sizes will vary widely with the species and the procedure to be performed (Figure€19.3). As a general rule, more instruments should be included in the pack than the surgeon anticipates using. Consultation with a professional familiar with surgical procedures should be sought before buying the instruments (Swindle 1983; College of Animal Welfare 1997; Hurov 1978).
Figure 19.3â•…Example of surgical instruments required for general surgery in large animals.
84550.indb 577
11/11/10 6:28:27 PM
578
Handbook of Laboratory Animal Science
Instruments should be thoroughly cleaned between procedures to remove blood and tissue. They should be packed in cloth or other suitable material designed to withstand autoclaving and then placed within a metal drum or autoclave bag. After they have been autoclaved in the appropriate manner, instruments should be stored in a clean, dry environment. If the integrity of the pack is broken, then all instruments must be autoclaved again. Rodents For rodent surgery, instruments are generally selected from those used in human pediatric, ophthalmic, and neurological surgery because of their small size (Waynforth and Flecknell 1992; College of Animal Welfare 1997; Hurov 1978). A basic set of instruments might consist of a size 3 scalpel handle or disposable scalpel with a number 10 blade, a pair of blunt-ended Mayo scissors, a pair of pointed scissors, two pairs of straight or curved medium-fine serrated dissecting forceps, and a similar pair of rat-toothed forceps. A more satisfactory (though more expensive) alternative is to use De Bakey forceps with their fine, relatively atraumatic tips, which provide a firm grip on tissue without causing trauma. Other instruments and accessories must be added to this basic set if a specific procedure requires them. Such instruments could include microsurgical scissors and forceps, eyelid retractors, hemostatic forceps, bulldog clips, neurological clips, and/or stainless steel screws. In addition, accessories such as cotton wool buds, drapes, gauze swabs, dental cement, plastic cannulae or other implants, or steel trocar needles may be needed. Instruments to close wounds will be essential, including needle holders, needles, and suture materials of various types. For closure of the skin, automatically applied stainless steel staples or Michel clips together with a suitable applicator are popular alternatives in appropriate circumstances. On occasion, additional equipment may be necessary to facilitate surgery. For example, a stereotaxic frame and drill will be required to carry out some craniotomy procedures. For microsurgery, a manual or foot-operated zoom microscope is particularly valuable. The particular surgical procedure being performed will dictate the type of specialist apparatus required. All surgical instruments and accessories to be handled by the surgeon need to be sterile before use. Sterilization is achieved using steam, heat, irradiation, or ethylene oxide gas (Eshleman 1968). The method chosen will generally depend on the material to be sterilized, safety considerations, and local availability. Some instruments and accessories, such as needles, sutures, surgical gloves, drapes, scalpels, and blades, can be obtained already sterilized from commercial sources. The instruments and accessories required for the surgical procedure should be prepackaged and, if appropriate, the complete pack should be sterilized. Each pack is then opened, preferably by the assistant, in a manner that will keep the contents sterile, and the contents are placed on a sterile drape in an orderly fashion. They are now ready for the surgeon to use (Figure€19.4). Care and Preparation of the Animal before Surgery Large Animals Before any surgery is performed, animals should be fasted for approximately 12 hours, depending on species, to prevent vomiting or regurgitation. Some abdominal procedures may require longer fasting periods. In most cases, water does not have to be restricted until a few hours before surgery. This can help prevent dehydration. For prolonged fasts, such as those that may be required for some lower gastrointestinal procedures, use of oral electrolyte/glucose solutions should be considered to maintain the animal’s hydration and to prevent hypoglycemia.
84550.indb 578
11/11/10 6:28:27 PM
Surgery
579
Figure 19.4â•…Arrangement of instruments placed on a sterile drape ready for surgery. Note the drape covering the whole animal with only the surgical site over the head being exposed.
No agent completely sterilizes the skin without damaging it. Thus, the agents used for skin preparation reduce the microflora to a level that minimizes danger from infection (Swindle 1988; Collins et al. 1981; Ghosh, Maisels, and Woodcock 1967). The animal should be clipped and/or shaved to remove all hair over an area some distance from the site of the surgical incision to ensure that unprepared areas cannot contaminate the incision. Antiseptic soap solutions, such as those containing iodine or chlorhexidine, are then used to scrub the area, starting along the surgical incision site and working outward in a circular fashion. This should be repeated three times. The soap can be removed with alcohol in the same manner. Some surgeons prefer then to wipe the area with iodine solution. An assistant should perform the final surgical preparation of the site within the operating room with sterile instruments and gauze sponges after donning sterile surgical garb. Iodine-impregnated adhesive drapes can also be applied to the dry scrubbed skin. The whole animal should be covered with a final sterile drape over all surfaces, including the head and feet. Figure€ 19.5 shows a pig draped for aseptic surgery. Rodents Rodents should be acclimated to their home environment for up to 7 days and must be healthy prior to undergoing surgery. There is no need to fast the animal before surgery, unless the procedure requires the gastrointestinal tract to be empty, since rodents are incapable of vomiting. Once the animal is anesthetized, the site of the incision and a small area around it are prepared. First, the hair is removed using small fine clippers. Then the area is swabbed with a soap-based antiseptic cleaner, followed by an antiseptic (0.5% alcoholic chlorhexidine or 10% alcoholic or aqueous povidone-iodine).
84550.indb 579
11/11/10 6:28:28 PM
580
Handbook of Laboratory Animal Science
Figure 19.5â•…Animal and surgeons appropriately draped and dressed for aseptic surgery.
Excess antiseptic must be removed with a sterile gauze pad to prevent inappropriate cooling of the surgical area by evaporation. The animal should then be covered with a sterile drape, even for the smallest incision. A small hole is cut in the drape to correspond to the incision area. In some instances, the drape may need to be secured to prevent it from slipping. Generally, the drape will cover the whole animal; when cloth drapes are used, it will be difficult to see that respiration is continuing normally. Therefore, respiration should be checked frequently and transparent or semitransparent plastic drapes should be used when possible. Drapes should be positioned such that instruments can remain sterile when placed next to the animal and on the drape (Figure€19.4). For surgical procedures that have the potential for contamination with a large microbiological load, an appropriate antibiotic should be injected prior to surgery in consultation with the attending veterinarian. Presurgical administration of the antibiotic will allow it to achieve its peak effect during surgery, when contamination is most likely (Morris 1995). Care of the Animal during Surgery Preparation of the Surgical Team The surgical team should wear scrub suits that are only used in the operating theater. Shoes that are dedicated to the area and that can be disinfected or shoe covers should be worn. A clean surgical cap and mask should be placed on the surgeon prior to the surgical scrub of the hands and forearms. The same principles regarding the use of disinfectant agents for the animal’s skin (discussed previously) apply to the surgeon’s skin. Using a sterile brush, the fingers and hands and then the forearms are thoroughly scrubbed and rinsed with disinfectant soap and running water. The hands are kept higher than the elbows to prevent water and soap from flowing back onto the fingers. After the scrub is performed, the surgeon dons a sterile surgical gown, tied at the back by an assistant. The surgeon’s fingers do not extend beyond the cuff of the sleeve until the sterile gloves are donned. A sterile assistant may hold the gloves and the surgeon then thrusts each hand into an open glove. Alternatively, the surgeon may manipulate his or her hand into a glove after placing it on the forearm with the fingers of the glove pointed toward the shoulder. The outsides of the gloves
84550.indb 580
11/11/10 6:28:28 PM
Surgery
581
should never be touched by bare fingers during this procedure. Once the hands are covered and the glove is over the cuff of the sleeve, the surgeon may use the gloved hands to fit the gloves properly over the fingers. Once the surgeon is gowned and gloved, the hands and arms are never held below the waist (Swindle 1983). Aseptic Surgery During the surgical procedure, the surgeon should never touch a nonsterile object (Swindle 1983; Strachan and Wise 1979). The hands and forearms should be kept over the draped area of the animal and the instrument tray, opened by an assistant. For major procedures, the surgical team generally includes a surgeon, an assistant surgeon, and an instrument nurse, in addition to an anesthetist. An operating room assistant may open packs, sutures, and other required supplies onto the sterile tray. The outer cover of any package being opened should never touch the sterile field or the surgical team. Handling of Tissues and Instruments Tissues should be handled using the principles set forth by Halsted over a century ago (Swindle 1983; Morris 1995). The principles of surgery are asepsis, gentle tissue handling, hemostasis, closure of dead space, careful approximation of tissues, avoiding tension, and minimizing foreign materials. Prevention of sepsis by using aseptic technique is far more reliable than using antibiotics to try to cover breaks in asepsis. When antibiotics are indicated during a procedure, the single most important dose is one that results in a blood level of the antibiotic at the time the skin incision is made (Swindle 1983; Morris 1995). Control of hemorrhage with electrocautery or ligatures is important to prevent hypovolemia as well as prevention of hematomas and seromas. Likewise, closure of dead space prevents seromas, which can be pockets for infection. Gentle handling of tissues is essential to prevent additional trauma and inflammation. Surgical instruments are designed to prevent undue trauma to the tissues and to aid in gentle handling. The appropriate instrument should be selected when a particular surgical manipulation is performed; this requires surgeons to familiarize themselves with the use of particular instruments prior to performing the procedure. Ringed instruments should be placed at the level of the most distal knuckle of the thumb and ring finger. Instruments can then be stabilized with the two digits between the rings. The scalpel should be held firmly between the tips of the fingers and thumb and the incision made with the belly of the blade, not the tip. The scalpel should be handed between personnel by the distal end of the handle, not near the blade. The tips of the instruments should be used in most cases and little surrounding tissue should be included when the instrument is closed and locked. This prevents trauma and necrosis, except in the tissue or structure of interest. All tissues should be kept moist using either gauze or towels moistened with warm saline. Only the edges of the incision should be exposed in the surgical field and drapes should be placed to prevent bare skin from being exposed to the surgeon. When gauze is used on exposed tissue, abrasive rubbing should be avoided. Instead, the moistened gauze should be used to press and hold against the bleeding edges (Swindle 1983). Wound Closure and Suturing The wound edges should be closed in precise anatomic apposition in layers. It is important to eliminate all dead space during the closure. Sutures should be placed so that the edges are in close approximation, but not tied so tightly as to induce pain and inflammation or to impede local blood supply. This is especially important in the skin. Inner layers of muscle and fascia, as well as vascular ligatures, should be tied tightly. Sutures should generally start at one end of the incision and proceed
84550.indb 581
11/11/10 6:28:28 PM
582
Handbook of Laboratory Animal Science
in a regular pattern and spacing to the opposite end. In general, sutures should be spaced twice the distance along the length as the distance of the suture bite from side to side of the incision (Swindle 1988, 1983; Westaby 1985; Zederfeldt and Hunt 1990). The suture pattern depends upon the area of the wound being closed. Simple interrupted sutures offer the advantage of security, since the loss of one suture will not open the entire wound. Continuous sutures are easier and quicker to tie and offer a better seal of the edges, but if one knot comes loose, the entire layer will dehisce. Everting sutures, such as the horizontal and vertical mattress patterns, have special indications for tension relief on wound edges. The horizontal mattress is used to relieve tension along a lengthy incision, while the vertical mattress is used in the skin to reduce tension and potential scarring. The Cushing and Connell suture patterns, two inverting patterns, are used to close hollow organs, such as the stomach and bladder. The Lembert pattern is used by itself in a single layer closure or as an inverting suture to oversew other patterns, such as the Cushing or Connell. These inverting patterns provide a waterproof seal. The indications for the various patterns need to be considered by the surgeon in advance of the procedure (Swindle 1988, 1983; Tera and Aberg 1976) (Figure€19.6). Choice of suture material is dependent upon the surgeon’s preference, the structure to be sutured, the animal species, and the size of the incision. Generally, suture material may be classified as absorbable or nonabsorbable (Swindle 1988, 1983; Irvin, Koffman, and Duthie 1976; Kjaergaard et al. 1976; Craig et al. 1975; Postlethwait, Willigan, and Ulin 1975). The synthetic absorbable materials are superior to natural products, such as surgical gut, because they have greater tensile strength and a longer period before they are absorbed after implantation, and they are less inflammatory. Synthetic, nonabsorbable suture materials are also generally superior to natural products, such as silk, for the same reasons. Suture material may be braided for extra strength. It may also be coated for easier passage through tissues. Monofilament (synthetic) sutures are widely used for the suturing of cardiovascular tissues and skin. Suture material is sized in (0) sizes. The most commonly used suture materials for general surgery are 0–3/0. Suture material diminishes in diameter as the (0) size increases, with those >8/0 used primarily in microsurgery and sizes 1–3 used for larger animals. Stainless steel, another suture material, is widely used in orthopedic procedures. Staples have replaced suture materials in many cases and come in a variety of sizes in both absorbable and nonabsorbable materials. Specialized devices are available for a wide range of procedures, including bowel anastomoses, vascular transfixion, and skin closure. Using staples in the skin may lead to an increased incidence of skin infections, since foreign materials tend to get caught in them and animals can catch them on their cages. Even so, staples can be used to particularly good effect in rodents (Swindle and Adams 1988). Approaches to Surgery Laparotomy The laparotomy approach selected (midline, paramedian, flank, and transverse) in both large animals and rodents depends upon the site of surgical interest. The midline approach is the least traumatic, with the incision made along the linea alba, thereby avoiding transection or dissection of muscles. Any other approach will involve either cutting or blunt dissection through the muscles of the abdominal wall. The flank approach is usually reserved for surgical approach to the kidneys or adrenals. Paramedian and transverse incisions are usually utilized for approaches to organs, such as the spleen, biliary system, uterus, or stomach in larger animals for which adequate exposure cannot be attained from a midline incision. Approaches for laparoscopic surgery depend
84550.indb 582
11/11/10 6:28:28 PM
Surgery
583
(a)
(b)
(c) Figure 19.6â•…Suture patterns commonly used in surgery. (a) Muscle; (b) simple interrupted; (c) simple continuous; (d) horizontal mattress; (e) vertical mattress; (f) subcuticular suture pattern; (g) simple continuous skin suture. (Reprinted with permission from Swindle, M. M. 2007. Swine in the Laboratory, Surgery, Anesthesia, Imaging, and Experimental Techniques, 2nd ed. Boca Raton, FL: CRC Press.)
84550.indb 583
11/11/10 6:28:30 PM
584
Handbook of Laboratory Animal Science
(d)
(e)
(f ) Figure 19.6â•…(Continued)
upon the procedure being performed; a full discussion of the relationship between approach and procedure is beyond the scope of this chapter (Swindle 1988, 1983). The principles of surgery are the same regardless of the approach and need to address all of the issues discussed previously. Midline incisions are made with the animal in dorsal recumbency. Large animals should be placed in a V-shaped trough, be supported by sandbags, or have their legs tied cranially or caudally to prevent rotation. An incision along the linea alba is relatively bloodless
84550.indb 584
11/11/10 6:28:34 PM
Surgery
585
(g)
Figure 19.6â•…(Continued)
and in some species, such as the dog, fat may be encountered within the abdomen, requiring some hemostatic procedures. The initial incision should be made completely through the skin along the entire length of the desired surgical wound. The remainder of the abdominal incision is made with either scissors or an electrocautery. Next, the midline is cut until the abdomen is to be entered. At the point of entering the abdominal cavity, the tissue on either side of the midline is lifted with forceps and a stab incision is made into the abdomen. This allows air to enter the abdominal space and the organs are retracted back into the cavity, preventing them from being accidentally incised during the procedure. When using scissors, the muscles should be tented upward so as to avoid inadvertently harming the underlying organs. At this point, the edges of the incision are covered with moistened gauze laparotomy sponges. Depending upon the site of interest and the size of the incision, self-retaining retractors, such as Balfours for large animals, are utilized to provide increased exposure (Swindle 1988, 1983). Abdominal organs and tissues should have minimal exposure to air or abrasive materials. Tissues should be frequently wetted during surgical procedures to prevent drying. Use of warmed saline for irrigation prevents hypothermia. The proper instruments, or fingers, should be used in handling abdominal organs because they tend to be friable and easily damaged (Swindle 1983, 1988). Thoracotomy Thoracotomy is generally performed through an intercostal space or via a median sternotomy (Swindle 1983, 1988). The choice of the incision site is the determining factor in most cases, and approaches to various thoracic organs and structures will vary among species. A ventilator is required for thoracic procedures. An experienced anesthetist is necessary in order to assess oxygenation, blood gas measurements, and ventilation rates and pressures adequately. Cardiopulmonary emergency drugs and devices, such as a defibrillator for large animals, should be available. A lateral thoracotomy is performed in an intercostal space parallel to and between two ribs. Because of differences in the angles of the ribs of the thoracic cages among species, this may be an almost oblique incision in some species, such as the pig. The skin and muscle are incised down to the level of the pleura while the animal is in lateral recumbency. When the pleura is reached, a stab incision is made in the pleura as the animal exhales. This may require momentarily turning off the
84550.indb 585
11/11/10 6:28:35 PM
586
Handbook of Laboratory Animal Science
ventilator. The ribs are retracted outward to avoid damage to the lungs. After the pleura is incised, scissors are utilized to extend the incision. Self-retaining rib retractors are placed to expand the surgical exposure after applying moist gauze pads to the edges of the incision. Special care should be taken when handling the lungs due to their friability. The lungs should be packed off with wetted gauze and the ventilation volume adjusted. As with other procedures, living tissue should be moistened with warm saline to prevent desiccation. The median sternotomy approach is generally utilized for exposure of the heart for procedures such as transplantation. This approach offers greater exposure of the heart without the interference of the lungs in the surgical field. It is more readily performed without splitting the manubrium sterni to minimize postsurgical pain due to movement of the sternum in the postoperative period. After a midline incision is made, the sternum is split with a surgical saw or chisel. It is best to position a metal spatula beneath the sternum when it is split to avoid damage to the heart. Selfretaining retractors are utilized as described previously. Lateral thoracotomies are closed first by preplacing large sutures around the ribs on either side of the incision. A chest tube is placed through the skin and muscle layers caudal and dorsal to the incision site, placing the tip so that it does not traumatize the lungs. The rib sutures are then tied tightly. The remaining muscle and subcutaneous and skin layers are closed routinely. The chest tube is then secured in place with a purse string suture in the skin and the air is evacuated from the chest, either with a syringe or by using a chest tube with a one-way valve. Removal of the chest tube will depend upon the procedure performed. It may be removed after only a few minutes if no air or blood is present in the thorax or it may have to stay in place for a number of days for more invasive procedures. For median sternotomies, the sternum is closed with wire or nonabsorbable sutures after they are preplaced in the same fashion as described for lateral thoracotomies. The subcutaneous and skin layers are closed in a routine fashion and a chest tube is utilized as described before. Catheterization/Cannulation Implantation of catheters is one of the most common research surgical procedures performed (Swindle and Adams 1988; Swindle 1988; Swindle et al. 1998; Bamstein et al. 1966; Bailie 1986; Girardet and Benninghoff 1973; Manolas et al. 1983; Nelson and Swan 1969; Snow and Tyner 1969; Witzel, Littledike, and Cook 1973; Farins et al. 1986; Faulkner et al. 1976; Olesen, Sjontoft, and Tronier 1989; Swindle et al. 2005). Short-term catheterization for a few days is relatively easy to accomplish and maintain; however, complications increase with the amount of time that the catheter is required to be maintained and are related to the site of placement. Discussion in this chapter will be limited to the general principles of implanting and maintaining chronic intravascular catheters. Design of the catheter is critical to the success of long-term catheterization. It is best to utilize manufactured catheters that are free of imperfections, rather than to attempt to make catheters in house. Catheters with coatings to minimize thrombogenesis are also available for long-term implantation. The two most common catheter materials are silicone and polyurethane. Silicone is more flexible and less traumatic, but can absorb contaminants if not handled properly prior to implantation. It is more difficult to guide into vessels because of its flexibility and is more thrombogenic. Tapered tips are preferred over sharp or angular tips, to prevent trauma to vascular walls. Catheters have to be tied securely in place to minimize movement, both at the entrance to the blood vessel and to adjacent connective tissue. The length of the catheter should be premeasured to ensure that the tip placement will be correct. Use of preplaced beads and cuffs that are secured to the catheter prior to sterilization greatly facilitates surgical implantation (Figure€19.7). In order to externalize the distal end of the catheter, the catheter track from the point of exit from the skin to the point of entry into the blood vessel should be made with a trochar (tunneling rod).
