Microbial Safety of Fresh Produce (Institute of Food Technologists Series)

Microbial Safety of Fresh Produce

The IFT Press series reflects the mission of the Institute of Food Technologists—to advance the science of food contributing to healthier people everywhere. Developed in partnership with Wiley-Blackwell, IFT Press books serve as leading-edge handbooks for industrial application and reference and as essential texts for academic programs. Crafted through rigorous peer review and meticulous research, IFT Press publications represent the latest, most significant resources available to food scientists and related agriculture professionals worldwide. Founded in 1939, the Institute of Food Technologists is a nonprofit scientific society with 22,000 individual members working in food science, food technology, and related professions in industry, academia, and government. IFT serves as a conduit for multidisciplinary science through leadership, championing the use of sound science across the food value chain through knowledge sharing, education, and advocacy. IFT Book Communications Committee Barry G. Swanson Syed S. H. Rizvi Joseph H. Hotchkiss Christopher J. Doona William C. Haines Ruth M. Patrick Mark Barrett John Lillard Karen Nachay IFT Press Editorial Advisory Board Malcolm C. Bourne Fergus M. Clydesdale Dietrich Knorr Theodore P. Labuza Thomas J. Montville S. Suzanne Nielsen Martin R. Okos Michael W. Pariza Barbara J. Petersen David S. Reid Sam Saguy Herbert Stone Kenneth R. Swartzel

A John Wiley & Sons, Inc., Publication

Microbial Safety of Fresh Produce Xuetong Fan, Brendan A. Niemira, Christopher J. Doona, Florence E. Feeherry, Robert B. Gravani EDITORS

A John Wiley & Sons, Inc., Publication

Edition first published 2009 © 2009 Blackwell Publishing and the Institute of Food Technologists Chapters 1, 5, 10, 12, 13, 14, 17, 18, and 22 remain with the U.S. Government. Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Editorial Office 2121 State Avenue, Ames, Iowa 50014-8300, USA For details of our global editorial offices, for customer services, and for information about how to apply for permission to reuse the copyright material in this book, please see our website at www.wiley.com/wiley-blackwell. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Blackwell Publishing, provided that the base fee is paid directly to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license by CCC, a separate system of payments has been arranged. The fee codes for users of the Transactional Reporting Service are ISBN-13: 978-0-8138-0416-3/2009. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks, or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Library of Congress Cataloging-in-Publication Data Microbial safety of fresh produce / editors, Xuetong Fan . . . [et al.]. p. cm. – (The IFT press series) Includes bibliographical references and index. ISBN 978-0-8138-0416-3 (hardback : alk. paper) 1. Fruit–Microbiology. 2. Vegetables–Microbiology. I. Fan, Xuetong. QR122.M537 2009 363.19'29–dc22 2009009719 A catalog record for this book is available from the U.S. Library of Congress. Set in 10 on 12 pt Times by SNP Best-set Typesetter Ltd., Hong Kong Printed in Singapore 1