84550.indb 586
11/11/10 6:28:35 PM
Surgery
587
Figure 19.7â•…Examples of implantable vascular access ports.
The trochar is passed from the site of the blood vessel through the subcutaneous tissue to the skin. A skin incision is made over the tip of the trochar and the catheter is then passed into the hollow tube. When the trochar is removed, the catheter should be in place with minimal trauma to the tissues. Subcutaneous vascular access ports can also be used and the catheter track is made in the same manner. The site of exit of the catheter depends upon the blood vessel of interest, the species of animal, and the use for which the catheter is intended. The exit site should be chosen to allow ready access by personnel while minimizing access for the animal to chew or scratch the site. The use of jackets and/ or harnesses may be required (Figure€19.8). Vessels frequently used for chronic catheterization include the: external and internal jugular veins, carotid artery, femoral artery, and femoral vein. However, virtually any internal or external
Figure 19.8â•…Pig with multiple catheterizations and a tether and harness system.
84550.indb 587
11/11/10 6:28:36 PM
588
Handbook of Laboratory Animal Science
vessel can be catheterized with a properly designed catheter. After the surgical approach is performed, the vessels are isolated and cleared of adventitial tissue. Elastic vessel loops or suture strands are placed into the vessel proximal and distal to the site of entrance of the catheter; they are retracted when necessary to interrupt blood flow. The vessel is allowed to fill with blood and a small nick incision (no more than one-third of the way through the vessel) is made with either iris scissors or a #11 scalpel blade. A vascular pick or small probe may be used to facilitate passing the catheter into the lumen of the vessel. Catheters should be prefilled with saline. When the catheter is passed toward the heart, the proximal vascular loop is released enough to allow atraumatic passage. Catheters are secured in place both proximally and distally with nonabsorbable sutures, after which blood is aspirated and then flushed to check adequate flow. A loop of the catheter is left in the subcutaneous tissue to ensure that the catheter is not under tension during growth or movement of the conscious animal. The surgical wounds are closed in the usual manner. The catheter exit site should be sealed to prevent infection and catheter migration. To prevent clotting within the catheter, a heparin–saline solution must be introduced and “locked” in place. This is accomplished by closing the end of the catheter with a pin or by attaching it to an injection port or a device such as an access port or a swivel. Craniotomy Craniotomies may be performed on rodents for a variety of purposes. Most commonly, they are performed for delivery of substances that affect the brain, measuring or stimulating electrical activity of specific brain areas, or analyzing neurotransmitters or test compounds. This may be for the investigation of and development of therapies for neurodegenerative and psychiatric disorders and other pathological processes that may be centrally mediated (e.g., metabolic disease). Craniotomy may be carried out to implant devices for recording electrical activity, dialysis, electrical stimulation, and/or administration of test substances (Remie 2000). The approach is usually via a small drill hole made in the center of the dorsal surface of the skull, although other approaches may be used in certain circumstances (e.g., zygomatic approach to access the middle cerebral artery). When it is necessary to target a specific area, its location should be mapped and stereotaxic coordinates obtained. Atlases with this information are available (Franklin and Paxinos 1997; Paxinos and Watson 1986); however, it is advisable to validate these for the particular species, strain, and body weight of the animal to be used. This can be done using cadavers and positioning cannulae stereotaxically, injecting dye, and examining brain sections histologically to determine the optimal position. The procedure for a dorsal craniotomy follows a standard pattern for most of the purposes mentioned earlier. The animal is anesthetized by injection or inhalation. When inhalation anesthesia is used, efforts must be devoted to ensuring the efficient delivery and scavenging of anesthetic gas (i.e., use of a purpose-made face mask that fits on the stereotaxic frame or endotracheal intubation). The skin is clipped and cleaned at the surgical site, taking care not to contaminate the eyes. The animal is positioned in the stereotaxic frame in ventral recumbency on a thermostatically controlled heat blanket to maintain body temperature. A drape is placed over the whole animal with only the surgical site exposed (Figure€19.4). An incision is made in the skin, from behind the eyes to between the ears. It is recommended that local anesthetic be injected into the periosteum, which can then be scraped away. The position of the entry site for the implant or needle is determined stereotaxically and holes are carefully drilled just through the skull at the entry point and at two to four surrounding points to accommodate steel anchoring screws.
84550.indb 588
11/11/10 6:28:36 PM
Surgery
589
Figure 19.9â•…Drilling a hole in the skull.
When implants are being used, up to four anchoring screws are placed in the skull to support the placing of the head cap (Figures€19.9–19.11). The implant is positioned to the required depth and a supporting head cap of dental cement is then constructed around the screws. The skin is sutured at the front and rear of the head cap, leaving both the cap and the implant exteriorized. Gentle tissue handling, judicious use of local anesthetic, and smooth sculpting of the dental cement to an oval shape will prevent tissue irritation and swelling. Perioperative analgesia and postoperative nursing care are given to aid recovery. When an intracerebroventricular cannula has been implanted, its position may be confirmed prior to use in studies. This can be accomplished by administering angiotensin via the cannula and observing a drinking response. When there is a good record of accurate cannula placement, the cannula’s position may be verified at necropsy at the end of the study by histology or by injection of dye via the cannula and observation of the dye’s location in the brain postmortem.
Figure 19.10â•…Stereotaxically determined drill hole in the skull.
84550.indb 589
11/11/10 6:28:38 PM
590
Handbook of Laboratory Animal Science
Figure 19.11â•…Stainless steel screws placed into the skull for anchoring and aiding the formation of the head cap.
Postoperative Care Provision of good postoperative care is an extension of good surgical technique and requires appropriate facilities, equipment, and trained personnel. All animals recovering from surgery and anesthesia require monitoring and specialized care, predicated on species-specific needs and the type of surgical procedure. This is most effectively achieved with a separate, dedicated recovery room. The focus of postoperative monitoring should be on returning the animal to and then maintaining normal physiologic parameters, given that anesthetics are known to disrupt homeostasis (Flecknell 1996). If homeostatic abnormalities are inadequately addressed postoperatively, recovery may be prolonged, resulting in further physiologic exacerbations or even death. Large Animals The Recovery Environment A dedicated recovery room should be warm and quiet, providing adequate lighting for staff to observe animals and perform nursing procedures. An ambient room temperature of 21–25°C is generally suitable for most mature animals, as long as supplemental heat sources, such as circulating water blankets and heat lamps, are available. Covering large animals with blankets or drapes also helps to warm the animal. Intensive care units may be utilized to control the ambient temperature for animals weighing less than 25 kg and have the additional advantage of permitting the easy administration of oxygen.
84550.indb 590
11/11/10 6:28:38 PM
Surgery
591
Frequent monitoring of rectal temperature should be performed postoperatively, until it can be determined that the animal has returned to normothermia and can maintain a normal body temperature in the absence of supplemental heat. If an animal has reached normal body temperature, yet becomes hypothermic after removal of supplemental heat, the animal is still physiologically unstable and additional underlying problems may exist. Specially designed cages are normally used for initial postsurgical recovery to facilitate observation and to limit activity. However, in some cases, dog transport cages are suitable for recovery of most large animals up to 20 kg. For larger animals, recovery in runs is more practical. In either case, the recovery area should be equipped to handle potential complications during recovery. Necessary items include intubation supplies in the event of apnea or airway obstruction, suction equipment to clear airway passages of excessive secretions, supplies for fluid administration, and equipment for supplemental oxygen administration. Additional equipment might also include an ECG monitor and pulse oximeter. Monitoring and Nursing Care It is useful to divide the postoperative period into (a) the time of recovery from anesthesia and (b) the period of time that includes recovery from the surgical procedures (general healing of surgical incisions). Vital signs and surgical incisions should be monitored and documented frequently during recovery from anesthesia. Once the animal is stabilized, observations should be performed at least twice daily. When a surgical procedure is performed with the intent of producing chronic disease, monitoring of project-specific parameters should continue for the life of the animal. During recovery from anesthesia, the greatest attention should be paid to the cardiovascular and respiratory systems. Observations of depth, rate, and pattern of respiration; mucous membrane color; capillary refill time; pulse quality; and auscultation of the chest are all clinical observations that can be used collectively to assess cardiopulmonary function. Clinical observations are augmented by ECG monitoring, the use of pulse oximetry to monitor hemoglobin saturation with oxygen, and noninvasive blood pressure monitoring. Documentation of vital signs and pertinent clinical observations should be made in the animal’s medical record. Extubation should be performed after the return of a strong swallowing reflex. During recovery, animals should be kept clean and dry. Surgical incisions should be monitored for drainage and any complications reported immediately. Administration of warmed fluids supports the cardiovascular system and also helps warm the animal. Urine and fecal output should be recorded. Reduced urinary output is cause for concern and can result from dehydration, urinary tract injury, or pain. Reduction of fecal output is not unusual postoperatively due to preoperative fasting; however, if it is prolonged, this may indicate the presence of more serious problems. Once the patient has stabilized, postoperative monitoring should continue until all surgical incisions have healed. Observations of surgical incisions, behavior, and activity; monitoring of vital signs, appetite, water intake, and urine and feces production; and evaluation of pain levels should be made daily. Comparisons of postoperative body weights with preoperative weights can be used to assess the adequacy of food and fluid intake. Although some reduction of body weight postoperatively can be attributed to reduced food intake, most postoperative weight loss is due to fluid deficit. Efforts should be made to encourage animals to eat and drink postoperatively. Offering palatable foods and hand feeding are generally successful for dogs and cats and can also be used for swine and small ruminants accustomed to human contact. Incisions should be cleansed if necessary and evaluated for signs of abnormal healing or infection. Postoperative medications should be given at the orders of the attending veterinarian. Nonabsorbable sutures, clips, or staples should be removed once all wounds have healed. It is important that all information and treatments be documented in the animal’s medical record.
84550.indb 591
11/11/10 6:28:38 PM
592
Handbook of Laboratory Animal Science
Acclimation to handling and human contact preoperatively will help reduce an animal’s anxiety postoperatively. In some cases, animals not used to human contact will require more intensive socialization prior to surgery. Preoperative acclimation should also involve specific dietary modifications, use of restraint techniques, or wearing of jackets or harnesses that will be implemented postoperatively. Ultimately, this aids both the animal and the caregivers during the postoperative period when close human contact is necessary. Monitoring of animals that have chronic induced disease as a result of surgery should be tailored to the needs of the project while minimizing the animals’ distress. Monitoring parameters should be defined when the protocol is designed and modified as clinical experience with the model develops. Parameters may include periodic echocardiography for heart failure models or periodic serum chemistry profiles for metabolic disease models. Surgically implanted catheters or vascular access ports should be used for projects requiring repetitive blood sampling or drug administration. Administration of analgesics or other drugs that could alleviate distress should be utilized in consultation with the veterinarian and scientists to avoid any that might be contraindicated scientifically (Swindle 1988). Rodents Since all rodents will require some degree of special attention in the postoperative period, it is preferable, as with large animals, to provide a separate recovery area. This not only enables more appropriate environmental conditions to be maintained, but also encourages individual attention and special nursing (Lane 1985; Taylor and McGehee 1995). Warmth and Comfort Supplemental heating to prevent hypothermia should be provided. Animal incubators are excellent for this purpose and also enable easy observation of the animals. The temperature for adult animals should be maintained between 25 and 30°C and between 35 and 37°C for neonatal animals. Care must be taken not to overheat the animal and a thermometer should be placed near its body. Postoperative bedding should be comfortable and provide effective insulation. Rats should never be allowed to recover from anesthetic directly on sawdust or similar bedding that could stick to the animal’s nose, eyes, and mouth. Special bedding material made out of cloth or synthetic fur is available commercially and is excellent for all species. Respiratory Depression Respiratory depression can develop postoperatively and be dangerous for the animal. Respiration should be monitored continuously by observation and treated appropriately with oxygen supplementation or intubation and ventilation if depression occurs. Fluid Therapy Excessive loss of fluid can occur very quickly intraoperatively, particularly in rats and mice. In addition, the animal may not drink normally for up to the first 24 hours in the postoperative period. Fluid should be given parenterally during anesthesia or postoperatively by subcutaneous, intraperitoneal, or intravenous administration in quantities that will replace estimated losses during surgery and the initial 12–24 hours of the postoperative period. Monitoring of body weight pre- and postoperatively can provide a good indication of the adequacy of food and fluid intake. Urine and fecal output should be recorded and any abnormalities investigated. Reduced urine output may be due to dehydration or urinary tract injury, or in response to pain. Failure of the animal to defecate
84550.indb 592
11/11/10 6:28:38 PM
Surgery
593
must always be investigated. Potential causes include paralytic ileus or bowel entrapment in any implanted intra-abdominal device that may be present. General Care Companion animals, such as dogs and cats, respond well to human contact, but rats and mice may find this stressful and prefer to be in a warm environment with subdued lighting and as few disturbances as possible. All clinical observations and drugs administered must be recorded on a patient record card kept near the animal for rapid reference. Observations (e.g., body weight, activity, integrity of the wound site) should be made over the first few days postoperatively to determine whether animals are recovering normally. Postoperative Pain Any surgery from which the animal recovers from the anesthetic could cause pain during the recovery period. It is now generally considered essential to provide pain relief for all surgical procedures in all species—no matter how brief—including surgery on rodents. It is important to make an assessment of the degree and probable duration of pain to enable the analgesic regimen to be tailored appropriately and its effectiveness monitored. In many cases, this may have to be done anthropomorphically, although clinical schemes for assessment of pain in animals have been published. The management of pain in laboratory animals has been discussed widely and this, together with the appropriate drug regimens to use, is well described in the scientific literature (Flecknell 1984; Crawford 2000; Liles and Flecknell 1993; Hampshire and Gonder 2007; Fish et al. 2008).
References Bailie, M. B. 1986. Vascular-access-port implantation for serial blood sampling in conscious swine. Laboratory Animal Science 36:431. Bamstein, J. J., R. S. Gilfillan, N. Pace, and D. F. Rahlmann. 1966. Chronic intravascular catheterization. Journal of Surgical Research 6:6–11. Bradfield, J. F., T. R. Schachtman, R. M. McLaughlin, and E. K. Steffan. 1992. Behavioral and physiological effects of inapparent wound infection in rats. Laboratory Animal Science 42:572. British Laboratory Veterinary Association. 1992. Experimental surgery slide program, slide sets and notes. Obtained from P. A. Flecknell, Comparative Biology Center, Medical School, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK. College of Animal Welfare. 1997. Veterinary surgical instruments. An illustrated guide. Oxford, England: Butterworth-Heinemann. Collins, C. H., M. C. Allwood, S. F. Bloomfield, and A. Fox. 1981. Disinfectants: Their use and evaluation of effectiveness. London: Academic Press. Craig, P. H., J. A. Williams, K. W. Davis, A. D. Magoun, A. J. Levy, S. Bogdansky, and J. P. Jones. 1975. A biologic comparison of polyglatin 910 and polyglycolic acid synthetic absorbable sutures. Surgery 141:1. Crawford, R. L. 2000. A reference source for the recognition and alleviation of pain and distress in animals. See Web site http:// www.nal.usda.gov/awic. Cunliffe-Beamer, T. L. 1983. Biomethodology and surgical technique. In The mouse in biomedical research, vol. 3, ed. H. L. Foster. J. D. Small, and J. G. Fox. New York: Academic Press. ———. 1989. Surgical techniques. In Guidelines for the well-being of rodents in research, ed. H. N. Guttman. Bethesda, MD: Scientists Center for Animal Welfare. De Boer, J., J. Archibald, and H. G. Downie. 1975. An introduction to experimental surgery. New York: American Elsevier Publishing Company. Dougherty, R. W. 1981. Experimental surgery in farm animals. Ames: Iowa State University Press.
84550.indb 593
11/11/10 6:28:39 PM
594
Handbook of Laboratory Animal Science
Eshleman, J. R. 1968. Methods used for sterilization or disinfection of instruments. Journal of Dental Education 32:330. Farins, L. R., E. S. Woodle, C. F. Frey, S. I. Nakayama, and R. E. Ward. 1986. A simple technique for experimental hepatic vein catheterization in swine. Laboratory Animal Science 36:406. Faulkner, R. T., W. P. Czajkowski, E. J. Rayfield, and R. L. Hickman. 1976. Technique for portal catheterization in rhesus monkey (Macaca mulatta). American Journal of Veterinary Research 37:473. Fish, R. E., M. J. Brown, P. J. Danneman, and A. Z. Karas. 2008. Anesthesia and analgesia in laboratory animals, 2nd ed. New York: Academic Press (Elsevier). Flecknell, P. A. 1984. The relief of pain in laboratory animals. Laboratory Animals 18:147. ———. 1996. Laboratory animal anesthesia. New York: Academic Press. Franklin, K. B. J., and G. Paxinos. 1997. The mouse brain in stereotactic coordinates. New York: Academic Press. Ghosh, J., D. O. Maisels, and A. S. Woodcock. 1967. Preoperative skin disinfection. British Journal of Surgery 54:551. Girardet, R. E., and D. L. Benninghoff. 1973. Surgical techniques for long-term effects of thoracic duct lymph circulation in dogs. Journal of Surgical Research 15:168. Green, C. J., and S. Simpkin. 1990. Basic microsurgical techniques. A laboratory manual. Obtained from the Surgical Research Group, MRC Clinical Research Center, Northwick Park Hospital, Harrow, Middlesex, UK. Hampshire, V., and J. C. Gonder, eds. 2007. Research animal anesthesia, analgesia and surgery. Bethesda, MD: Scientists Center for Animal Welfare (SCAW). Hecker, J. F. 1974. Experimental surgery on small ruminants. South Hampton, England: Butterworth & Co. Hurov, L. 1978. Handbook of veterinary surgical instruments and glossary of surgical terms. Philadelphia, PA: W. B. Saunders Co. Irvin, T. T., C. G. Koffman, and H. L. Duthie. 1976. Layer closure of laparotomy wounds with absorbable and nonabsorbable suture materials. British Journal of Surgery 63:793. Kjaergaard, J., N. P. Laursen, C. M. Madsen, A. Tilma, and C. Zimmermann-Nielsen. 1976. Comparison of dexon and mersilene sutures in the closure of primary laparotomy incisions. Acta Chirurgica Scandinavica 142:315. Knecht, C. D., A. R. Allen, D. J. Williams, and J. H. Johnson. 1981. Fundamental techniques in veterinary surgery. Philadelphia, PA: W. B. Saunders Company. Lambert, R. 1982. Surgery of the digestive system of the rat. Springfield, IL: Charles C Thomas. Lane, D. R., ed. 1985. Jones’s animal nursing, 4th ed. New York: Pergamon Press. Liles, J. H., and P. A. Flecknell. 1993. The influence of buprenorphine or bupivicaine on the postoperative effects of laparotomy and bile-duct ligation in rats. Laboratory Animals 27:374. Lumley, J. S. P., C. J. Green, and J. E. Angell-James. 1990. Essentials of experimental surgery. Stoneham, MA: Butterworth. Manolas, K. J., H. M. Farmer, J. Cussen, and R. B. Welbourn. 1983. An experimental model for simultaneous chronic sampling of portal and systemic blood and gastrointestinal lymph via cannulae in conscious swine. Cornell Veterinarian 73:333. Morris, T. 1995. Antibiotic therapeutics in laboratory animals. Laboratory Animals 29:16. Nelson, A. W., and H. Swan. 1969. Long-term catheterization of the thoracic duct in the dog. Archives of Surgery 98:83. Olesen, H. P., E. Sjontoft, and B. Tronier. 1989. Simultaneous sampling of portal, hepatic and systemic blood during intragastric loading and tracer infusion in conscious pigs. Laboratory Animal Science 39:429. Paxinos, G., and C. Watson. 1986. The rat brain in stereotactic coordinates, 2nd ed. New York: Academic Press. Petty, C. 1982. Research techniques in the rat. Springfield, IL: Charles C Thomas. Postlethwait, R. W., D. A. Willigan, and A. W. Ulin. 1975. Human tissue reaction to sutures. Annals of Surgery 181:144. Remie, R. 2000. Experimental surgery. In The laboratory rat, ed. G. J. Krinke. New York: Academic Press. Santiesteban, R., D. Hutson, and R. S. Dombro. 1983. Chronic catheterization of the portal vein in dogs. Laboratory Animal Science 33:373. Sirek, A., and O. V. Sirek. 1983. A new technique for hepatic portal sampling in the conscious dog. Proceedings of Society for Experimental Biology and Medicine 172:397. Slatter, D. H. 1985. Textbook of small animal surgery. Philadelphia, PA: W. B. Saunders Company. Snow, H. D., and J. G. Tyner. 1969. Chronic arterial and venous catheterization in sheep. American Journal of Veterinary Research 30:2241. Strachan, C. J. L., and R. Wise, eds. 1979. Surgical sepsis. New York: Academic Press.
84550.indb 594
11/11/10 6:28:39 PM
Surgery
595
Swindle, M. M. 1983. Basic surgical exercises using swine. Philadelphia, PA: Praeger Scientific. ———. 1988. Surgery, anesthesia and experimental techniques in swine. Ames: Iowa State University Press. ———. 2007. Swine in the laboratory, surgery, anesthesia, imaging, and experimental techniques, 2nd ed. Boca Raton, FL: CRC Press. Swindle, M. M., and P. J. Adams. 1988. Experimental surgery and physiology. Baltimore, MD: Williams & Wilkins. Swindle, M. M., T. Nolan, A. Jacobsen, P. Wolf, and A. C. Smith. 2005. Vascular access port (VAP) usage in large animal species. Contemporary Topics in Laboratory Animal Science 44 (3): 7–17. Swindle, M. M., A. C. Smith, and J. A. Goodrich. 1998. Chronic cannulation and fistulation procedures in swine: A review and recommendations. Journal of Investigative Surgery 11:7–20. Taylor, R., and R. McGehee. 1995. Manual of small animal postoperative care. Baltimore, MD: Williams and Wilkins. Tera, H., and C. Aberg. 1976. The strength of suture knots after one week in vivo. Acta Chirurgica Scandinavica 142:301. Van Dongen, J. J., R. Remie, J. W. Rensema, and G. H. J. van Wannik, eds. 1990. Manual of microsurgery on the laboratory rat. Part 1. General information and experimental. New York: Elsevier. Waynforth, H. B. 1995. Basics of surgery. In Laboratory Animals. An Introduction for Experimenters, 2nd ed., ed. A. A. Tuffrey. New York: John Wiley & Sons. Waynforth, H. B., and P. A. Flecknell. 1992. Experimental and surgical technique in the rat, 2nd ed. New York: Academic Press. Westaby, S. 1985. Wound care. London: William Heinemann Medical Books Ltd. Witzel, D. A., E. T. Littledike, and H. M. Cook. 1973. Implanted catheters for blood sampling in swine. Cornell Veterinarian 63:432. Zederfeldt, B. H., and T. K. Hunt. 1990. Wound closure. Wayne, NJ: Davis & Geck.
84550.indb 595
11/11/10 6:28:39 PM
84550.indb 596
11/11/10 6:28:39 PM
Chapter 20
Microsurgical Procedures in Experimental Research
Daniel A. Steinbrüchel Contents Introduction..................................................................................................................................... 597 Basic Requirements for Experimental Microsurgical Projects...................................................... 598 Technical Requirements............................................................................................................. 598 Microscopes.......................................................................................................................... 598 Microsurgical Instruments and Accessories......................................................................... 598 Laboratory Facilities............................................................................................................. 599 Basic Techniques of Microsurgical Vascular Anastomosis............................................................600 End-to-End Anastomosis...........................................................................................................600 Interrupted Suture Technique................................................................................................600 End-to-Side Anastomosis...........................................................................................................602 Vein Anastomosis.......................................................................................................................603 Cuff Technique...........................................................................................................................603 Splint Technique........................................................................................................................603 Selection of Often Used Microsurgical Models for Transplantation Research..............................604 Heterotopic Heart Transplantation in the Rat............................................................................605 Kidney Transplantation in the Rat.............................................................................................605 Other Microsurgical Transplantation Models............................................................................606 Concluding Remarks and Comments.............................................................................................607 References.......................................................................................................................................609 Introduction The history and development of microsurgical procedures essentially reflect recurrent attempts over several decades to solve the problem of establishment or reestablishment of vascular continuity in vessels of decreasing diameter. At the same time, this development illustrates the mutually beneficial influence of clinical experience and experimental microsurgical results, where clinical data initiated detailed investigations in microsurgical animal models, while experimental experience could be applied directly to clinical procedures. Carrel and Guthrie first demonstrated the feasibility of vascular anastomosis (Carrel 1902; Carrel and Guthrie 1906; Guthrie 1908). Before this time, a major vascular lesion in an extremity 597
84550.indb 597
11/11/10 6:28:39 PM
598
Handbook of Laboratory Animal Science
usually resulted in amputation. In spite of the introduction of the microscope to clinical use in 1921 (Nylén 1972) and a gradual development of specialized instruments and accessories (clamps, suture materials), successful vascular anastomosis of vessels in the 2–3 mm range was technically not feasible until 1960, when Jacobson and Suarez (1960) demonstrated the successful anastomosis of blood vessels of 1 mm in external diameter. Subsequently, the application of microsurgical and microvascular procedures progressed rapidly, both clinically and experimentally. Successful replantation of amputated extremities (Kleinert and Kasdan 1963; Horn 1964; Bunke and Schulz 1965) and the transfer of free skin flaps as composite grafts were reported (Krizek et al. 1965; Strauch and Murray 1967; McLean and Buncke 1972; McGregor and Morgan 1973; O’Brien et al. 1973; Acland and Smith 1976). Different laboratories described a variety of microsurgical models that focused on the transplantation of primary vascularized organs in rodents (Lee and Fisher 1961; Miller et al. 1962; Abbott et al. 1964; Fisher and Lee 1965; Lee 1967; Reemtsma et al. 1968; Mikaeloff et al. 1969; Ono and Lindsey 1969; Lee et al. 1970; Monchick and Russel 1971). Today, microsurgical techniques are widely used in ophthalmologic, otologic, and reconstructive and plastic surgeries. Experimentally, microsurgical models are applied in studies focusing on physiological aspects and processes in the microvasculature subsequent to free tissue transfer and in transplantation research, where microvascular models (performed as routine procedures) form the basis of testing new immunosuppressing and immunomodulating treatment strategies and permit a more detailed study of the processes involved in allogenic and xenogeneic rejection of transplanted organs. Basic Requirements for Experimental Microsurgical Projects Performing microsurgical procedures with success, in terms of generating valid and relevant experimental data, is dependent on several factors, including not only the necessary technical equipment, but also research assistants with motivation, a certain persistence, and a genuine scientific interest. It needs training, qualified supervision, and a positive attitude to teamwork because scientifically interesting projects, which include microsurgery, will consist of an interdisciplinary approach to often complex problems, in collaboration with immunologists, pathologists, physiologists, and clinicians. Technical Requirements Microscopes Preference has to be given to the operating microscope compared to magnifying glasses. It offers the advantage of greater magnification (stepwise or zoom function), built-in illumination, and the possibility of documentation (video, photograph). On the other hand, the field of view is limited and the depth of focus moderate, which can partly be compensated for by foot switch control. The cost is clearly a disadvantage, but it will prove to be a good investment in the long run. Furthermore, diploscopes make the performance of complex microsurgical procedures easier and allow detailed training supervision. Microsurgical Instruments and Accessories A few high-quality instruments with which you are familiar and that you use routinely are sufficient. An extensive variety of different instruments usually has no beneficial effect on technical accuracy and efficiency. As a rule, as simple as possible is best. The choice of optimal clamps, suture material, and needles for different vessel and tissue types must be taken into account.
84550.indb 598
11/11/10 6:28:39 PM
Microsurgical Procedures in Experimental Research
599
Figure 20.1â•…A basic set of microsurgical instruments with needle holder, dissecting scissors, and a pair of microsurgical forceps.
Figure€20.1 illustrates a simple set of essential microsurgical instruments consisting of a needle holder, a dissecting scissors, and a pair of microsurgical forceps. A satin finish to avoid glare is an advantage, and the instruments should have sufficient length to make a convenient pen grip possible. Two single clamps and a twin-clip approximating clamp are shown in Figure€20.2, and the size is illustrated by a match. Microsurgical instruments are rather expensive and very delicate, but appropriate use and care guarantee excellent function for many years (for further reading concerning more detailed instrument description, see Engemann, Deltz, and Thiede 1982; Silber 1979). Laboratory Facilities The importance of optimal facilities for observation and housing of animals (with respect to microsurgical models, most often rodents) must not be underestimated. Advanced projects often include intensive monitoring that is not limited to the immediate postoperative period, but rather proceeds for several months. Qualified full-time technical assistance is therefore necessary to achieve optimal benefits in terms of valid, complete, and reproducible experimental results.
Figure 20.2â•…Two single clamps and a twin-clip approximating clamp (match for comparison of size).
84550.indb 599
11/11/10 6:28:41 PM
600
Handbook of Laboratory Animal Science
Basic Techniques of Microsurgical Vascular Anastomosis Atraumatic technique is an essential prerequisite for successful microsurgical procedures. The preparation of arteries and veins includes dissection from surrounding tissue, where a sharp division of structures (with scissors) on the basis of knowledge of natural cleavages is the principle. Blunt dissection and unnecessary manipulation of the vessels must be avoided. The choice of clamps has to be adapted to the type and size of vessel; an optimal clamp exerts a minimal necessary degree of compression, diminishing the risk of endothelial damage. After division of the vessel with one clear cut, the stumps are rinsed with saline solution. Addition of heparin is not decisive; the important factor for a successful anastomosis is the surgical technique. Subsequently, adjacent adventitia is removed and the stumps can, if necessary, be gently dilated. Any instrumental manipulation with risk of intimal damage during preparation as well as suturing increases the possibility for thrombosis of the anastomosis. End-to-End Anastomosis The infrarenal aorta and the femoral artery in rats are the most suitable objects for the beginner in experimental microsurgery. Interrupted Suture Technique Several methods described in the literature are discussed in the following sections. Eccentric Biangulation Technique This technique, first described by Cobbet (1967), is today very popular and can be recommended as the initial type of suture technique for beginners in the field of microsurgical vessel anastomosis. As illustrated in Figures€20.3A and 20.3B, the initial two stay sutures are applied 120° apart. One or two interposed sutures will finish the anterior wall, after which the vessel is rotated 180°. The posterior wall can now be sutured. The use of a twin-clip approximating clamp is a clear advantage for this type of suture technique. Biangulation Technique It is not always possible to place the initial stay sutures in the exact eccentric position, especially when there is a major discrepancy between vessel diameters. The use of the biangulation technique (Harashina 1977) (Figure€20.4) can therefore be an advantage; the two initial stay sutures are placed 180° apart, hereby defining an appropriate adaptation of the two different vessel diameters. Interpositioning of sutures in the anterior wall and, after rotation, in the posterior wall will complete the anastomosis. Successive Interrupted Sutures (Ship’s Wheel Type) This technique (Fujino and Aoyagi 1975) is especially suitable when interrupted sutures are preferred but anatomical circumstances do not allow a free rotation of the vessel (Figure€20.5). It is an advantage to place the first suture in the posterior wall. After the suture has been tied, one end is cut near the knot, while a gentle traction of the long end will facilitate the exact positioning of the next adjacent suture. This procedure is repeated until the anastomosis is finished.
84550.indb 600
11/11/10 6:28:42 PM
Microsurgical Procedures in Experimental Research
601
2 3 1
5 4 2 3
Figure 20.3â•…(Top) Initial stay sutures are placed 120° apart (1,2). (Bottom) The vessel is turned 180° and the anastomosis can be accomplished from the front.
1 5 4 3 2
Figure 20.4â•…Initial stay sutures are placed 180° apart (1,2), and the remaining sutures are hereafter interposed.
1
2 3
Figure 20.5â•…The former stay suture is used to assist the placement of the next suture in a continuous way.
84550.indb 601
11/11/10 6:28:44 PM
602
Handbook of Laboratory Animal Science
(a)
(b)
(c)
Figure 20.6â•…(a) Positioning of primary sutures, biangulated or eccentric. (b) Suturing the anterior wall. (c) Completing the posterior wall after 180° rotation.
Running Suture Technique This technique (Biemer and Schmidt-Tintemann 1982) shortens anastomosis time and improves primary hemostasis, but this technique includes the risk of stenosis in less trained hands (Figure€20.6). End-to-Side Anastomosis The principles of suture technique are basically identical to those of end-to-end anastomosis. Interrupted or running sutures can be used. However, microsurgical organ transplantation models usually do not permit a free rotation of vessels, so anastomosis technique includes suturing of the posterior wall from the luminal side, which is done more easily using a running type of suture. The technique is illustrated in Figure€20.7.
84550.indb 602
11/11/10 6:28:45 PM
Microsurgical Procedures in Experimental Research
603
(a)
(b)
(c)
Figure 20.7â•…(a) Placing of angle sutures. (b) Suture of the posterior wall from the inside. (c) Completing the anastomosis from the front.