2009

Titles in the IFT Press series • Accelerating New Food Product Design and Development (Jacqueline H. Beckley, Elizabeth J. Topp, M. Michele Foley, J.C. Huang, and Witoon Prinyawiwatkul) • Advances in Dairy Ingredients (Geoffrey W. Smithers and Mary Ann Augustin) • Biofilms in the Food Environment (Hans P. Blaschek, Hua H. Wang, and Meredith E. Agle) • Calorimetry and Food Process Design (Gönül Kaletunç) • Nondigestible Carbohydrates and Digestive Health (Teresa M. Paeschke and William R. Aimutis) • Food Ingredients for the Global Market (Yao-Wen Huang and Claire L. Kruger) • Food Irradiation Research and Technology (Christopher H. Sommers and Xuetong Fan) • Food Laws, Regulations and Labeling (Joseph D. Eifert) • Food Risk and Crisis Communication (Anthony O. Flood and Christine M. Bruhn) • Foodborne Pathogens in the Food Processing Environment: Sources, Detection and Control (Sadhana Ravishankar and Vijay K. Juneja) • Functional Proteins and Peptides (Yoshinori Mine, Richard K. Owusu-Apenten, and Bo Jiang) • High Pressure Processing of Foods (Christopher J. Doona and Florence E. Feeherry) • Hydrocolloids in Food Processing (Thomas R. Laaman) • Microbial Safety of Fresh Produce (Xuetong Fan, Brendan A. Niemira, Christopher J. Doona, Florence E. Feeherry, and Robert B. Gravani) • Microbiology and Technology of Fermented Foods (Robert W. Hutkins) • Multiphysics Simulation of Emerging Food Processing Technologies (Kai Knoerzer, Pablo Juliano, Peter Roupas, and Cornelis Versteeg) • Multivariate and Probabilistic Analyses of Sensory Science Problems (Jean-François Meullenet, Rui Xiong, and Christopher J. Findlay • Nondestructive Testing of Food Quality (Joseph Irudayaraj and Christoph Reh) • Nanoscience and Nanotechnology in Food Systems (Hongda Chen) • Nonthermal Processing Technologies for Food (Howard Q. Zhang, Gustavo V. BarbosaCànovas, V.M. Balasubramaniam, Editors; C. Patrick Dunne, Daniel F. Farkas, and James T.C. Yuan, Associate Editors) • Nutraceuticals, Glycemic Health and Type 2 Diabetes (Vijai K. Pasupuleti and James W. Anderson) • Packaging for Nonthermal Processing of Food (J. H. Han) • Preharvest and Postharvest Food Safety: Contemporary Issues and Future Directions (Ross C. Beier, Suresh D. Pillai, and Timothy D. Phillips, Editors; Richard L. Ziprin, Associate Editor) • Processing and Nutrition of Fats and Oils (Ernesto M. Hernandez and Afaf Kamal-Eldin) • Processing Organic Foods for the Global Market (Gwendolyn V. Wyard, Anne Plotto, Jessica Walden, and Kathryn Schuett) • Regulation of Functional Foods and Nutraceuticals: A Global Perspective (Clare M. Hasler) • Sensory and Consumer Research in Food Product Design and Development (Howard R. Moskowitz, Jacqueline H. Beckley, and Anna V.A. Resurreccion) • Sustainability in the Food Industry (Cheryl J. Baldwin) • Water Activity in Foods: Fundamentals and Applications (Gustavo V. Barbosa-Cànovas, Anthony J. Fontana Jr., Shelly J. Schmidt, and Theodore P. Labuza) • Whey Processing, Functionality and Health Benefits (Charles I. Onwulata and Peter J. Huth)

Contents

Contributors

xi

Preface

xv

Section I.

Microbial Contamination of Fresh Produce

3

Chapter 1. Enteric Human Pathogens Associated with Fresh Produce: Sources, Transport, and Ecology Robert E. Mandrell

5

Chapter 2. The Origin and Spread of Human Pathogens in Fruit Production Systems Susan Bach and Pascal Delaquis

43

Chapter 3.

Internalization of Pathogens in Produce Elliot T. Ryser, Jianjun Hao, and Zhinong Yan

55

Section II.

Preharvest Strategies

81

Chapter 4. Produce Safety in Organic vs. Conventional Crops Francisco Diez-Gonzalez and Avik Mukherjee

83

Chapter 5.

The Role of Good Agricultural Practices in Produce Safety Robert B. Gravani

101

Chapter 6.

Effectively Managing through a Crisis Will Daniels and Michael P. Doyle

119

Chapter 7.

The Role of Water and Water Testing in Produce Safety Charles P. Gerba

129

Chapter 8.

The Role of Manure and Compost in Produce Safety Xiuping Jiang and Marion Shepherd

143

Section III. Chapter 9.

Postharvest Interventions Aqueous Antimicrobial Treatments to Improve Fresh and Fresh-Cut Produce Safety Joy Herdt and Hao Feng

167 169

vii

viii

Contents

Chapter 10. Irradiation Enhances Quality and Microbial Safety of Fresh and Fresh-Cut Fruits and Vegetables Brendan A. Niemira and Xuetong Fan Chapter 11. Biological Control of Human Pathogens on Produce John Andrew Hudson, Craig Billington, and Lynn McIntyre Chapter 12. Extension of Shelf Life and Control of Human Pathogens in Produce by Antimicrobial Edible Films and Coatings Tara H. McHugh, Roberto J. Avena-Bustillos, and Wen-Xian Du Chapter 13. Improving Microbial Safety of Fresh Produce Using Thermal Treatment Xuetong Fan, Bassam A. Annous, and Lihan Huang Chapter 14. Enhanced Safety and Extended Shelf Life of Fresh Produce for the Military Peter Setlow, Christopher J. Doona, Florence E. Feeherry, Kenneth Kustin, Deborah Sisson, and Shubham Chandra Section IV.

Produce Safety during Processing and Handling

Chapter 15. Consumer and Food-Service Handling of Fresh Produce Christine M. Bruhn Chapter 16. Plant Sanitation and Good Manufacturing Practices for Optimum Food Safety in Fresh-Cut Produce Edith H. Garrett Chapter 17. Third-Party Audit Programs for the Fresh-Produce Industry Kenneth S. Petersen

191 205

225

241

263

289 291

307 321

Chapter 18. Applications of Immunomagnetic Beads and Biosensors for Pathogen Detection in Produce Shu-I Tu, Joseph Uknalis, Andrew Gehring, and Peter Irwin

331

Section V.

349

Public, Legal, and Economic Perspectives

Chapter 19. Public Response to the 2006 Recall of Contaminated Spinach William K. Hallman, Cara L. Cuite, Jocilyn E. Dellava, Mary L. Nucci, and Sarah C. Condry

351

Chapter 20. Produce in Public: Spinach, Safety, and Public Policy Douglas A. Powell, Casey J. Jacob, and Benjamin Chapman

369

Chapter 21. Contaminated Fresh Produce and Product Liability: A Law-in-Action Perspective Denis W. Stearns

385

Contents

ix

Chapter 22. The Economics of Food Safety: The 2006 Foodborne Illness Outbreak Linked to Spinach Linda Calvin, Helen H. Jensen, and Jing Liang

399

Section VI.

419

Research Challenges and Directions

Chapter 23. Research Needs and Future Directions Brendan A. Niemira, Christopher J. Doona, Florence E. Feeherry, Xuetong Fan, and Robert B. Gravani

421

Index

427

Contributors

Bassam A. Annous, Chapter 13 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 Roberto J. Avena-Bustillos, Chapter 12 USDA, Agricultural Research Service Western Regional Research Center 800 Buchanan St. Albany, CA 94710 Susan Bach, Chapter 2 Agriculture and Agri-Food Canada Pacific Agri-Food Research Centre 4200 Highway 97 South Summerland, British Columbia, V0H 1Z0 Canada Craig Billington, Chapter 11 Food Safety Programme, ESR Ltd. P. O. Box 29-181 Christchurch, New Zealand Christine M. Bruhn, Chapter 15 Center for Consumer Research Department of Food Science and Technology, University of California, Davis One Shields Ave. Davis, CA 95616-8598 Linda Calvin, Chapter 22 USDA, Economic Research Service Specialty Crops Branch, Rm 5040s 1800 M Street NW Washington, D.C. 20036-5831

Shubham Chandra, Chapter 14 Chandra Associates, Milford, MA 01757 Present mailing address: U.S. Army Natick Soldier Research, Development, and Engineering Center Department of Defense Combat Feeding Directorate Systems and Equipment Engineering Team Natick, MA 01760-5018 Benjamin Chapman, Chapter 20 Department of Plant Agriculture University of Guelph Guelph, ON, N1G 2W1 Canada Sarah C. Condry, Chapter 19 Food Policy Institute Rutgers University ASB III, 3 Rutgers Plaza New Brunswick, NJ 08901 Cara L. Cuite, Chapter 19 Food Policy Institute Rutgers University ASB III, 3 Rutgers Plaza New Brunswick, NJ 08901 Will Daniels, Chapter 6 Earthbound Farm 1721 San Juan Highway San Juan Bautista, CA 95045 xi

xii

Contributors

Pascal Delaquis, Chapter 2 Agriculture and Agri-Food Canada Pacific Agri-Food Research Centre 4200 Highway 97 South Summerland, British Columbia, V0H 1Z0, Canada Jocilyn E. Dellava, Chapter 19 Department of Psychiatry, School of Medicine University of North Carolina at Chapel Hill 101 Manning Drive CB #7160 Chapel Hill, NC 27599 Francisco Diez-Gonzalez, Chapter 4 Department of Food Science and Nutrition University of Minnesota 1334 Eckles Avenue St. Paul, MN 55108 Christopher J. Doona, Chapters 14, 23 U.S. Army Natick Soldier Research, Development, and Engineering Center Department of Defense Combat Feeding Directorate Food Safety and Defense Team Natick, MA 01760-5018 Michael P. Doyle, Chapter 6 Center for Food Safety Department of Food Science & Technology University of Georgia 1109 Experiment St. Griffin, GA 30223-1791 Wen-Xian Du, Chapter 12 USDA, Agricultural Research Service Western Regional Research Center 800 Buchanan St. Albany, CA 94710