Vein Anastomosis Veins are fragile in rodents and must be handled carefully. The technique of anastomosis is essentially the same as that applied in arteries. Cuff Technique This is an alternative, nonsuture method for vascular anastomosis (Figure€20.8) (Heron 1971; Kamada and Calne 1979). In principle, one vessel end is everted over a polyethylene cuff, the other is pulled over the endothelialized cuff, and the anastomosis is secured by a circular ligation. Splint Technique This useful method has preferably been used for ureter and bile duct end-to-end anastomosis or the insertion of a stented ureter or choledochus into the recipient bladder or duodenum, respectively (Daniller, Buchholz, and Chase 1968; Zimmermann et al. 1981). The suture and anastomosis techniques illustrated here are the basic approaches to microvascular surgery, which will naturally be the
84550.indb 603
11/11/10 6:28:47 PM
604
Handbook of Laboratory Animal Science
(a)
(b)
(c) Figure 20.8â•…(a) A plastic cuff of adequate size is pushed over the vessel end. (b) Evertion of the vessel stump over the cuff. (c) Completing the anastomosis with a circular ligation.
object of modifications, preferences, and improvements in accordance with personal experience and increasing surgical skill. Selection of Often Used Microsurgical Models for Transplantation Research The introduction of microsurgical procedures to organ transplantation research has made the investigation of specific immunological questions and immunosuppressing or immunomodulating treatment strategies possible to study specific problems of donor organ preservation or to focus on more physiologic aspects of solid organ transplantation. The use of inbred rodent strains (mainly rats and mice) combines the possibility of technically feasible whole organ transplantation in donor–recipient combinations where the genetic disparity in respect to major histoincompatibility is identical between random individuals from the same strain or where genetically manipulated
84550.indb 604
11/11/10 6:28:48 PM
Microsurgical Procedures in Experimental Research
605
animals (transgenic or gene knockout) are available. Therefore, the outcome of organ transplantation between two rodent strains that are, for all practical purposes, genetically identical is reproducible and predictable (Günther 1985). At the same time, rats (and, to a lesser degree, mice) are of a size that makes it possible for trained microvascular surgeons to achieve patency rates (preferably >95% for simple procedures) that satisfy the statistical and scientific demands for reproducible and valid results, where technical failure does not cover real observation data. Transplantation research using microsurgical animal models will often include the use of polyclonal or monoclonal antibodies, genetic transfection, different sera, immunological reagents, and a variety of methods and procedures applied as organ recipient treatment, as well as for immunological and histological analysis. Many of these reagents are commercially available for rats and mice. This is an enormous advantage when planning transplantation research projects in laboratories, which initially do not have the possibility of producing these reagents. From a more practical but no less important point of view, rats and mice are cheap, housing is uncomplicated, and the animals seem to tolerate surgical and anesthesiological stress well and are highly resistant to postoperative infection. This section will shortly deal with the two major microsurgical transplantation models (heart and kidney). Finally, a few more complex procedures will be mentioned. Heterotopic Heart Transplantation in the Rat Cardiac transplantation in rats is the most frequently used model in transplantation research. In principle, it consists of a short circuit of normal heart hemodynamics. The donor heart is excised after ligation of the inferior and superior caval veins and the pulmonary veins, and after division of the aorta and pulmonary artery. On the recipient side, the donor aorta is anastomosed end to side to the infrarenal aorta, and the pulmonary artery end to side to the inferior caval vein. As recipient vessels, the common carotid artery and external jugular vein can be used when preference is given to the heterotopic cervical heart transplantation model. Figure€20.9 illustrates the result after heterotopic heart transplantation to the recipient abdominal vessels. The transplanted heart is perfused via the coronary arteries, draining to the right atrium and through the right ventricle to the venous system of the recipient. But the left ventricle has no physiological function because there is no ventricular inlet. The model is suitable for immunological and histological studies or investigations focusing on cardioplegic methods and organ preservation. In terms of palpable heartbeat, graft function is easily monitored. However, heterotopic cardiac transplantation is a less adequate model for hemodynamic or functional studies (Konertz, Thiede, and Bernhard 1980; Bernhard and Konertz 1983). The creation of an atrial septal defect gives some left ventricular inlet, preventing the often observed formation of an intraventricular thrombus. But this model is not truly functional (Steinbrüchel et al. 1996). Several modifications have been described using the recipient abdominal or cervical vessels for anastomosis or placing the donor heart as left ventricular bypass (Abbott et al. 1964; Ono and Lindsey 1969; Lee et al. 1970; Heron 1971; Steinbrüchel et al. 1990; Konertz, Semik, and Bernhard 1980). However, the original techniques have been shown to be highly resistant to innovation and recent studies use the same techniques described for more than 20 years (Niimi 2001; Richter et al. 2001). Kidney Transplantation in the Rat In contrast to the heterotopic heart transplantation model, renal transplantation offers, in addition to histological and immunological monitoring, the possibility of differentiated functional assessment of a transplanted organ (Salomon et al. 1996; Steinbrüchel et al. 1987, 1992). After bilateral nephrectomy of the recipient’s kidneys, renal function will exclusively depend on the graft.
84550.indb 605
11/11/10 6:28:48 PM
606
Handbook of Laboratory Animal Science
ICV A
Figure 20.9â•…Heterotopic heart transplantation to recipient abdominal vessels with an aorta-to-aorta and a pulmonary artery-to-inferior caval vein anastomosis.
Several techniques of organ harvesting and reimplantation have been described; basically, they differ on the site of anastomosis. In the heterotopic renal transplantation model, the donor kidney is harvested with or without an aortic or caval vein segment or patch and anastomosed end to side to the recipient infrarenal aorta and caval vein (Fisher and Lee 1965; Lee 1967; Jakubowski 1985). The procedure is illustrated in Figure€20.10. The orthotopic model makes use of an end-to-end anastomosis between the donor and recipient renal vessels, thereby replacing the recipient’s kidney with a graft (Fabre, Lim, and Morris 1971; Kamada 1985) (Figure€20.11). The urinary tract can be reconstituted by ureteric implantation into the recipient bladder, bladder-to-bladder anastomosis, or a direct end-to-end ureter anastomosis. I think that excellent results can be achieved by any of these different techniques and that major or minor variations are less important compared to the necessity of atraumatic microsurgical practice and perfectionism, as well as a reduction of warm ischemia time to zero. Other Microsurgical Transplantation Models This section summarizes a few more complex microsurgical models used for transplantation research. For more detailed description of techniques and methods, see the references at the end of this chapter. • Lung transplantation (Asimacopoulos, Molokiha, and Peck 1971; Mark and Wildevuur 1981) • Orthotopic liver transplantation (Kamada and Calne 1979; Lee et al. 1973; Hansen, Kim, and Lie 1979; Houssin et al. 1979) • Small intestine transplantation (Monchick and Russel 1971; Deltz and Thiede 1985) • Pancreas transplantation (Lee et al. 1972; Nolan et al. 1983) • Esophagus replacement (Parsa and Spira 1978; Uchida and Harii 1989) • Multivisceral grafts, including liver, pancreas, stomach, omentum, small intestine, and colon (Murase et al. 1990)
84550.indb 606
11/11/10 6:28:49 PM
Microsurgical Procedures in Experimental Research
ICV
ICV
607
A
A
Figure 20.10â•…(Top) Donor nephrectomy for heterotopic left kidney transplantation with an aortic cuff and a vein patch of the inferior caval vein. (Bottom) End-to-side anastomosis from donor to recipient aorta and from the renal vein to the inferior caval vein. Insertion of the ureter into the bladder.
Concluding Remarks and Comments Technical procedures and methods used in experimental microsurgery are essentially the same as those applied in the clinic; however, the basic approach is principally different. In the clinic, the surgeon faces an individual problem with a subsequent optimal solution to this problem. Experimental microsurgery tries to elucidate a specific biological question where the microsurgical procedures per se are not actually interesting, and this implies a preferably 100% standardization and reproducibility of the individual experiments. The aim of these studies is the observation and
84550.indb 607
11/11/10 6:28:51 PM
608
Handbook of Laboratory Animal Science
ICV
A
(a) ICV
A
(b) Figure 20.11â•…(a) Donor nephrectomy for orthotopic left kidney transplantation with division of the renal artery, vein, and ureter. (b) End-to-end anastomosis of renal vessels after recipient nephrectomy and end-to-end ureter anastomosis using the splint technique.
determination of biological variation, rather than the monitoring of intraoperative modifications. In other words, detailed repetition of procedures is necessary, and only experimental series with high success rates are scientifically acceptable. There is no mystery about successful performance of microsurgical procedures. It is the result of training, perfectionism, and persistence. The basic rule is “as simple as possible, as fast as possible.” Projects of major scientific interest and value that include microsurgical models are based on an interdisciplinary approach to often complex problems. Cooperation and teamwork are therefore essential. Care has to be taken with respect to animal observation and housing. If optimal laboratory facilities and full-time technical assistance are not available or are neglected, time-consuming and difficult microsurgical procedures can end up surgically successful but with disappointing and useless results from a scientific point of view.
84550.indb 608
11/11/10 6:28:52 PM
Microsurgical Procedures in Experimental Research
609
References Abbott, C. P., E. S. Lindsey, O. Creech, Jr., and C. W. DeWitt. 1964. A technique for heart transplantation in the rat. Archives of Surgery 89:645. Acland, R., and P. Smith. 1976. Microvascular surgical techniques used to provide skin cover over an ununited tibial fracture. Journal of Bone and Joint Surgery 58:471. Asimacopoulos, P. J., F. A. S. Molokiha, and C. A. S. Peck. 1971. Lung transplantation in the rat. Transplant Proceedings 3:583. Bernhard, A., and W. Konertz. 1983. Experimental heart transplantation. Journal of Thoracic and Cardiovascular Surgery 86:314. Biemer, E., and U. Schmidt-Tintemann. 1982. Anatomische und funktionelle Grundlagen für die Wahl von Nahtmaterialien und Nahttechniken in der klinischen Mikrochirurgie. In Nahtmaterialien und Nahttechniken, ed. A. Thiede and H. Hamelmann, 400. Heidelberg: Springer–Verlag. Bunke, H. J., Jr., and W. P. Schulz. 1965. Experimental digital amputation and reimplantation. Plastic and Reconstructive Surgery 36:62. Carrel, A. 1902. La technique opératoire des anastomoses vasculaire et la transplantation des viscères. Lyon Medical 98:859. Carrel, A., and C. C. Guthrie. 1906. Complete amputation of the thigh with replantation. American Journal of Medical Science 131:297. Cobbet, J. 1967. Small vessel anastomosis. British Journal of Plastic Surgery 20:16. Daniller, A., R. Buchholz, and R. A. Chase. 1968. Renal transplantation in rats with use of microsurgical techniques: A new method. Surgery 63:956. Deltz, E., and A. Thiede. 1985. Microsurgical technique for small-intestine transplantation. In Microsurgical models in rats for transplantation research, ed. A. Thiede, E. Deltz, R. Engemann, and H. Hamelmann, 51. Heidelberg: Springer-Verlag. Engemann, R., E. Deltz, and A. Thiede. 1982. Nahtmaterialien und Nahttechniken in der experimentellen Mikrochirurgie. In Nahtmaterialien und Nahttechniken, ed. A. Thiede and H. Hamelmann, 90. Heidelberg: Springer–Verlag. Fabre, J., S. H. Lim, and P. J. Morris. 1971. Renal transplantation in the rat: Details of a technique. Australian and New Zealand Journal of Surgery 41:69. Fisher, B., and S. Lee. 1965. Microvascular surgical techniques in research, with special reference to renal transplantation. Surgery 58:904. Fujino, T., and F. Aoyagi. 1975. A method of successive interrupted suturing in microvascular anastomoses. Plastic Reconstructive Surgery 55:240. Günther, E. 1985. Immunogenetic aspects of organ transplantation in the rat. In Microsurgical models in rats for transplantation research, ed. A. Thiede, E. Deltz, R. Engemann, and H. Hamelmann, 83. Heidelberg: Springer–Verlag. Guthrie, C. C. 1908. Some physiologic aspects of blood vessel surgery. Journal of the American Medical Association 51:1658. Hansen, H. H., Y. Kim, and T. S. Lie. 1979. Orthotopic liver transplantation in the rat—Special reference to arterialization. Excerpta Medica International Congress Series 465:394. Harashina, T. 1977. Use of the untied suture in microvascular anastomosis. Plastic Reconstructive Surgery 59:134. Heron, I. 1971. A technique for accessory cervical heart transplantation in rabbits and rats. Acta Pathologica Microbiologica Scandinavica [A] 79:366. Horn, J. S. 1964. Successful reattachment of a completely severed forearm. Lancet 1:1152. Houssin, D., M. Gigou, D. Franco, A. M. Szekely, and H. Bismith. 1979. Spontaneous long-term survival of liver allografts in inbred rats. Transplant Proceedings 11:567. Jacobson, J. H., and E. L. Suarez. 1960. Microsurgery in anastomosis of small vessels. Surgery Forum 11:243. Jakubowski, H. D. 1985. Renal transplantation in the rat. In Microsurgical models in rats for transplantation research, ed. A. Thiede, E. Deltz, R. Engemann, and H. Hamelmann, 47. Heidelberg: Springer–Verlag. Kamada, N. 1985. A description of cuff technique for renal transplantation in the rat. Transplantation 39:93. Kamada, N., and R. Y. Calne. 1979. Orthotopic liver transplantation in the rat. Transplantation 28:47.
84550.indb 609
11/11/10 6:28:53 PM
610
Handbook of Laboratory Animal Science
Kleinert, H. E., and M. L. Kasdan. 1963. Salvage of devascularized upper extremities including studies on small vessel anastomosis. Clinical Orthopedics 29:29. Konertz, W., M. Semik, and A. Bernhard. 1980. Left ventricular bypass in inbred rats—A new experimental model in microsurgery. Operative technique and hemodynamic evaluation. Thoracic Cardiovascular Surgery 28:277. Konertz, W., A. Thiede, and A. Bernhard. 1980. Heterotopic heart transplantation in rats—An improved technique of functional evaluation. Excerpta Medica International Congress Series 465:359. Krizek, T. J., T. Tani, J. D. DesPrez, and C. L. Kiehn. 1965. Experimental transplantation of composite grafts by microsurgical vascular anastomoses. Plastic Reconstructive Surgery 36:38. Lee, S. 1967. An improved technique of renal transplantation in the rat. Surgery 61:771. Lee, S., A. C. Charters, J. G. Chandler and M. J. Orloff. 1973. A technique for orthotopic liver transplantation in the rat. Transplantation 16:664. Lee, S., K. S. K. Tung, H. Koopmans, J. G. Chandler, and M. J. Orloff. 1972. Pancreaticoduodenal transplantation in the rat. Transplantation 13:421. Lee, S., W. F. Willoughby, C. J. Smallwood, A. Dawson, and M. J. Orloff. 1970. Heterotopic heart and lung transplantation in the rat. American Journal of Pathology 59:279. Lee, S. H., and B. Fisher. 1961. Portacaval shunt in the rat. Surgery 50:668. Mark, K. W., and C. R. H. Wildevuur. 1981. Lung transplantation in the rat. I. Technique and survival. Annals of Thoracic Surgery 34:74. McGregor, I. A., and G. Morgan. 1973. Axial and random pattern flaps. British Journal of Plastic Surgery 26:202. McLean, D. H., and H. J. Buncke. 1972. Autotransplant of omentum to a large scalp defect with microsurgical revascularization. Plastic Reconstructive Surgery 49:268. Mikaeloff, P. P., R. Levrat, P. Nesmoz, J. P. Rassat, M. Philippe, L. M. Dubernard, and A. Bel. 1969. Heterotopic liver transplantation in the rat. Value, technique, results of about 70 cases. Lyon Chiropractic 68:133. Miller, B. F., E. Gonzales, L. J. Wilchins, and P. Nathan. 1962. Kidney transplantation in the rat. Nature 194:310. Monchick, G., and P. S. Russel. 1971. Transplantation of small bowel in the rat: Technical and immunological considerations. Surgery 70:693. Murase, N., A. J. Demetris, D. G. Kim, S. Todo, J. J. Fung, and T. E. Starzl. 1990. Rejection of multivisceral allografts in rats: A sequential analysis with comparison to isolated orthotopic small-bowel and liver grafts. Surgery 108:880. Niimi, M. 2001. The technique for heterotopic cardiac transplantation in mice: Experience of 3000 operations by one surgeon. Journal of Heart Lung Transplant 20:1123. Nolan, M. S., N. J. Lindsey, C. P. Savas, A. Herold, S. Beck, D. N. Slater, and M. Fox. 1983. Pancreatic transplantation in the rat. Long-term study following different methods of management of exocrine drainage. Transplantation 36:26. Nylén, C. O. 1972. The otomicroscope and microsurgery 1921–71. Acta Otolaryngologica 73:453. O’Brien, B. M., A. M. MacLeod, J. W. Hayhurst, and W. A. Morrison. 1973. Successful transfer of a large island flap from the groin to the foot by microvascular anastomoses. Plastic Reconstructive Surgery 52:271. Ono, K., and E. S. Lindsey. 1969. Improved technique of heart transplantation in rats. Journal of Thoracic Cardiovascular Surgery 57:225. Parsa, F. D., and M. Spira. 1978. Experimental esophagal reconstruction in rats with a free groin flap. Plastic Reconstructive Surgery 62:271. Reemtsma, K., N. Gialdo, D. A. Depp, and E. J. Eichwald. 1968. Islet cell transplantation. Annals of Surgery 168:438. Richter, M., H. Richter, M. Skupin, F. W. Mohr, and H. G. Olbrich. 2001. Do vascular compartments differ in the development of chronic rejection? AT1 blocker Candesartan versus ACE blocker Enalapril in an experimental heart transplant model. Journal of Heart Lung Transplant 20:1092. Salomon, S., D. A. Steinbrüchel, B. Nielsen, and E. Kemp. 1996. Hamster to rat kidney transplantation: Technique, functional outcome and complications. Urology Research 24:211. Silber, S. J. 1979. Microsurgical technique. In Microsurgery, ed. S. J. Silber, 1. Baltimore, MD: Williams and Wilkins. Steinbrüchel, D., H. Dieperink, E. Kemp, H. Starklint, and S. Larsen. 1987. Rat kidney allotransplantation without warm ischemia. Postoperative recovery and glomerulotubular function. European Surgery Research 19-S1:80.
84550.indb 610
11/11/10 6:28:53 PM
Microsurgical Procedures in Experimental Research
611
Steinbrüchel, D. A., S. Larsen, T. Kristensen, H. Starklint, C. Koch, and E. Kemp. 1992. Survival, function, morphology and serological aspects of rat renal allografts. Effect of short-term treatment with cyclosporin A, anti-CD4 and anti-interleukin-2 receptor monoclonal antibodies. Acta Pathologica Microbiologica et Immunologica 100:682. Steinbrüchel, D. A., H. H. Madsen, B. Nielsen, S. Larsen, C. Koch, J. C. Jensenius, C. Hougesen, and E. Kemp. 1990. Treatment with total lymphoid irradiation, cyclosporin A and a monoclonal anti-T-cell antibody in a hamster-to-rat heart transplantation model. Transplant International 3:36. Steinbrüchel, D. A., B. Nielsen, S. Salomon, and E. Kemp. 1996. Heterotopic cardiac transplantation in rodents: A new model with graft atrial septectomy. In Recent advances in experimental microsurgery, ed. T. Kajimoto and H. Kitatani, 183–186. Kanazawa: PSKMU. Strauch, B., and D. E. Murray. 1967. Transfer of composite graft with immediate suture anastomosis of its vascular pedicle measuring less than 1 mm in external diameter using microsurgical techniques. Plastic Reconstructive Surgery 40:325. Uchida, L., and K. Harii. 1989. Experimental replacement of the cervical esophagus in rats with a jejunal free transplantation. Laryngoscope 99:837. Zimmermann, F. A., K. Obermüller, J. M. Gokel, and S. Dorn-Kling. 1981. Die Gallengangrekonstruktion bei der Ratte durch Choledocho-Choledochostomie über einen verlorenen Drain. Zeitschrift für Experimentelle Chirurgie 14:241.