Xuetong Fan, Chapters 10, 13, 23 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 Florence E. Feeherry, Chapters 14, 23 U.S. Army Natick Soldier Research, Development, and Engineering Center Department of Defense Combat Feeding Directorate Food Safety and Defense Team Natick, MA 01760-5018 Hao Feng, Chapter 9 Department of Food Science and Human Nutrition University of Illinois at Urbana-Champaign 1304 W. Pennsylvania Avenue Urbana, IL 61801 Edith H. Garrett, Chapter 16 Edith Garrett & Associates, Inc. P. O. Box 1470 Arden, NC 28704 Andrew Gehring, Chapter 18 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 Charles P. Gerba, Chapter 7 Department of Soil, Water and Environmental Science University of Arizona Tucson, AZ 85721 Robert B. Gravani, Chapters 5, 23 Department of Food Science Cornell University Ithaca, NY 14853

Contributors

William K. Hallman, Chapter 19 Food Policy Institute Rutgers University ASB III, 3 Rutgers Plaza New Brunswick, NJ 08901 Jianjun Hao, Chapter 3 Department of Plant Pathology Michigan State University East Lansing, MI 48824 Joy Herdt, Chapter 9 Ecolab Inc. 655 Lone Oak Drive Eagan, MN 55121 Lihan Huang, Chapter 13 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 John Andrew Hudson, Chapter 11 Food Safety Programme, ESR Ltd. P. O. Box, 29-181 Christchurch, New Zealand Peter Irwin, Chapter 18 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 Casey J. Jacob, Chapter 20 Department of Diagnostic Medicine/ Pathobiology Kansas State University Manhattan, KS 66506 Helen H. Jensen, Chapter 22 Dept. of Economics Iowa State University Center for Agricultural & Rural Development 578 Heady Hall Ames, IA 50011-1070

xiii

Xiuping Jiang, Chapter 8 Department of Food Science & Human Nutrition Clemson University 217 Poole Ag. Center Clemson, SC 29634 Kenneth Kustin, Chapter 14 Department of Chemistry Emeritus Brandeis University Waltham, MA 02254-9110 Present mailing address: 5210 Fiore Ter L-111 San Diego, CA, 92122-5686 Jing Liang, Chapter 22 Department of Economics Iowa State University Center for Agricultural & Rural Development 578 Heady Hall Ames, IA 50011-1070 Robert E. Mandrell, Chapter 1 USDA, Agricultural Research Service Western Regional Research Center 800 Buchanan St. Albany, CA 94710 Tara H. McHugh, Chapter 12 USDA, Agricultural Research Service Western Regional Research Center 800 Buchanan St. Albany, CA 94710 Lynn McIntyre, Chapter 11 Food Safety Programme, ESR Ltd. P. O. Box, 29-181 Christchurch, New Zealand Avik Mukherjee, Chapter 4 Department of Food Technology Haldia Institute of Technology HIT/ICARE Campus, Hatiberia Haldia, East Midnapur West Bengal 721657 India

xiv

Contributors

Brendan A. Niemira, Chapters 10, 23 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 Mary L. Nucci, Chapter 19 Food Policy Institute Rutgers University ASB III, 3 Rutgers Plaza New Brunswick, NJ 08901 Kenneth S. Petersen, Chapter 17 USDA, Agricultural Marketing Service Fruit and Vegetable Programs 1400 Independence Avenue SW, Room 1661, Stop 0240 Washington, D.C. 20250-0240 Douglas A. Powell, Chapter 20 International Food Safety Network Department of Diagnostic Medicine/ Pathobiology Kansas State University Manhattan, KS, 66506 Elliot T. Ryser, Chapter 3 Department of Food Science and Human Nutrition National Food Safety and Toxicology Center Michigan State University East Lansing, MI 48824 Peter Setlow, Chapter 14 Molecular, Microbial and Structural Biology University of Connecticut Health Center Farmington, CT 06030-3305 Marion Shepherd, Chapter 8 Department of Food Science & Human Nutrition Clemson University C233 Poole Ag. Center Clemson, SC 29634

Deborah Sisson, Chapter 14 U.S. Army Natick Soldier Research, Development, and Engineering Center Department of Defense Combat Feeding Directorate Systems and Equipment Engineering Team Natick, MA 01760-5018 Denis W. Stearns, Chapter 21 Marler Clark LLP PS 6600 Bank of America Tower 701 Fifth Avenue Seattle, WA, 98104 Shu-I Tu, Chapter 18 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 Joseph Uknalis, Chapter 18 USDA, Agricultural Research Service Eastern Regional Research Center 600 E. Mermaid Lane Wyndmoor, PA 19038 Zhinong Yan, Chapter 3 Department of Food Science and Human Nutrition National Food Safety and Toxicology Center Michigan State University East Lansing, MI 48824

Preface

Fresh and fresh-cut fruits and vegetables contain rich sources of many nutrients and provide numerous health benefits, so nutritionists and health professionals highly recommend increasing consumption of these important foods. However, fresh produce has also been the source of recent outbreaks of foodborne illnesses, which have caused sickness, hospitalizations, and deaths of consumers, as well as serious adverse economic impact on growers and processors. The intense media attention devoted to outbreaks of Escherichia coli O157:H7 in spinach and lettuce and Salmonella linked to hot peppers and tomatoes demonstrates the importance of food safety in the mind of the public. Whether diners at the table, growers on the farm, or military personnel on shipboard or in distant global deployments, the consumer trusts the innovations of science to lead the advances dedicated to ensuring the safety of fresh produce. The increasing outbreaks of foodborne diseases associated with consumption of fresh and fresh-cut produce underscore the urgent need for significantly improving the understanding of the ecology of pathogens and for developing improved farm-to-table strategies to ensure the safety of these foods. Fresh fruits and vegetables risk contamination because they are generally grown in open fields with potential exposure to enteric pathogens from soil, irrigation water, manure, wildlife, workers, and/or other sources. Additionally, fresh produce is often eaten raw, without cooking or other treatments that could eliminate pathogens that may be present. Although the reasons for the recent increase in fresh produce– associated foodborne illnesses are not fully understood, many reasons have been suggested, including an increase in global trade, a longer and more complex food supply chain, centralized processing plants with wide distribution networks, genetic changes increasing the pathogenicity of microorganisms, and even an aging population that is more susceptible to foodborne illness. This book evolved from several Symposia we organized at the 2007 Institute of Food Technologists (IFT) Annual Meeting in Chicago. It provides information on all aspects of produce safety, including pathogen ecology, good agricultural and manufacturing practices, preharvest and postharvest interventions, third-party audits, economic concerns, consumer perceptions, education, legal concerns, and policy issues. The contributors of this book are nationally and internationally renowned experts in the field of produce safety who have shared their perspectives on a variety of important issues. Given the importance of produce safety to the country and the world, research in this area is extremely active, involving scientists from industry, government, and academia. New and novel intervention technologies and strategies xv