84550.indb 611
11/11/10 6:28:53 PM
84550.indb 612
11/11/10 6:28:53 PM
Chapter 21
Postmortem Procedures
Ricardo E. Feinstein and Kimberly S. Waggie Contents Introduction..................................................................................................................................... 614 Necropsy and Laboratory Safety.................................................................................................... 614 Instruments and Materials......................................................................................................... 615 Description of Lesions............................................................................................................... 616 Guidelines for the Necropsy of Rodents and Rabbits................................................................ 616 External Examination; Skin and Subcutaneous Tissues....................................................... 617 Abdomen and Pelvis.............................................................................................................. 618 Male Genital Organs and Urinary Bladder........................................................................... 618 Female Genital Organs and Urinary Bladder....................................................................... 620 Spleen and Pancreas.............................................................................................................. 620 Stomach and Intestines.......................................................................................................... 620 Liver, Kidneys, and Adrenal Glands..................................................................................... 620 Mouth, Neck, and Thorax..................................................................................................... 621 Head and Spinal Cord........................................................................................................... 622 Muscles and Joints................................................................................................................. 624 Sampling Techniques...................................................................................................................... 624 Sampling for Morphological Examinations............................................................................... 625 Histology............................................................................................................................... 625 Sample Collection for Microbiology.......................................................................................... 626 Parasitology........................................................................................................................... 626 Sampling for Cultivation....................................................................................................... 627 Sampling for Serology............................................................................................................... 629 Sampling for PCR in Diagnostic Microbiology......................................................................... 629 Sampling for Nutrient Analysis and Toxicology........................................................................ 629 Conclusion....................................................................................................................................... 630 Acknowledgment............................................................................................................................ 630 References....................................................................................................................................... 630 Useful Web Pages........................................................................................................................... 633
613
84550.indb 613
11/11/10 6:28:53 PM
614
Handbook of Laboratory Animal Science
Introduction Postmortem procedures (PMPs) include necropsy, collection of samples and tissue specimens, and recording of lesions observed. The major uses of PMPs are in diagnosis, health quality control, and toxicology studies (Everitt and Gross 2006). In biomedical research, PMPs are also essential to many experiments, not merely as techniques for collecting samples and specimens for examination, but also as means for improving experimental reliability. Postmortem procedures are indispensable to evaluation and characterization of mutant animals, and detailed descriptions of PMPs in work with transgenic animals have been published (Wood 2000; Relyea et al. 2000; Brayton, Justice, and Montgomery 2001; Ward et al. 2000). Sampling techniques may considerably influence the results of an experiment. It is best to define what is being looked for and, in collaboration with the study pathologist, the methods to evaluate the change before starting the study. The protocols for the sacrifice and necropsy of the animals, tissue collection, time required to perform PMPs, and types of fixatives and fixation times are some of the topics that must be thoroughly considered in the design of animal experiments (for details on animal euthanasia, see Chapter 20, this volume). Necropsies should follow a standard protocol, which should be available when PMPs are being performed. Obviously, any deviation from the protocol should be documented (Everitt and Gross 2006; Mann, Hardisty, and Parker 2002; Bucci 2002). In laboratory animals, infections by pathogenic organisms and the resulting lesions are uncontrolled variables that should always be investigated. The additional costs to research of diagnostic necropsies are likely to be low compared to the adverse effects of a permissive attitude (i.e., do not wait until several animals die to perform a diagnostic necropsy) (Jacoby and Lindsey 1997). Diagnostic necropsies, however, should be entrusted to individuals with specialized training or to diagnostic laboratories because diseases cannot be investigated without knowledge of lesions, causative factors, and disease mechanisms. The availability of specialized laboratories, such as histology, microbiology, clinical chemistry, and serology, is essential for disease diagnosis as well as for thorough evaluation of experimental models and toxicology studies. Although most diagnostic necropsies are prompted by detection of sick or dead animals at the breeding colony, during transportation, upon arrival at their site of use, or during the course of experiments, clinical signs may or may not be observed. Apparently healthy animals are not necessarily free from lesions that could hamper procedures or influence experimental results. The impact of infections will depend on the natures and aims of experiments, but researchers should be aware that a careful postmortem examination of the animals, including seemingly healthy individuals, is the most effective way to answer the question of whether complicating lesions are present. The systematic use of PMPs in diagnosis and health quality control will result in better laboratory animals. Defined, high-quality animals and carefully designed experiments also further reduce the number of animals required for experimentation and testing (Homberger et al. 1999; Nicklas et al. 2002). Although this chapter focuses on necropsy of laboratory rodents and rabbits, the general principles presented are relevant to all mammals. Reviews specifically pertaining to necropsy of domestic, wildlife, and laboratory animals have been published (Relyea et al. 2000; Cooper and Cooper 2007; King et al. 2005; Munson 1988; Strafuss 1988; Feldman and Seely 1988; Seymour et al. 2004). Necropsy and Laboratory Safety Necropsies should be performed in a specially equipped room because cadavers and tissue specimens are potential sources of infection to humans and animals. Due to the risk of microbial contamination and also of exposure to allergens and high concentrations of harmful substances such as anesthetics, fixatives, and solvents, containment facilities and strict adherence to hygiene practices are necessary. Containment standards for postmortem rooms, necropsy, and cleaning and disposal
84550.indb 614
11/11/10 6:28:53 PM
Postmortem Procedures
615
procedures have been published (Canadian Food and Inspection Agency 2009), and two most useful biosafety manuals (USDHHS 2007; WHO Laboratory 2004) may be found on the World Wide Web (WWW) at http://www.inspection.gc.ca/english/sci/lab/convet/convete.shtml and http://www. who.int/csr/resources/publications/biosafety/WHO_CDS_CSR_LYO_2004_11/en/print.html, and at http://www.unh.edu/ehs/pdf/CDC-NIH-BMBL-5.pdf. Methods of decontamination, cleansing routines, and personal hygiene should be described in the necropsy laboratory’s procedure manual. PMPs should be performed according to accepted laboratory practices. It is not within the scope of this chapter to provide a complete list of safety measures in necropsy work, but some measures will be described. Briefly, in necropsy work, closed-front protective clothing and surgical gloves, as well as protective glasses, should be worn when working close to the surface of organs. Specialized gloves may be necessary when working with toxic agents (a review of these may be found at http://physchem.ox.ac.uk/ MSDS/glovesbychemical.html). Wristwatches, bracelets, or rings should not be worn. Instruments and all other necessary equipment should be prepared before starting the necropsy. The exteriors of tubes, containers, plastic bags, etc. must be protected from contamination and spills. These items should be placed within reach but not beside the cadaver. Other objects, such as telephones, doorknobs, pencils, etc., should not be touched during the necropsy. No one performing a necropsy should ever pipette by mouth, touch unprotected body areas, or apply contact lenses or wear them without protective eyewear such as goggles. Also, one should not eat, drink, smoke, or apply cosmetics in the necropsy room. The necropsy of animals known or suspected to harbor zoonotic agents or organisms hazardous to humans or other animals may require additional containment systems and safety measures, such as the use of biological safety cabinets, laminar flow hoods, or personal protection equipment like a closely fitting, ventilated helmet provided with high-efficiency filters. After the necropsy, hands should be washed thoroughly. Protective clothing and gloves must always be taken off when leaving the necropsy room. Local regulations regarding disposal procedures and labeling of containers of biological material should be consulted. Cadavers and tissues can be autoclaved or incinerated. Contaminated disposable items, such as gloves, should be sterilized or transported in leak-proof containers to an appropriate plant for sterilization and destruction. Used needles, scalpel blades, and glass waste should be placed in appropriate “sharps” containers for decontamination and destruction according to local biosafety regulations (Canadian Food Inspection Agency 2009; USDHHS 2007; WHO Laboratory 2004; Demiryürek et al. 2002). The design, building materials, equipment, and environment of the necropsy room require special attention. Surfaces, including floors, walls, necropsy table, safety cabinets, laminar flow hoods, or working benches, must withstand frequent washing and be easy to clean and disinfect. Electrical equipment, such as computers and balances for weighing animals and organs as well as electrical wiring and sockets, must be shielded from liquids. Since animals for necropsy may be submitted alive, equipment and instruments for euthanasia should be available. The use of inhalational anesthetics imposes additional design considerations such as space for anesthetic machines, gas cylinders, or hose connections and proper venting of waste gases. Water services, lighting (intensity and spectrum of light and position of light sources), room ventilation and air movements, and temperature are some of the factors to be considered (Howard and Foucher 2009). The ergonomics of necropsy work is important because personnel may spend many hours performing movements or assuming awkward positions that could result in injury from repetitive motion, postural, or articular factors. Instruments and Materials Surgical instruments are appropriate for most necropsies, although certain procedures, such as dissection of very small organs, require microsurgical instruments. Tubes and containers for
84550.indb 615
11/11/10 6:28:53 PM
616
Handbook of Laboratory Animal Science
samples must be identified clearly and indelibly (not on the covers). If necessary, a code number should be used. The following materials and instruments are commonly used: • • • • • • • •
Sharp knife, scalpel blades, and handle Dissecting scissors and small operating scissors Bone-cutting forceps, serrated forceps, and toothed forceps Sterile instruments for collecting samples for microbiological testing Syringes (1, 2, 5, and 10 mL) and needles Tubes for liquid samples (3, 5, and 12 mL) Container of fixative (for routine fixation of tissues in 10% buffered neutral formaldehyde solution) Leak-proof containers for tissue specimens (bacteriology, mycology, parasitology, virology, polymerase chain reaction [PCR], chemistry) • Squeeze bottle of 70% alcohol and squeeze bottle of saline • Swabs, for sampling purposes (see sampling techniques) • Plastic bags of various sizes and paper towels
A balance or balances having weight ranges suitable for cadavers and organs are part of the minimal equipment of the necropsy laboratory. A stereoscopic microscope is an invaluable aid in examining small animals, organs, lesions, and some parasites. An electric drill with a cutting disk is a practical aid for cutting bony structures and teeth (protective glasses must always be worn when such drills are used). Description of Lesions Lesions observed at necropsy must be documented. The written lesion description should allow a reader to form a mental picture of the changes. The location, appearance, number, and severity of the lesions should be described in a precise and concise manner. The location of lesions must be described according to the organ and lobe, area of the skin, portion of the intestines, etc. Anatomical structures are used as reference points (Strafuss 1988). For paired organs, it should be mentioned which of them is affected. The appearance of lesions is described in terms of size, shape, color, appearance of the surface and of the cut surface, consistency, demarcation from surrounding tissues, and distribution (e.g., focal, multifocal, diffuse). The size should be measured in two or three dimensions in linear units (millimeters or centimeters), volume (milliliters), weight (grams), or relative weight. In hollow organs and lesions, the amount, appearance, and odor of the contents should be detailed. Diagnosis of the lesion should be made separately from the description. Photographing tissue specimens is a most useful aid for description, documentation, and teaching purposes. Each photograph should include a size marker, such as a ruler, for reference. Digital cameras take high-quality pictures and are a valuable tool for photography of biomedical specimens (Edwards 1988; Weinberg 1997). In addition, modern communication technologies allow a distant observer instantly to visualize digitized images, not only of gross lesions but also of whole histological slides. The applications, benefits, and possibilities of these rapid advances in the field of telepathology have been described (Saikia, Gupta, and Saikia 2008). Guidelines for the Necropsy of Rodents and Rabbits Recently killed animals are preferable for necropsy, to avoid the effects of autolysis and putrefaction. If the necropsy is delayed, the cadaver should be refrigerated. Freezing of cadavers should be avoided because the freeze–thaw cycle causes marked tissue damage. The anatomical differences between rodents and rabbits do not preclude using a similar necropsy technique. The general necropsy protocol can be modified, depending on the aims of the studies, but changes of the protocol must always be documented, preferably prior to the start of the necropsy.
84550.indb 616
11/11/10 6:28:54 PM
Postmortem Procedures
617
The order in which the necropsy is conducted depends on a number of factors, including purpose of the necropsy, samples to be collected for further study, and personal preference. For example, if tissues that autolyze rapidly, such as the pancreas and digestive mucosa, are to be included in the study, they should be collected and preserved as soon as possible. However, the prosector may wish to take samples for bacteriology immediately from the respiratory tract to minimize the chance of contamination with the normal intestinal flora. Detailed guidelines for the necropsy of rodents and rabbits have been published (Everitt and Gross 2006; Relyea et al. 2000; Feldman and Seely 1988; Seymour et al. 2004). General guidelines for the necropsy of mice, which also are applicable to other rodents, are presently available on the World Wide Web at “the virtual mouse necropsy” page at http://www3.niaid.nih.gov/labs/aboutlabs/ cmb/InfectiousDiseasePathogenesisSection/mouseNecropsy/. Special techniques are sometimes used in conjunction with the necropsy to evaluate specific tissues or processes. For example, techniques of vascular perfusion of rodents for adequate fixation of the nervous tissues and also to preserve embryos have been described (Casella, Hay, and Lawson 2007). Vital staining, the use of colored substances injected into the vascular system, has been employed to investigate vascular permeability to different substances. Vital dyes have been found useful in the search for delicate or very small organs such as the thoracic duct or the paraganglia (LeVeen and Fishman 1947; Coleridge, Coleridge, and Howe 1967; Richardson 1969; Clasen, Pandolfi, and Has 1990; McDonald and Blewett 1981). Another vital dye, 2,3,5-triphenyltetrazolium chloride (TTC), has been used in myocardial infarct and stroke studies to differentiate viable from dead tissue (Bederson et al. 1986; Greve and Saetersdal 1991). Small organs, such as the pituitary gland, adrenals, or lymph nodes, may be placed in histological cassettes directly after removal and immersed in fixative to prevent loss as well as drying. For a general necropsy, one convenient method is to examine the organs in the following order: external examination, skin and subcutaneous tissues, abdomen, pelvis, mouth, neck, thorax, head, spinal cord, muscles, and joints. After removal from the body, the organs should be laid out on the necropsy table for further examination and sampling. External Examination; Skin and Subcutaneous Tissues A general necropsy is started by reviewing background information and inspecting the cadaver. The animal species, strain, animal identity (i.e., necropsy number, ear tags, tattoos, etc.), sex, and body weight are recorded. Postmortem changes are scored; for example, on a scale ranging from one to five, one corresponds to mild decomposition of tissues, such as that observed in animals sacrificed immediately before the necropsy, and five corresponds to a pronounced autolysis of tissues. The appearance of the skin, hair coat, body openings, and visible mucous membranes is observed. Abnormalities such as loss of hair, changes in the color of the skin or mucous membranes (icterus, cyanosis, pallor, etc.), presence of discharges from natural orifices, and masses are noted. In hamsters, the flank organs, which are a male secondary sexual characteristic, are inspected. The skin and subcutaneous tissues are palpated for abnormalities. The amount of fat in the body depots and the muscular volume are observed to score the nutritional state of the animal. The body condition may be rapidly scored by observing the volume of muscular masses covering osseous protuberances: (1) emaciation, absence of fat in the body depots; (2) undernourished, bad nutritional condition; (3) good nutritional condition; (4) over condition; and (5) obesity. This method is also applicable for body condition scoring of live animals (UllmanCulleré and Foltz 1999). The cadaver is placed on its back and pinned to a dissection board (in rabbits, this is not necessary). The skin is moistened with alcohol, and a midline incision is made from the symphysis of the mandible to the anus, avoiding the penis in male animals. The skin is reflected on both sides of the incision (Figure€21.1). The subcutaneous tissues are inspected. Skinning of the cadaver can be completed at
84550.indb 617
11/11/10 6:28:54 PM
618
Handbook of Laboratory Animal Science
Figure 21.1â•…After doing a midline incision, the skin is reflected on both sides. The arrows point at the clitoral glands.
this stage or at the end of the necropsy. In guinea pigs and autolyzed cadavers, the digestive tract is friable and the stomach and the intestines may rupture during the skinning procedure. The mandibular and cervical lymph nodes, the salivary glands (mandibular, parotid, and sublingual), and the extraorbital lacrimal glands are observed. The mammary glands are inspected (mice have five pairs of mammary glands, rats six, hamsters six or seven, guinea pigs one, and rabbits four pairs). In female rodents, the clitoral glands are inspected (Figure€21.1). In males, the penis, prepuce, and the preputial glands are examined. The inguinal and axillary lymph nodes (usually embedded in subcutaneous fat tissues) are observed. Abdomen and Pelvis The abdomen is opened by a midline incision through the abdominal wall, from the sternum to the pelvis, and by two cuts through the muscles along the costal arcs (Figure€21.2). Care should be taken to avoid unintentional incision of the underlying viscera. The floor of the pelvis is removed after making a sagittal cut on each side of the midline (Figure€ 21.3). The abdominal and pelvic organs are examined in situ. The appearance of the serous membrane and the occurrence of abnormal contents, such as serous fluid, blood, fibrin, or adhesions between organs, are observed. Male Genital Organs and Urinary Bladder The scrotum is cut open, and the testicles and epididymides are extracted. The fibrous ligaments anchoring the tail of the epididymis to the scrotum are cut. The vas deferens is cut, and the
84550.indb 618
11/11/10 6:28:55 PM
Postmortem Procedures
619
Figure 21.2â•…The abdominal cavity with the organs in situ. Samples for microbial culture should be taken before handling the viscera.
testicles and epididymides are removed for inspection. The ureters are cut and the remaining genital organs are removed in block—that is, seminal vesicles, coagulating glands, bulbourethral glands, prostate gland, urethra, and penis (male rodents have large accessory sex glands that should not be mistaken for uterine horns). The urine is collected from the urinary bladder. Rabbit and guinea pig urine is normally turbid due to its high concentration of mineral crystals. In rodents, the presence
UT-H
Rectum
UT-H
Figure 21.3â•…Pelvis with the organs in situ. UT-H: uterine horn. The arrow points at the urinary bladder.