xvi

Preface

will be developed to further minimize the risk of pathogen contamination and increase consumer confidence in fresh produce. This book reviews the challenges of the recent, high-profile outbreaks associated with fresh produce, including the possible internalization of pathogens by plant tissues, and explains how human pathogens survive and multiply in water, in soils, and on fresh fruits and vegetables. Understanding the ecology of human pathogens will help scientists develop effective intervention strategies to enhance produce safety. Several authors discuss the latest proactive measures and strategies that the industry is taking to improve the safety of produce. With the latest advances in scientific research, all sectors of the produce industry may be able to develop and adopt integrative practices that are specific, measurable, and verifiable to assure the farm-to-table safety of produce. To minimize the risk of human pathogens on fresh produce, emphasis on preharvest strategies such as the implementation of good agricultural practices (GAPs) and a risk analysis of irrigation waters and supply lines are discussed. Postharvest interventions—from current industry practices using chemical sanitizers, to promising innovative technologies such as irradiation and biological controls—are also presented. In addition, authors address the impact of foodborne outbreaks on public health and the fresh produce industry in terms of economic impact, consumer acceptance, and legal considerations. This book offers readers comprehensive reviews of the many challenges associated with produce safety and provides strategies to minimize the risk associated with consumption of fresh produce. This text is a useful reference for all who are interested in produce safety, from consumers to university professors, students, government scientists, produce industry personnel, agricultural advisors and policy makers, consultants, equipment suppliers, third-party auditors, food retailers, and workers in the food-service industry, and it is replete with references to original source materials. Our deepest appreciation and thanks go to the chapter authors who contributed significant amounts of their time, talents, knowledge, and expertise to bring this book to life from concept to hardcover in the pursuit of increased produce safety. We are indebted to them for the excellence of their contributions, without which this book would never have been published. We are also grateful to the thoughtful guidance, helpful advice, and dedicated assistance that Ms. Susan Engelken and Mr. Mark Barrett at Wiley-Blackwell have given in the preparation and completion of this book. We dedicate this book to all those individuals who work diligently devoting their talents and efforts to reducing the microbial risks and hazards in the production, harvesting, packing, transporting, and merchandising of fresh fruits and vegetables and who value the importance of keeping these important foods safe for the benefit and well-being of all consumers. Xuetong Fan Brendan A. Niemira Christopher J. Doona Florence E. Feeherry Robert B. Gravani

Microbial Safety of Fresh Produce

Section I Microbial Contamination of Fresh Produce

1 Enteric Human Pathogens Associated with Fresh Produce: Sources, Transport, and Ecology Robert E. Mandrell

Introduction Now in the cold parts of the country, don’t you think people get to wanting perishable things in the winter—like peas and lettuce and cauliflower? In a big part of the country they don’t have those things for months and months. And right here in Salinas valley we can raise them all the year round. … Do you know we could ship lettuce right to the east coast in the middle of winter?” John Steinbeck In 1952, John Steinbeck through his character Adam Trask in “East of Eden” commented on the desirability of fresh produce and the uniqueness of the climate and soil conditions of the Salinas Valley of California for providing leafy greens and other vegetables year-round to the rest of the nation. The development of this region on the central coast of California, known as the “Salad Bowl of America,” is linked closely to the growth of fresh produce consumption in the U.S. as a result of increased seasonal availability, new varieties of domestic and imported produce, and increased interest in the nutritional and health benefits of fresh produce (Clemens 2004). The growing global economy has continued demand for fresh produce and involves shipping produce long distances rapidly. Increased mechanization and efficiency of production, new and improved cultivars, and new chemicals to treat plant disease and new products have been developed to meet this demand. Minimally processed, bagged produce is a relatively recent new product to help meet the growing demand for fresh produce (USDA-ERS 2001). An unintended consequence of increased consumption of fresh and bagged produce, however, is an increase in illnesses and outbreaks, including some multistate and multicountry outbreaks. Some of the higher profile outbreaks have been caused by E. coli O157:H7–contaminated leafy vegetables, in addition to outbreaks caused by Salmonella-contaminated tomatoes, cantaloupe, and other produce items. Investigations of some of these outbreaks have led some to conclude that contamination occurred probably in the field, i.e., preharvest contamination (CalFERT 2007a,b, 2008; Hedberg and others 1999; Gupta and others 2007; Greene and others 2008; Castillo and others 2004). The leafy green outbreaks appear not to be associated simply with an increase in consumption. Leafy green consumption between 1996 and 2005 increased 9% compared to the previous decade, but outbreaks associated with leafy greens increased 38.6%, with a majority of them caused by E. coli O157:H7 (Herman and others 2008). 5

6

Section I. Microbial Contamination of Fresh Produce

Outbreaks associated with these commodities occurring since 2000 have led to proposals and active studies to identify the risk factors that may enhance preharvest contamination of produce. However, no single risk factor can explain these multiple outbreaks associated with different production environments, processes, produce items, and pathogens. Rather, it is probable that a convergence of multiple dynamic events involving more than one factor are required to cause major, noticeable outbreaks. Each outbreak may be caused by one or more events different from other outbreaks, even though some common factors are suspected, such as the probable source (e.g., livestock, wild animal) and mechanisms of transport from a source to a field (e.g., watersheds, animal intrusions, aerosols). However, the mechanisms of survival of pathogens in complex environments, and locations and conditions where amplification of their numbers might occur, have not been well documented. Reviews describing the sources, fate, and transport of pathogens as potential risk factors relevant to preharvest contamination have been published previously; they provide background and specific details that will be summarized in this review. Studies of the incidence and fitness of E. coli O157:H7 and Salmonella in the produce production environment associated with leafy vegetables, tomatoes, and cantaloupes will be emphasized since they have been associated with multiple outbreaks suspected of being caused by preharvest contamination in the U.S. and Mexico. However, the same environmental factors described for these two pathogens and implicated commodities will apply generally to other pathogens associated with produce contamination, except for specific fitness characteristics that might be linked to a specific commodity. Information related to the incidence and survival of bacterial pathogens and fecal indicators in the production environment, and potential transport processes and risk factors associated with growing fresh produce in dynamic, agricultural regions are presented.

Outbreaks Associated with Selected Fresh Produce Commodities An unintended consequence of the increased production and consumption of fresh produce is an increase in the number of outbreaks of foodborne illness (CSPI 2007; Sewell and Farber 2001; Sivapalasingam and others 2004). The produce items and types of pathogens associated most frequently with outbreaks in the U.S. (Sivapalasingam and others 2003) and other industrialized countries (Sewell and Farber 2001) have been reported previously, and documented in previous review articles about this subject (Nguyen and Carlin 1994; Beuchat 1996, 2006; Seymour and Appleton 2001; Harris and others 2003; Mandrell and Brandl 2004; Johnston and others 2006b). However, selected data related to outbreaks linked with fresh leafy vegetables and tomatoes will be emphasized in this review in support of the theory that multiple recent outbreaks have resulted from preharvest contamination, especially large multistate or multicountry outbreaks (Table 1.1). The total number of cases of foodborne illness in the United States has been estimated to be approximately 76 million illnesses per year, associated with 325,000 hospitalizations and 5000 deaths (Mead and others 1999). In a recent review of outbreaks associated specifically with fresh produce, the U.S. Centers for Disease Control and Prevention (CDC) analyzing data from the CDC Foodborne Outbreak Surveillance

Table 1.1. Selected outbreaks associated with enteric human pathogens and fresh producea Locationb

E. coli O157:H7

MonthYear Jul-95

MT

74

E. E. E. E.

coli O157:H7 coli O157:H7 coli O157:H7 coli O157:H7

Sep-95 Sep-95 Oct-95 May-96

ME ID OH IL, CT

E. E. E. E. E. E. E. E.

coli O157:H7 coli O157:H7 coli O157:H7 coli O157:H7 coli O157:H7 coli O157:H7 coli O157:H7 coli O157:H7

Jun-96 May-98 Sep-98 Feb-99 Sep-99 Sep-99 Oct-99 Oct-99

E. coli O157:H7 E. coli O157:H7 E. coli O157:H7

Oct-99 Jul-02 Nov-02

E. coli O157:H7

Pathogen

Reference

Known or Suspected Vehicle Lettuce, Romaine

Source Regionc MT, WA

30 20 11 61

Lettuce, Iceberg Lettuce, Romaine Lettuce Lettuce, Mesclun mix

Unknown Unknown Unknown CA (SV)

NY CA MD NE CA WA OH, IN OR

7 2 4 65 8 6 47 3

Unknown Unknown Unknown Unknown CA (SV) CA (SV) Unknown CA (SV)

41 29 24

CA (SV) CA (SV) CA (SJoV)

CDC 1999 CDC 2002 CDC 2002

Sep-03

PA WA IL, WI, MN, SD, UT CA

Lettuce, Mesclun Lettuce, salad Lettuce Lettuce, salad Lettuce, Romaine Lettuce, Romaine Lettuce, salad Lettuce, Romaine hearts Lettuce, Romaine Lettuce, Romaine Lettuce

Ackers and others 1998 CDC 1995 CSPI 2008 CDC 1995 Hilborn and others 1999 CDC 1996 CDC 1998 CDC 1998 CDC 1999 CDC 1999 CDC 1999 CDC 1999 CDC 1999

CA (SV)

E. coli O157:H7

Sep-03

ND

5

CDHS 2004a CDC 2003

E. coli O157:H7

Oct-03

CA

16

Lettuce, Iceberg/ Romaine Lettuce, mixed with Romaine Spinach

E. coli O157:H7 E. coli O157:H7

Nov-04 Sep-05

NJ MN

6 11

CA (SV) CA (SV)