84550.indb 619
11/11/10 6:28:57 PM
620
Handbook of Laboratory Animal Science
of a urethral plug (a whitish, rather hard cast) in the lumen of the proximal urethra and sometimes extending into the bladder is considered normal in sexually mature males (Kunstyr et al. 1982). Female Genital Organs and Urinary Bladder The ovaries with the oviducts are located caudal to the kidneys in the peritoneal cavity. The genital organs and supporting ligaments are partly embedded in fat, but the ovaries stand out from the fat tissues as more reddish in color. The vulva and vagina are dissected free from the skin and rectum. The vagina is caught with a forceps, and the supporting ligaments of the vagina, uterus, oviduct, and ovary are cut. The genital organs and the urinary bladder may be opened for examination of the mucous membranes. Spleen and Pancreas The spleen is removed by cutting the omentum and the ligamentum gastrolienalis along the greater stomach curvature. The pancreas is examined. It is a rather diffuse and richly lobulated organ located in the supporting ligaments of the stomach and the small intestine. It is firmer and grayer in color than fat tissues. Stomach and Intestines In mice, rats, hamsters, and gerbils, the stomach is divided into two distinct regions: the foreÂ� stomach (also designated as the cardiac, cutaneous, or proventicular region) and the glandular region. The glandular region has a thicker wall than the forestomach. In the guinea pig, these regions are not clearly demarcated. The rabbit stomach is not divided into distinct areas (Ghoshal and Bal 1989). The intestines of rodents have few unique features (in guinea pigs, the cecum is a thin-walled, voluminous organ). Rabbits have long intestines, with most of the gut-associated lymphoid tissues located in the last portion of the small intestine (sacculus rotundus of the ileocecal tonsil) and in the appendix vermiformis, a cecal diverticulum. The intestinal wall is thicker at the level of the sacculus rotundus and the appendix than in the rest of the intestines. To remove the gastrointestinal tract, the anus is dissected free from the surrounding skin. The rectum is gently caught with a forceps, and the mesentery and the supporting ligaments are cut. The supporting ligaments along the whole length of the large intestine, the small intestine, and the stomach are cut. It is convenient to cut the supporting ligaments as close to the intestines as possible. The stomach is caught with a forceps and, while the stomach is gently pulled away from the diaphragm, the esophagus is cut. Then the whole gastrointestinal tract is removed, taking care not to spill contents onto the other organs, and placed on the necropsy table for examination and sampling (Figure€21.4). Specimens and samples for further studies may be taken at this stage, to minimize the impact of autolysis on the gut mucosa. The stomach is opened along the greater curvature. If appropriate, the whole length of the intestines is opened, and the gastrointestinal contents are collected and examined. The digestive mucosa is inspected, including the Peyer’s patches. Lesions in the mucosa, such as erosions, can be covered by adherent contents. The surface can be rinsed with saline to remove the contents. The mesentery, including the mesenteric lymph nodes, is then removed. Liver, Kidneys, and Adrenal Glands The liver is removed by cutting the hepatic ligaments, and the hepatic tissues can be examined by deep cuts in different lobes. The gall bladder and its contents are examined (rats do not have this organ). The right and left adrenals, which are located cranially and medially to the respective
84550.indb 620
11/11/10 6:28:57 PM
Postmortem Procedures
621
Stomach Duod
Ileum
Cecum Colon
Jejun
Rectum Anus
Figure 21.4â•…Gastrointestinal tract placed on the necropsy table for examination. Duod: duodenum; Jejun: jejunum. The arrows point at some of the Peyer’s patches.
kidney, are removed together with the kidneys. Each kidney is removed by cutting the ureters and the renal vessels at the level of the renal hilus. The renal capsule is removed, and the appearance of the cortical surface is observed. If possible, the kidneys are cut and the cortex, medulla, and the renal pelvis are examined. If the right kidney is cut transversally and the left kidney longitudinally, it will be possible to distinguish between them in histological sections (Relyea et al. 2000). Mouth, Neck, and Thorax The mandibular muscles are cut on both sides, and the mandible is pulled backward and removed, if necessary. The oral cavity is inspected. The larynx, trachea, and esophagus are dissected by cutting the muscles in the ventral part of the neck. To open the thorax, the xiphoid cartilage is lifted with a forceps and the sternum removed by cutting on both sides along the costochondral junctions (Figure€21.5). The sternum is a convenient tissue for histological examination of the bone marrow. The thorax and its organs are inspected in situ. In mice, hamsters, and rabbits, the thymus is located in the anteroventral portion of the thorax, close to the midline. Rats, in addition, have a smaller cervical portion, which lies ventrally to the trachea. In guinea pigs, the thymus is entirely located in the ventral part of the neck. The tongue is caught with a forceps and pulled backward, and the soft palate and pharynx are cut with a scalpel. The tongue, esophagus, larynx, trachea with thyroid and parathyroid glands, thymus, mediastinal and bronchial lymph nodes, lungs, and heart are removed in block by gently pulling backward after severing the thoracic aorta and the caudal vena cava at the level of the diaphragm. In small animals such as mice and hamsters, it is often best to harvest the heart and lower respiratory tract intact; after the gross inspection, these organs are fixed to be evaluated by microscopic examination. If the heart is inspected grossly, the pericardial sac is opened first and its contents examined. In larger species, before the heart is opened, the left and right sides should be identified. The right atrium may be opened by making an incision from the sinus venosus into the auricle. The
84550.indb 621
11/11/10 6:28:58 PM
622
Handbook of Laboratory Animal Science
TRA
THY
LU
Heart
LU
Figure 21.5â•…The neck and thorax with the sternum removed and thoracic organs in situ. TRA: trachea; THY: thymus; LU: lung. The thyroids are marked by arrows.
incision is extended into the right ventricle by cutting its wall parallel to the interventricular septum toward the apex, and from the apex to the pulmonary artery. The left atrium is opened, starting the incision at the entrance of the pulmonary veins. The wall of the auricle is cut, and the incision is extended into the left ventricle along the interventricular septum toward the apex, and from the apex into the aortic artery. The myocardial cut surface, the heart cavities, and the atrioventricular and semilunar valves are inspected. Clots are removed and the endocardium is inspected. The aortic trunk is opened and, if necessary, the whole length of the aorta. The whole length of the esophagus, the larynx, the trachea, and the major bronchi are then opened, and the mucosa of these organs, as well as the cut surface of the lungs, are examined. Head and Spinal Cord The skin is cut transversally over the neck, and the cranium is skinned and severed caudally to the occipital protuberance. The cranium of young animals and of adult mice can be opened with scissors. For adult animals of other species, bone cutters are preferable. An electric drill with a cutting disk is also practical. The cranium is opened by first making two cuts from the foramen magnum to the medial part of both orbits. Then the frontal bone is cut transversally at a level just behind the orbits, which coincides with the anterior border of the cranial cavity. To extract the brain and cerebellum from the cranium, the calvaria is removed (Figure€21.6) and the olfactory lobes are cut with small operating scissors inserted under the anterior portion of the brain. The optic nerves and the remaining cranial nerves on both sides are then cut. The brain and cerebellum are removed (artifacts produced by the manipulation of fresh brain tissues can be avoided by perfusion fixation or fixation of the brain in situ). Detailed guidelines for the extraction of the central nervous system in adult rats and in embryos and fetuses have been published (Casella et al. 2007). After removal of the brain, the pituitary gland appears on the floor of
84550.indb 622
11/11/10 6:29:00 PM
Postmortem Procedures
623
Figure 21.6â•…Cranium with the brain in situ. The calvaria was removed by making two lateral cuts: from the foramen magnum to the medial part of both orbits, and a transverse cut just behind the orbits.
the cranium, attached to the sella turcica (Figure€21.7). In rodents, there is a thin layer of dura mater covering the pituitary gland. The dura around the gland is cut with a scalpel. Using the scalpel blade as a shovel, the gland is lifted out of the sella turcica. In rabbits, the pituitary gland is covered by a bony projection (dorsum sellae), which has to be removed prior to removal of the gland. The tympanic cavity is inspected. The wall of the tympanic bulla is thin and can be opened ventrally with scissors (rodents) or bone cutters (rabbits) after disinfection of the surface. The middle ear can be examined histologically after decalcification of the skull. Using small operating scissors or a scalpel inserted in the orbit, the eyeball is freed from the surrounding tissues, and the optic nerve is cut. A forceps is used to lift the eyeball, with the Harderian
Figure 21.7â•…Dorsal endocranial view of the skull after removal of the brain and cerebellum. The pituitary gland in the sella turcica is marked by an arrow.
84550.indb 623
11/11/10 6:29:01 PM
624
Handbook of Laboratory Animal Science
Figure 21.8â•…Ventral view of the spinal cord in situ, after removal of the vertebral bodies.
gland and the smaller intraorbital lacrimal gland, out of the orbit. In guinea pigs, the zygomatic salivary gland located ventral to the globe of the eye is also collected. To inspect the nasal cavity, the cranium can be divided by a midline cut, yielding sagittal sections. However, if the nasal region is to be examined histologically, it should be cut transversally together with palatine structures for reference points, after fixation and decalcification (Mery et al. 1994; Harkema, Carey, and Wagner 2006). The spinal cord should be fixed in situ because it is easily damaged if it is removed from the vertebral canal. In adult animals, except for mice, the spinal cord can be removed with small operating scissors or bone cutters. Starting at the first cervical vertebra, incisions are made alternately on the right and left side to remove the vertebral arches and then the roots of the spinal nerves are cut. An alternative is to approach the spinal cord ventrally by removing the vertebral bodies (Figure€21.8). Muscles and Joints The sublumbar muscles and the muscles of the thigh region are inspected by longitudinal and transverse cuts. The major limb joints are examined. The periarticular muscles are removed, the articular capsule is swabbed with alcohol, and the joints are opened with a scalpel. Samples for culturing may be obtained by scraping and swabbing the articular surfaces. If necessary, joints can be removed and fixed for histological examination. Sampling Techniques Once the in-life phase and gross necropsies for an animal study are completed, all that remains are the accumulated records and preserved specimens (Noel 1982). Correct sampling technique is essential to postmortem studies. The results obtained from experimental studies and diagnostic
84550.indb 624
11/11/10 6:29:02 PM
Postmortem Procedures
625
workups depend on the care with which findings are recorded and specimens are collected. Diagnostic laboratories use necropsy report checklists to organize the description of gross findings; a good example of a necropsy record and checklist may found at http://www.afip.org/consultation/vetpath/ ddpdf/dd1626.pdf. In the following sections, we briefly describe the most basic sampling techniques for histology, electron microscopy, microbiology, molecular biology, and chemical studies. These are general guidelines, but specific tissues, microorganisms, or substances may require particular techniques for which specialized literature or laboratories should be consulted. We have grouped the sampling methods into sampling for morphological examinations and sampling for microbiology. Sampling for Morphological Examinations Histology Histology is universally used to evaluate tissue changes in animals subjected to experimental setups or toxicology studies or for disease diagnosis. Microscopy of fresh specimens or fixed stained tissues is also a common tool to diagnose infections by demonstrating microbial agents directly or the lesions that they cause. Tissues for microscopic evaluation must be collected in a consistent manner according to diagnostic or experimental study protocol. Published guidelines for sampling and trimming procedures of specific organs constitute a most valuable aid for standardizing these procedures (Relyea et al. 2000; Ruehl-Fehlert et al. 2003; Kittel et al. 2004; Morawietz et al. 2004; Renne et al. 2009). Both macroscopically normal tissues, which may contain microscopic lesions, and lesions with surrounding normal tissue should be collected. Select organs may also be weighed prior to fixation. Prior to weighing, fat, excess blood, and other irrelevant tissue should be removed to ensure accuracy. Tissue specimens should be obtained with a scalpel or a sharp knife and should be handled carefully to avoid artifacts caused by stretching or compression of tissues. Artifactual distortion of tissues can destroy, alter, mimic, or mask changes. It is always advisable to check with the pathology laboratory to determine its preference for fixative and fixation procedures. For routine histology, tissue specimens are fixed by immersion in 10% buffered neutral formaldehyde solution for at least 24 h. The volume of fixative should exceed that of the specimens by at least 10 times (formaldehyde solution) or 20 times (alcohol-based fixatives). For an adequate penetration of the fixative, tissue specimens should not be thicker than 5–6 mm (Figure€21.9).
Figure 21.9â•…Liver specimen for histology. Tissue specimens for histology should not be thicker than 6 mm for an adequate penetration of the fixative.
84550.indb 625
11/11/10 6:29:03 PM
626
Handbook of Laboratory Animal Science
The choice of fixative as well as the duration of fixation is essential for immunohistochemistry. For example, overfixation of tissues to be used for immunohistochemistry can result in alteration of epitopes, rendering them undetectable by the desired antibody. For electron microscopy, specimens not thicker than 3 mm can be fixed in glutaraldehyde or Karnovsky fluid (Relyea et al. 2000). The use of a microwave oven can considerably shorten the time required for fixation and subsequent histological procedures. In addition, microwave oven heating of tissue sections often improves tissue antigen detection by immunohistochemistry in formalin-fixed, paraffin-embedded tissues (Shi, Key, and Kalra 1991). In situ hybridization (ISH) and in situ PCR are increasingly used as tools for diagnostics (e.g., detection of pathogens) and experimental biology (e.g., gene expression). A full review of these methods is beyond the scope of this chapter. However, it should be noted that most ISH and PCR studies can employ aldehydes for tissue fixation (Hofler and Mueller 1994; Teo and Shaunak 1995). The postmortem decomposition of tissues is a hindrance to microscopic studies. It occurs rapidly in the intestines (Scheifele, Bjornson, and Dimmick 1987; Seaman 1987). Intestinal specimens are best preserved when collected as soon as possible after the animal’s death by injecting formalin into the lumen of an unopened intestinal segment. The lumen should not be left distended because this results in altered morphology (Fenwick and Kruckenberg 1987). The “Swiss roll” is a practical technique that allows the examination of the whole intestine of rodents. Immediately after excision, the intestine is opened and divided into segments of a length that, when rolled, will easily fit into a tissue cassette. With the mucosa facing out, the intestine is rolled from end to end. The result resembles the head of a lollipop (Moolenbeek and Ruitenberg 1981). This procedure may also be performed on an unopened intestine. Optimum fixation of the lung also requires special attention. Immersion fixation can be used, but it may result in collapsed alveolar spaces that obscure detail, especially in small animals such as mice. Expansion of the lungs with fixative by perfusion through the trachea preserves the alveolar spaces and results in better sections for microscopic examination. Detailed guidance for the dissection and preparation of the respiratory organs for histological examination in inhalational studies may be found on the World Wide Web at http://www.oecd.org/dataoecd/45/2/43822718.pdf (Renne et al. 2009). Optimal preservation of tissue morphology and tissue antigens can be achieved by perfusion fixation. In this method, the animal is deeply anesthetized and the fixative is injected into the vascular system. The type and amount of fixative, perfusion pressure, and injection site will depend on various factors, including the aim of the study, the animal species, and body weight (Relyea et al. 2000; Casella et al. 2007). For small tissue specimens, a different method, based on freeze substitution and low-temperature plastic embedding, also results in high-quality morphology and optimum antigen preservation (Murray and Ewen 1991). Sample Collection for Microbiology Parasitology The sampling site is important in parasitological examinations because parasites have specific predilection sites. In addition, simultaneous infection with various types of parasites is not uncommon. Thus, skin scrapings for ectoparasites should be obtained from different areas. If possible, the whole skin should be examined, including ears, eyelids, and nasal cavity. The skin should be placed in a hermetically sealed container or in a plastic bag and refrigerated until the moment of examination. For endoparasites, samples from gastrointestinal contents should be collected during the necropsy. Samples should be placed in clean, leak-proof containers. Fresh samples are always preferable (Smith et al. 2007). Microscopic examination of gastrointestinal contents or feces (wet preparation) permits the identification of parasites, larvae, and eggs. Specimens examined immediately after collection
84550.indb 626
11/11/10 6:29:03 PM
Postmortem Procedures
627
are appropriate for the demonstration of motile protozoans or trophozoite stages of protozoans. Samples for examination of protozoan oocysts, helminth eggs, and adult helminths may be refrigerated (Smith et al. 2007; Lassen and Weiser 2004). Most endoparasites located in organs other than the intestine are diagnosed by microscopic examination of tissue sections. For certain parasites, such as the protozoan Encephalitozoon cuniculi, serological methods and immunohistochemical techniques are also available (Boot et al. 2000). Molecular assays for routine diagnosis of rodent pinworms are still uncommon. A PCR offered high specificity, but its sensitivity was lower than the detection of pinworms by necropsy followed by direct microscopic intestinal examination (Feldman and Bowman 2007). Blood smears are useful for examining for blood parasites. Smears should be of a good quality because no diagnostic skill can compensate for a poorly made blood film. Preferably, smears should be prepared from fresh blood, recently obtained, without anticoagulants. Films that are not stained immediately should be fixed and stored in a protected place. Sampling for Cultivation The results from postmortem cultures depend on the care with which specimens are collected. The time between euthanasia and necropsy is especially important in rodents because autolysis occurs rapidly. To avoid accidental contamination of tissues and the postmortem growth of contaminant bacteria, specimens for bacterial (and viral) culture and molecular techniques should be obtained before the organs are handled (i.e., as early as possible during the necropsy) (Figure€21.10). The laboratory receiving the samples should be consulted regarding which tissues to examine and the best conditions for transport of the specimens. Bacteria Most bacterial infections are diagnosed by bacterial culture. The successful isolation of bacteria depends on various factors, but a correct sampling technique is essential. Inadequate sampling techniques can result in the overgrowth of a causative agent by contaminant bacteria. Other important factors are the sampling site, type of disease and its duration, whether the animals have been treated with antibiotics, etc. The elapsed time since death should also be considered because the viability of pathogenic bacteria and mycoplasma in tissues decreases, while bacteria of the normal flora rapidly invade the tissues and may overgrow pathogenic agents. The agonal or postmortem invasion
Figure 21.10â•…Cutting a piece of liver for bacteriology. Samples for bacteriology should be taken with sterile instruments and preferably before the organs are handled.
84550.indb 627
11/11/10 6:29:04 PM
628
Handbook of Laboratory Animal Science
of tissues by resident bacteria must be distinguished from an infection occurring before death. Therefore, culture results must be evaluated together with necropsy findings, such as the presence of lesions consistent with those attributed to the bacteria isolated (Songer and Post 2008). Tissue specimens for bacterial culture should be obtained with sterile instruments or swabs. The selected body cavity is opened aseptically, and the surface of the tissues can be seared with a red-hot spatula or a flame. Bacterial specimens may consist of cut pieces of tissues (Figure€21.10). The instruments must be sterilized before collecting each specimen. In the case of hollow organs (intestine, uterus), a segment is cut after ligation at both ends. Sterile swabs inserted into the tissues are also used, but the swabs should be processed within a few hours. Cotton swabs should be avoided because substances present in the cotton may hinder the growth of certain bacteria. Swabbing is also convenient for sampling serosal and mucosal surfaces (e.g., pericardial sac, joints, genital tract, conjunctiva). Organs with a thick capsule as well as abscesses and pustules can be opened with a sterile scalpel after disinfection of the surface. The contents are then sampled by thorough swabbing against the inside of the capsule. Body fluids, such as urine or blood, can be sampled by swabbing or aspirated with a sterile needle and syringe (Songer and Post 2008). Obtaining samples for culturing mycoplasmas and bacteria by washing the respiratory and the genital mucosa has been described in detail (Cassell et al. 1983). Bacterial specimens from organs and serosal or mucosal surfaces can be transferred to a culture medium using a sterile loop. For print cultures, the surface of the tissues is pressed against a culture medium or blotting paper is pressed against the tissues and then transferred to a culture medium. General techniques and sampling procedures for bacteriological cultures from specific organs have been described (Songer and Post 2008; Cassell et al. 1983; Kornerup Hansen 2000). The use of transport media and the conditions of transport of the specimens to the laboratory are of considerable importance. Commercially available swabs containing transport media for specific purposes are a convenient way to submit specimens to distant laboratories. Swabs should not be allowed to dry because this rapidly reduces the viability of many bacteria and mycoplasmas. Transport in sterile phosphate-buffered saline (PBS) at 4°C has been found to maintain the stability of murine mycoplasmas and various pathogenic bacteria, except for Pasteurella multocida and P. pneumotropica (Shimoda et al. 1991). Specimens should be refrigerated or kept on wet ice until processed, to prevent bacterial overgrowth. Containers for transportation of cultural specimens should be leak proof and labeled using indelible ink. Specimens should be submitted to the laboratory with a form specifying type of examination required (i.e., cultivation of aerobic or anaerobic bacteria, or a specific pathogen), tissue submitted, animal species, sex, age, disease history, and other pertinent information, such as whether the animal is immune deficient or has been treated with antibiotics (Songer and Post 2008). Fungi Specimens for fungal culture, except dermatological specimens, should be kept moist with sterile distilled water or saline. Specimens should be obtained aseptically, as has been described for bacterial sampling. Dermatological specimens should be free from contamination by blood. Hairs and the base of hair shafts and skin scrapings can be collected in petri dishes after washing the skin with 70% alcohol. Skin scrapings should include the center and the periphery of lesions (Songer and Post 2008). Viruses Necropsy specimens for virus isolation should be collected aseptically, using the same collection methods as for bacterial culture (see above), placed in leak-proof sterile containers without preservative, and chilled. Specimens, except blood, should be stored frozen, preferably at –70°C. The
84550.indb 628
11/11/10 6:29:04 PM
Postmortem Procedures
629
postmortem decomposition of tissues inactivates many viruses. Thus, specimens should be obtained shortly after death, preferably during an early stage of infection. Electron microscopy is also a straightforward method to detect viruses in tissues and clinical specimens; the accuracy of virus identification may be enhanced by techniques of immunoelectron microscopy (Murphy et al. 1999). Sampling for Serology Serology is the method of choice for detecting antibodies to viruses and Mycoplasma pulmonis and for a few bacteria and protozoan parasites. It is commonly used in health monitoring programs and for diagnosis (Nicklas et al. 2002). A positive result indicates that an animal has been exposed to the agent, but a single test does not discriminate between present and past infection. Since seroconversion takes time, a disadvantage of serology in diagnostic setups is that test results may be negative during the initial stage of an infection. The results from serologic tests are retrospective and should not be interpreted in isolation. In general, it is advisable to consult and follow guidelines from the laboratory where the serum will be submitted. Techniques to collect adequate samples have been described (Hrapkiewicz and Medina 2007). Blood samples for serology should be obtained from live animals. In recently sacrificed animals, blood can be aspirated with a syringe or a Pasteur pipette from the heart, from the thorax after severing the posterior vena cava or the thoracic aorta, or from the axillary or inguinal areas after severing the arteries at these areas. Blood samples for serology should be obtained aseptically and without additives. To avoid hemolysis, the blood must be obtained and processed carefully (blood must not be forced through a small-gauge needle, and the separation of serum from the rest of the blood must not be delayed). Serum samples can be maintained for a short time at 4°C or stored frozen, preferably at –70°C. Large volumes of serum should be fractioned to avoid repeated freezing and thawing. Sampling for PCR in Diagnostic Microbiology PCR technology offers high sensitivity, specificity, and rapidity, and it is becoming essential in diagnosis and also screening for infections by important pathogens. Organisms can be detected in tissues, feces, or body fluids. An advantage of the PCR is that microbes may be detected at an early stage of infection. Unlike serology, the PCR is not based on seroconversion and is therefore applicable in immunodeficient animals that cannot produce antibodies (Compton and Riley 2001; Macy et al. 2009; Kelmenson et al. 2009). Given the high sensitivity of molecular techniques such as the PCR, contamination of specimens with nucleic acid during sampling and during specimen preparation at the diagnostic laboratory is a serious hazard. In addition to strict hygiene aimed at preventing the contamination of the necropsy room and working surfaces, it is essential to avoid cross contamination with nucleic acids between specimens. To destroy nucleic acid remnants, instruments can be sterilized by heat or dipped in a 10% solution of bleach or disinfectant for at least 5 min prior to sampling. Specimens placed in sterile tubes or containers should be frozen rapidly at –80°C (Compton and Riley 2001). Sampling for Nutrient Analysis and Toxicology Feed and bedding should be examined when nutrient excesses or deficiencies or contamination by pesticides, herbicides, heavy metals, mycotoxins or other substances that might influence biological processes is suspected. Serum and tissue specimens collected without chemical contamination should be placed in clean, leak-proof tubes or containers. Each organ must be placed in a separate container. Polyethylene bags and rubber stoppers are not always appropriate because they absorb
84550.indb 629
11/11/10 6:29:05 PM
630
Handbook of Laboratory Animal Science
or are permeable to various organic substances and the plasticizers used in their manufacture may contaminate the sample. Cross contamination between different animals and tissue specimens may be avoided by using procedures similar to those described for preventing microbial contamination. Serum and tissue specimens should be frozen for transportation to the laboratory. Whole blood should not be frozen. Blood collected in tubes containing fluoride and oxalate as preservative and anticoagulant can be transported chilled. Specimens preferred for chemical examination are liver, kidney, blood or serum, urine, and stomach with its contents. Bone should be collected when a pesticide or metal is suspected. The laboratory should be consulted about the type and amount of specimen necessary for each analysis. When possible, all toxicants to be analyzed should be defined before the necropsy, in order to determine target tissues and sample volumes, appropriate containers for samples, and any other factors necessary to obtain optimal results (Cooper and Cooper 2007; Munro 1998; Galey and Talcott 2006). Conclusion We are aware that some experiments will require more detailed necropsies than described in this chapter, but in many situations, a necropsy that is less thorough than presented here will suffice. A partial necropsy will be better than no necropsy at all. This chapter is intended to encourage investigators to spend more time in gross morphologic evaluation of their animals, as well as guide them in taking meaningful samples. Acknowledgment We thank Bengt Ekberg for the photographs presented in this chapter. The photographs shown were taken during the necropsy of a female Sprague–Dawley rat, aged approximately 3 months. References Bederson, J. B., L. H. Pitts, S. M. Germano, M. C. Nishimura, R. L. Davis, and H. M. Bartkowski. 1986. Evaluation of 2,3,5-triphenyltetrazolium chloride as a stain for detection and quantification of experimental cerebral infarction in rats. Stroke 17:1304. Boot, R., A. K. Hansen, C. K. Hansen, N. Nozari, and H. C. W. Thuis. 2000. Comparison of assays for antibodies to Encephalitozoon cuniculi in rabbits. Lab Anim 34:281. Brayton, C., M. Justice, and C. A. Montgomery. 2001. Evaluating mutant mice: Anatomic pathology. Veterinary Pathology 38:1. Bucci, T. J. 2002. Basic techniques. In Handbook of toxicologic pathology, 2nd ed., ed. W. M. Haschek, C. G. Rousseaux, and M. A. Wallig, chap. 8. San Diego, CA: Academic Press. Canadian Food Inspection Agency. Containment standards for veterinary facilities. This publication may be found at http://www.inspection.gc.ca/english/sci/lab/convet/convete.shtml. Cassell, G. H., M. K. Davidson, J. K. Davis, and J. R. Lindsey. 1983. Recovery and identification of murine mycoplasmas. In Methods in mycoplasmology, vol. 2, Diagnostic mycoplasmology, ed. J. G. Tully and S. Razin, 129. New York: Academic Press. Casella, J. P., J. Hay, and S. J. Lawson. 2007. The rat nervous system. An introduction to preparatory techniques. Chichester, England, John Wiley & Sons Ltd. Clasen, R. A., S. Pandolfi, and G. M. Has. 1990. Vital staining, serum albumin and the blood-brain barrier. Journal of Neuropathology & Experimental Neurology 29:266. Coleridge, H., J. C. G. Coleridge, and A. Howe. 1967. A search for pulmonary arterial chemoreceptors in the cat, with a comparison of the blood supply of the aortic bodies in the newborn and adult animal. Journal of Physiology 191:353.
84550.indb 630
11/11/10 6:29:05 PM
Postmortem Procedures
631
Compton, S. R., and L. K. Riley. 2001. Detection of infectious agents in laboratory rodents: Traditional and molecular techniques. Comparative Medicine 51:113. Cooper, J. E., and M. E. Cooper. 2007. Introduction to veterinary and comparative forensic medicine. Ames, IA: Blackwell Publishing. Demiryürek, D., A. Bayramoğlu, and S. Ustaçelebi. 2002. Infective agents in fixed human cadavers: A brief review and suggested guidelines. Anatomical Record 269:194. Edwards, W. D. 1988. Photography of medical specimens: Experiences from teaching cardiovascular pathology. Mayo Clinic Proceedings 63:42. Everitt, J. I., and E. A. Gross. 2006. Euthanasia and necropsy. In The laboratory rat, 2nd ed., ed. M. A. Suckow, S. H. Weisbroth, and C. L. Franklin, chap. 20. San Diego, CA: Academic Press. Feldman, D. B., and J. C. Seely. 1988. Necropsy guide: Rodents and the rabbit. Boca Raton, FL: CRC Press. Feldman, S. H., and S. Bowman. 2007. Molecular phylogeny of the pinworms of mice, rats and rabbits, and its use to develop molecular beacon assays for the detection of pinworms in mice. Lab Animal Europe 7:29. Fenwick, B. W., and S. Kruckenberg. 1987. Comparison of methods used to collect canine intestinal tissues for histological examination. American Journal of Veterinary Research 48:1276. Galey, F. D., and P. A. Talcott. 2006. Effective use of a diagnostic laboratory. In Small animal toxicology, 2nd ed., ed. M. E. Peterson and P. A. Talcott. St. Louis, MO: Elsevier Saunders. Ghoshal, N. G., and H. S. Bal. 1989. Comparative morphology of the stomach of some laboratory mammals. Lab Anim 23:21. Greve, G., and T. Saetersdal. 1991. Problems related to infarct size measurements in the rat heart. Acta Anatomica (Basel) 142:366. Harkema, J. R., S. A. Carey, and J. G. Wagner, 2006. The nose revisited: A brief review of the comparative structure, function, and toxicologic pathology of the nasal epithelium. Toxicologic Pathology 34:252. Hofler, H., and J. Mueller. 1994. In situ hybridization in pathology. Verhandlungen der Deutschen Gesellschaft für Pathologie 78:124. Homberger, F., R. Boot, R. E. Feinstein, A. Kornerup-Hansen, and J. van der Logt. 1999. FELASA guidance paper for the accreditation of laboratory animal diagnostic laboratories, Report of the Federation of European Laboratory Animal Science Associations (FELASA) Working Group on Accreditation of Diagnostic Laboratories. Lab Anim 33 (s1):19. This publication may also be found at http://www.lal.org.uk/pdffiles/LAfel4.pdf Howard, H., and Y. K. Foucher. 2009. Animal use space. In Planning and designing research animal facilities, ed. J. R. Hessler and N. D. M. Lehner, chap. 19. London: Academic Press. Hrapkiewicz, K., and L. Medina. 2007. Clinical laboratory animal medicine. An introduction, 3rd ed. Oxford, England: Blackwell Publishing. Jacoby, R. O., and J. R. Lindsey. 1997. Health care for research animals is essential and affordable. FASEB Journal 11:609. Kelmenson, J. A., D. P. Pomerleau, S. Griffey, W. Zhang, M. J. Karolak, and J. R. Fahey. 2009. Kinetics of transmission, infectivity, and genome stability of two novel mouse norovirus isolates in breeding mice. Comparative Medicine 59:27. King, J. M., L. Roth-Johnson, D. C. Dodd, and N. E. Newson. 2005. The necropsy book, 4th ed. Gurnee, IL: Charles Louis Davis, D.V.M., Foundation. Kittel, B., C. Ruehl-Fehlert, G. Morawietz, et al. 2004. Revised guides for organ sampling and trimming in rats and mice—Part 2. Experimental and Toxicologic Pathology 55:413. This publication may also be found at http://reni.item.fraunhofer.de/reni/trimming/index.php?lan=en Kornerup Hansen, A. 2000. Handbook of laboratory animal bacteriology, chap. 2. Boca Raton, FL: CRC Press. Kunstyr, I., W. Küpper, H. Weisser, S. Naumann, and C. Messow. 1982. Urethral plug. A new secondary male sex characteristic in rat and other rodents. Lab Anim 16:151. Lassen, E. D., and G. Weiser. 2004. Laboratory technology for veterinary medicine. In Veterinary hematology and clinical chemistry, ed. M. A. Thrall et al., chap. 1. Philadelphia, PA: Lippincott Williams & Wilkins. LeVeen, H. H., and W. H. Fishman. 1947. Combination of Evans blue with plasma protein: Its significance in capillary permeability studies, blood dye disappearance curves, and its use as a protein tag. American Journal of Physiology 151:26. Macy, J. D., F. X. Paturzo, L. J. Ball-Goodrich, and S. R. Compton. 2009. A PCR-based strategy for detection of mouse parvovirus. Journal of A.A.L.A.S. 48:263.
84550.indb 631
11/11/10 6:29:05 PM
632
Handbook of Laboratory Animal Science
Mann, P. C., J. F. Hardisty, and M. D. Parker. 2002. Managing pitfalls in toxicologic pathology. In Handbook of toxicologic pathology, 2nd ed., ed. W. M. Haschek, C. G. Rousseaux, and M. A. Wallig, chap. 9. San Diego, CA: Academic Press. McDonald, D. M., and R. W. Blewett. 1981. Location and size of carotid body-like organs (paraganglia) revealed in rats by the permeability of blood vessels to Evans blue dye. Journal of Neurocytology 10:607. Mery, S., E. A. Gross, D. R. Joyner, M. Godo, and K. T. Morgan. 1994. A tool for recording the distribution of nasal lesions in rats and mice. Toxicologic Pathology 22:353. Moolenbeek, C., and E. J. Ruitenberg. 1981. The “Swiss roll”: A simple technique for histological studies of the rodent intestine. Lab Anim 15:57. Morawietz, G., C. Ruehl-Fehlert, B. Kittel, et al. 2004. Revised guides for organ sampling and trimming in rats and mice—Part 3. Experimental and Toxicologic Pathology 55:433. This publication may also be found at http:// http://reni.item.fraunhofer.de/reni/trimming/index.php?lan=en. Munro, R. 1998. Forensic necropsy. Seminars in Avian and Exotic Pet Medicine 7:201. Munson, L. 1988. Necropsy of wild animals. This publication may be found at http://www.vetmed.ucdavis.edu/ whc/pdfs/necropsy.pdf. Murphy, F. A., E. P. Gibbs, M. C. Horzinek, and M. J. Studdert. 1999. Laboratory diagnosis of viral diseases. In Veterinary virology, 3rd ed., chap. 12. San Diego, CA: Academic Press. Murray, G. I., and S. W. B. Ewen. 1991. A novel method for optimum biopsy specimen preservation for histochemical and immunohistochemical analysis. American Journal of Clinical Pathology 95:131. Nicklas, W., P. Baneux, R. Boot, et al. 2002. Recommendations for the health monitoring of rodent and rabbit colonies in breeding and experimental units. Laboratory Animals 36:20. This publication may also be found at http://www.felasa.eu. Noel, R. B. 1982. Toxicity testing, hazard assessment, and data quality assurance in respect to use of laboratory animals. In Animals in toxicological research, ed. I. Bartosek, A. Guaitani, and E. Pacei, 45. New York: Raven Press. Relyea, M. J., J. Miller, D. Boggess, and J. Sundberg. 2000. Necropsy methods for laboratory mice: Biological characterization of a new mutation. In Systematic approach to evaluation of mouse mutations, ed. J. P. Sundberg and D. Boggess, chap. 5. Boca Raton, FL: CRC Press. Renne, R. A., J. Y. Everitt, J. R. Harkema, C. G. Plopper, and M. Rosenbruch. 2009. OECD guidance document on histopathology for inhalation studies. Draft 1/36 ENV 1, 2009. This paper may be found at http://www.oecd.org/dataoecd/45/2/43822718.pdf. Richardson, K. C. 1969. The fine structure of autonomic nerves after vital staining with methylene blue. Anatomical Record 164:359. Ruehl-Fehlert, C., B. Kittel, G. Morawietz, et al. 2003. Revised guides for organ sampling and trimming in rats and mice—Part 1. Experimental and Toxicologic Pathology 55:91. This publication may also be found at http://reni.item.fraunhofer.de/reni/trimming/index.php?lan=en. Saikia, B., K. Gupta, and U. M. Saikia. 2008. The modern histopathologist: In the changing face of time. Diagnostic Pathology 25:3. Scheifele, D., G. Bjornson, and J. Dimmick. 1987. Rapid postmortem gut autolysis in infant rats: A potential problem for investigators. Canadian Journal of Veterinary Research 51:404. Seaman, W. J. 1987. Postmortem change in the rat: A histologic characterization, chaps. 1, 2, 4. Ames: Iowa State University Press. Seymour, R., T. Ichiki, I. Mikaelian, et al. 2004. Necropsy methods. In The laboratory mouse, ed. H. Hedrich, chap. 30. London: Elsevier Academic Press. Shi, S.-R., M. E. Key, and K. L. Kalra. 1991. Antigen retrieval in formalin-fixed, paraffin-embedded tissues: An enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections. Journal of Histochemistry and Cytochemistry 39:741. Shimoda, K., K. Maejima, T. Kuhara, and M. Nakagawa. 1991. Stability of pathogenic bacteria from laboratory animals in various transport media. Lab Anim 25:228. Smith, P. H., E. W. Seklau, J. B. Malone, and C. M. Monahan. 2007. Collection, preservation, and diagnostic methods. In Flynn’s parasites of laboratory animals, 2nd ed., ed. D. G. Baker, chap. 1. Ames, IA: Blackwell Publishing. Songer, J. G., and K. W. Post. 2008. General Principles of bacterial disease diagnosis. In Veterinary microbiology. Bacterial and fungal agents of animal disease, chaps. 2, 44. St. Louis, MO: Elsevier Saunders.
84550.indb 632
11/11/10 6:29:05 PM
Postmortem Procedures
633
Strafuss, A. C. 1988. Necropsy. Procedures and basic diagnostic methods for practicing veterinarians. Springfield, IL: Charles C Thomas. Teo, I. A., and S. Shaunak. 1995. Polymerase chain reaction in situ: An appraisal of an emerging technique. Histochemical Journal 27:647. Ullman-Culleré, M. H., and C. J. Foltz. 1999. Body condition scoring: A rapid and accurate method for assessing health status in mice. Laboratory Animal Science 49:319. USDHHS (U.S. Department of Health and Human Services). 2007. Biosafety in microbiological and biomedical laboratories, 5th ed. Washington, D.C. This publication may also be found at http://www.unh.edu/ ehs/pdf/CDC-NIH-BMBL-5.pdf. Ward, J. M., F. Mahler, R. R. Maronpot, J. Sundberg, and R. M. Frederickson. 2000. Pathology of genetically engineered mice. Ames: Iowa State University Press. Weinberg, D. S. 1997. Digital imaging as a teaching tool for pathologists. Clinical Laboratory Medicine 17:229. Wood, P. A. 2000. Phenotype assessment: Are you missing something? Comparative Medicine 50:12. WHO Laboratory Biosafety Manual, 3rd ed. 2004. Geneva. This publication may also be at http://www.who. int/csr/resources/publications/biosafety/WHO_CDS_CSR_LYO_2004_11/en/print.html.
Useful Web Pages http://reni.item.fraunhofer.de/reni/trimming/index.php?lan=en http://www.unh.edu/ehs/pdf/CDC-NIH-BMBL-5.pdf http://www.felasa.eu/ http://www.ourtrent.com/docs/dna/biocontainment-docs/cfia-biocontainment-standards.pdf http://www3.niaid.nih.gov/labs/aboutlabs/cmb/InfectiousDiseasePathogenesisSection/mouseNecropsy/ http://WWW.oecd.org/dataoecd/45/2/43822718.pdf
84550.indb 633
11/11/10 6:29:05 PM
84550.indb 634
11/11/10 6:29:05 PM
Chapter 22
Alternatives Refinement, Reduction, and Replacement of Animal Uses in the Life Sciences
Lisbeth E. Knudsen, Marlies Leenaars, Bart S. Savenije, and Merel Ritskes-Hoitinga Contents Introduction..................................................................................................................................... 636 Use of Animals for Research and for Testing: Statistics, etc.......................................................... 637 The Three Rs..................................................................................................................................640 Regulatory Requirements.......................................................................................................... 641 Alternatives in Research................................................................................................................. 641 Continuing Professional Development....................................................................................... 642 Imaging...................................................................................................................................... 642 Systematic Reviews.................................................................................................................... 642 Information Sharing and Networking........................................................................................ 642 Humane End Point..................................................................................................................... 643 Training and Positive Reinforcement......................................................................................... 643 Statistics and Meta-analysis....................................................................................................... 643 “-omics”..................................................................................................................................... 643 In Vitro.......................................................................................................................................644 Serum-Free Media.....................................................................................................................644 Studies with Humans.................................................................................................................644 Lower Species............................................................................................................................ 645 Alternatives in Regulatory Testing................................................................................................. 645 QSARs.......................................................................................................................................646 In Silico......................................................................................................................................646 Step-by-Step Approach..............................................................................................................646 Statistics.....................................................................................................................................646 Combining Treatments...............................................................................................................646 In Vitro....................................................................................................................................... 647 International Harmonization...................................................................................................... 647 Validation................................................................................................................................... 647 Alternatives in Education................................................................................................................648 635
84550.indb 635
11/11/10 6:29:06 PM
636
Handbook of Laboratory Animal Science
Demonstration............................................................................................................................649 Dummies....................................................................................................................................649 Film/DVD/CD-ROM................................................................................................................. 650 Interactive DVD......................................................................................................................... 650 Skills Training by Simulation Programs.................................................................................... 650 Alternative Methods Centers.......................................................................................................... 650 References....................................................................................................................................... 651 Introduction Attention to the use of animals in life sciences and toxicological testing has increased with increasing public concern about the safety of products (ensuring no adverse health effects) and the humane treatment of laboratory animals used for development and testing of products. In addition to ethical considerations, the limitations in extrapolation from animal studies to human risk and the high costs of animal testing have promoted efforts to develop nonanimal alternatives to animal testing. This has taken place by introducing mathematical and computational models predicting toxicity from structural relationships with known compounds, by developing in vitro test systems, by considering data from high-throughput analyses and from in silico analyses (with “-omics” technologies), and by applying “intelligent,” integrated testing strategies. Each animal experiment will need to be justified by taking all available data into account. The inclusion of actual exposures in prioritization of testing is taken into account—for example, in the European REACH (registration, evaluation, and authorization of chemicals) testing program (Lilienblum et al. 2008). Data sharing will also promote decreased animal use and will avoid unnecessary repetitions of studies. The regulatory authorities play a major role in acceptance of information derived from other sources apart from animal studies by acknowledging comparable information from such other sources (e.g., in vitro, computer models). Most university studies within life and health sciences include risk assessment modules where toxicological information from animal studies is included. In physiology and biochemistry, animal experiments traditionally took place using vertebrates as well as invertebrates. Currently, considerable effort is devoted to making use of demonstration videos and similar aids to avoid the unnecessary use of animals. Challenges to the successful establishment of alternatives to the use of animals in life science research include the development of new analytical tools (e.g., -omics technologies), introduction of genetically modified organisms (GMOs), techniques for optimally handling products of human origin, and the best way to utilize samples and information from studies involving patients and healthy volunteers. The most pragmatic approach to reduce experiments on animals is to introduce alternative methods that eventually replace animal testing (replacement alternatives). If replacement is not possible, then every effort should be made to apply methods that use the smallest number of animals (reduction alternatives) and/or cause the least harm to the animals (refinement alternatives). Newly developed alternative methods must be validated to assess their relevance and reliability. Once validated, they can be made available for regulatory purposes. It is a challenge to determine the strategies already in operation within the field of laboratory science that might have an impact on the three Rs. It is also a challenge to determine how to evaluate strategies once they are implemented. Tracking relevant three Rs for a specific scientific question to minimize the use and pain and distress of animals is a big task in itself: About 100 databases with three Rs information are already available worldwide, each with its specific content and search strategies (Leenaars et al. 2009).
84550.indb 636
11/11/10 6:29:06 PM
Alternatives
637
Furthermore, refinement alternatives must be implemented in such a way that the welfare of the animals and the experimental hypotheses, methods, and results are promoted simultaneously. Choosing the proper cage enrichment for a given situation should improve welfare and experimental results; however, if cage enrichment is not suitable for a certain species or individual, then it may increase stress and compromise experimental results. Successful implementation of refinement procedures requires thorough preparation, evaluation of the literature, and training and experience within the field of laboratory animal science. Before experiments begin, a thorough literature search needs to be performed in order to design experiments that incorporate all available knowledge, to avoid unnecessary duplication of animal studies, and to provide optimal patient protection. A thorough literature search is a systematic review. Until now, systematic reviews were not commonly conducted when animal studies were performed. This is different from the medical field, where the Cochrane database was established decades ago. It is time for systematic reviews to be implemented routinely when animal studies are performed. Moreover, improved study designs for animal research are urgently needed (Kilkenny et al. 2009). This is an essential part of the responsible use of animals that will improve the translational value of animal studies to humans, as discussed in Chapter 13 in this volume. Appropriate study designs and careful reporting of animal research will also facilitate metaanalyses of these works (Hooijmans, Leenaars, and Ritskes-Hoitinga in press). This is currently a difficult task because much of the published data are not sufficiently detailed. Examples of three Rs that can be implemented in research, regulatory testing, and education are presented in Table€22.1. Use of Animals for Research and for Testing: Statistics, etc. The total number of animals used for experiments in the EU in 2005 was 12.1 million (with 1 of 25 member states reporting from 2004). As in previous years, by far the biggest group of animals used was rodents and rabbits, representing more than three-fourths of all animals used (78%). Table€22.1╅Three Rs Topics within Regulatory Testing, Research, and Education
84550.indb 637
Regulatory
Research
Refinement
Humane end points (HEP) Step-by-step approach
Reduction
International harmonization Step-by-step approach Statistics, meta-analysis Combining treatments (e.g., combination serology) -omics
Replacement
In vitro Risk assessment approach Qsars “Lower” species
Continuing professional development (CPD) Imaging (early detection of pathologies, related to HEP) Systematic reviews Information sharing, networking HEP Training/positive reinforcement “Lower” species Systematic reviews Statistics/experimental design Combining research -omics “Lower” species (e.g., Drosophila instead of vertebrates) In vitro Qsars In silico Systematic reviews “Lower” species (e.g., Drosophila instead of vertebrates)
Education Demonstration Dummies Skills training
Skills training
Film/DVD/CDROM interactive DVD Dummies
11/11/10 6:29:06 PM
638
Handbook of Laboratory Animal Science
Carnivores 0.33% Artio + perissodactyla 1.1%
Prosimians + monkeys + apes 0.09% Other mammals 0.08%
Birds 5.4% Cold-blooded animals 15% Rabbits 2.6% Other rodents 0.8%
Mice 53%
Guinea pigs 2.1%
Rats 19%
Figure 22.1â•…A color version of this figure follows page 336. Percentages of animals used in research. (From http://ec.europa.eu/environment/chemicals/lab_animals/pdf/staff_work_doc_sec1455.pdf.)
The group of cold-blooded animals is the second largest group, accounting for over 15% of all animals, followed by birds with 5%. As in 2002, no great apes were used in experiments in the EU in 2005. Figures from the fifth report can be found at http://ec.europa.eu/environment/chemicals/ lab_animals/pdf/staff_work_doc_sec1455.pdf. Figures€22.1–22.3 present the percentages of animals used in research, the percentages of animals used for selected purposes, and the percentages of animals used for toxicological or other safety evaluations broken down into types of products for which testing was required. The Cosmetics Directive in Europe foresees a regulatory framework with the aim of phasing out animal testing. It establishes a ban on testing finished cosmetic products and cosmetic ingredients on animals (testing ban) and a ban on marketing finished cosmetic products and ingredients included in cosmetic products tested on animals in the European Community (marketing ban). The testing ban on finished cosmetic products has been in effect since September 11, 2004; the testing ban on ingredients or combinations of ingredients will become effective in a step-by-step manner as soon as alternative methods are validated and adopted. There is a maximum cutoff date of 6 years after the effective date of the directive (i.e., March 11, 2009), regardless of the availability of alternative nonanimal tests. The marketing ban will also be applied in a step-by-step fashion as soon as alternative methods are validated and adopted in EU legislation with due regard to the OECD (Organization for Economic Cooperation and Development) validation process. The marketing ban also will be introduced within 6 years of the effective date of the directive (i.e., March 11, 2009) for all human health effects, with the exception of repeated-dose toxicity, reproductive toxicity, and toxicokinetics. For these specific health effects, a deadline of 10 years after the effective date of the directive is foreseen (i.e., March 11, 2013), regardless of the availability of alternative nonanimal tests.
84550.indb 638
11/11/10 6:29:07 PM
Alternatives
639
Other 8%
Education and training 1.6% Diagnosis of disease 2%
Fundamental biology studies 33%
Toxicological and other safety evaluation 8% Production and quality control veterinary medicine 3.5% Production and quality control human medicine and dentistry 11.8%
Research and develop human + veterin + dentist 31%
Figure 22.2â•…A color version of this figure follows page 336. Percentages of animals used for selected purposes. (From http://ec.europa.eu/environment/chemicals/lab_animals/pdf/staff_work_doc_ sec1455.pdf.)
3,000,000 2,500,000 2,000,000 Other
1,500,000
Education and training Diagnosis of disease
1,000,000
Cold-blooded animals
Birds
Other mammals
Prosimians + monkeys + apes
Artio + perissodactyla
Carnivores
Rabbits (Oryctolagus cuniculus)
Other rodents
Rats (Rattus norvegicus)
0
Mice (Mus musculus)
500,000
Toxicological and other safety evaluations (including safety evaluation of products) Research, development and quality control of products and devices for human medicine and dentistry and for veterinary medicine Biological studies of a fundamental nature
Figure 22.3â•…A color version of this figure follows page 336. Percentages of animals used for toxicological or other safety evaluations broken down into types of products for which testing was required. (From http://ec.europa.eu/environment/chemicals/lab_animals/pdf/staff_work_doc_sec1455.pdf.)
84550.indb 639
11/11/10 6:29:09 PM
640
Handbook of Laboratory Animal Science
The Three Rs The concept of the three Rs in animal-based research is an ethical framework that was first described in 1959 by Professor William Russell and Dr. Rex Burch in their book, The Principles of Humane Experimental Techniques (1959): Replacement means the substitution for conscious living higher animals of insentient material. Reduction means reduction in the numbers of animals used to obtain information of a given amount and precision. Refinement means any decrease in the incidence or severity of inhumane procedures applied to those animals which still have to be used. There are clearly areas of overlap between these categories. Consider the use of animal tissue cultures in virology. In a fundamental sense, we are here replacing animals by insentient material, and the method has been classified as replacement for the present purpose. But since one animal may be used to provide many cultures, each providing more information than a single whole animal used directly, we might legitimately speak of reduction. Finally, the animal used as the source for the cultures may be painlessly killed, instead of being exposed to the risk of a virus disease; so we might also label the procedure as a refinement.
As animal research has advanced, so has the interpretation of the three Rs. The three Rs help to ensure humane and responsible science, as well as good science, since if animals suffer, then that suffering may well affect the scientific outcomes being measured. This obviously does not apply when the suffering itself is the object of the research (e.g., studies of pain and/or distress). The U.S. Animal Welfare Act (AWA) regulations (specifically, the 1985 amendment) require the principal investigators to consider alternatives to procedures that may cause more than momentary or slight pain or distress to the animals, and to provide a written narrative of the methods used and sources consulted to determine the availability of alternatives, including refinements, reductions, and replacements. The search for alternatives refers to the three Rs: • Refinement. The use of analgesics and analgesia, the use of remote telemetry to increase the quality and quantity of data gathered, and humane end points for the animals are examples of refinements. • Reduction. The use of shared control groups, preliminary screening in nonanimal systems, innovative statistical packages, or consultations with a statistician are examples of reduction alternatives. • Replacement. Alternatives such as in vitro, cell culture, tissue culture, models, simulations, etc. are examples of replacement. This is also where one might look for any nonmammalian animal models—fish or invertebrates, for example—that would still provide the data needed.
The Council Directive of November 24, 1986—on the approximation of laws, regulations, and administrative provisions of the member states regarding the protection of animals used for experimental and other scientific purposes (86/609/EEC)—specifies legal, ethical, and scientific obligations to consider and incorporate the three Rs in any European research program. • Replacement: “An experiment shall not be performed if another satisfactory method of obtaining the result sought, not entailing the use of an animal, is reasonably and practicably available.” Some examples of replacements are given and could include: • Improved use of existing information by literature searches and publication of in-house data • Predictions based on the physical and chemical properties of molecules (e.g., pH, osmolality, comparison with other similar molecules) • In silico methods: mathematical and computer models, quantitative structure activity relationships (QSARs), molecular modeling • Organisms of a lower neurophysiological sensitivity (e.g., some invertebrates, plants) • Early developmental stages of vertebrates for teratological research • In vitro methods: organ cultures, tissue slices, cell cultures. • Human studies: volunteers, epidemiology; some companies use human volunteers to test finished products to confirm safety
84550.indb 640
11/11/10 6:29:10 PM
Alternatives
641
• Reduction: to use the minimum number of animals to meet the scientific objective within given confidence limits: • It is strongly recommended that professional statistical advice be sought before submitting the grant application and be detailed in the application to show that due consideration has been given to reduction. There should also be some form of ongoing review during the research program. • Reuse can reduce the number of animals, but care should be taken to avoid misleading results and increasing animal suffering. Reuse should not be confused with “continued use,” which implies that the chosen animal must be used because of some prior treatment, whereas with reuse, any animal can be chosen. • Good literature searches and avoiding unnecessary duplication of experimental groups, especially controls, are important. • Mutually recognized experimental standards promote data acceptability. • Results should be published or disseminated in another way (e.g., on a Web site) to prevent repetition of work, including studies that “failed.” • Refinement: methods that aim to avoid, alleviate, or minimize the potential pain, distress, or other adverse effects suffered by animals and enhance animal well-being. All invasive animal research should be conducted under general or local anesthetic unless that would be incompatible with the scientific objective or the use of an anesthetic would cause more suffering than the proposed research procedure: • Recognition of animal suffering is a critical first step in refinement. Three ways to reduce suffering are to −â‹™ Avoid it in the first place, through pilot studies and thorough literature reviews −â‹™ Ensure alleviation whenever compatible with the scientific objective −â‹™ Ensure that all those involved in the research are trained, are competent, and can distinguish between good and poor practices • The welfare of animals may be reduced through poor health. The use of specific pathogen free animals, or use of healthy animals is recommended. • The housing of animals in barren cages of an inappropriate size will induce abnormal behaviors and poor welfare. • Inappropriate husbandry such as social isolation and inappropriate training or handling will cause poor welfare. • Poor c