E. coli O157:H7

Aug/ Sep-05

Sweden

Lettuce Romaine, also vegetables Lettuce, iceberg

E. coli O157:H7

Aug/ Sep-06 Nov-06

Multi (26 states) NJ, NY, PA, DE MN, IA, WI WA

CA (SJuV)

71

Spinach, baby, bagged Lettuce, Iceberg

81

Lettuce, Iceberg

CA (CentV)

10

Lettuce, Romaine

CA (SV)

E. coli O157:H7 E. coli O157:H7 E. coli O157:H7

Nov/ Dec-06 May-08

No. Ill

57

135

>200

Unknown CA (SV)

Sweden

CA (CentV)

S. Saphra

Feb/ May-97

Multi

24

Cantaloupe

Mexico

S. Poona

Spring00–02d

Multi, Canada

58

Cantaloupe

Mexico

CDHS 2004b CDC 2004 MDPH 2006 Soderstrom and others 2008 CalFERT 2007b,c CalFERT 2007a CalFERT 2008 WDOH 2008 MohleBoetani and others 1999 MMWR 2002

7

Table 1.1. Continued Pathogen S. Litchfield

MonthYear Jan/Mar-08

Locationb

No. Ill

Multi, Canada U.K.

51

Known or Source Suspected Vehicle Regionc Cantaloupe Honduras

Reference

19

Vegetables, bagged

Italy, Spain

21

Rucola (arugula)

Italy

Sagoo and others 2003 Nygard and others 2008 Campbell and others 2001 Hedberg and others 1999 Hedberg and others 1999 Cummings and others 2001 Srikantiah 2002; Gupta and others 2007 Greene and others 2008 Gupta and others 2007 Gupta and others 2007 MMWR 2007a; Greene and others 2008 MMWR 2007a MMWR 2007a MMWR 2007a Isaacs and others 2005 MMWR 2004

S. Newport

May/ Jun-01

S. Thompson

Oct/Dec-04

Multi, Europe

S. Thompson

Mar-99

CA

741

Cilantro

Mexico (suspected)

S. Javiana

Jun/ Aug-90

IL, MI, MN, WI

176

Tomatoes

SC

S. Montevideo

Jun/ Aug-93

IL, MI, MN, WI

100

Tomatoes

SC

S. Baildon

Dec-98– Jan-99

Multi

86

Tomatoes

FL

S. Javiana

Jun/Jul-02

FL

141

Tomatoes, prediced

?

S. Newport

Sep/Oct-02

Multi

510

Tomatoes

VA

S. Braenderup

Jul-04

Multi

125

Tomatoes

FL

S. Javiana and other serovars

Jul-04

Multi

429

Tomatoes, presliced

?

S. Newport

Jul/Nov-05

Multi

72

Tomatoes

VA

S. Braenderup

Multi

82

Multi

115

Tomatoes, prediced Tomatoes

FL

S. Newport

Nov/ Dec-05 Jul/Nov-06

S. Typhimurium

Sep/Oct-06

Multi

190

Tomatoes

OH

S. Enteritidis

Oct-00– Jul-01

Multi, Canada

168

Almonds, raw

CA

S. Enteritidis

Sep-03– Apr-04

Multi, Canada

29

Almonds, raw

CA

8

?

CDC 2008a

Enteric Human Pathogens Associated with Fresh Produce

9

Table 1.1. Continued MonthYear Dec-05– Aug-06

Locationb

S. St. Paul

Apr/Jul-08

Multi

Shigella flexneri

May-01

NY

886

Shigella sonnei

Aug-04

Multi

116

Yersinia pseudotuberculosis

Oct-98

Finland

47

Yersinia pseudotuberculosis

Aug/ Sep-06

Finland

>400

Pathogen S. Enteritidis

Sweden

No. Ill 15

>1200

Known or Suspected Vehicle Almonds, raw

Source Regionc CA

Pepperse Tomatoes

Mexico (suspected) FL

Carrots

CA?

Lettuce, iceberg

Finland

Carrots

Finland

Reference Ledet Muller and others 2007 CDC 2008b Reller and others 2006 Gaynor and others 2009 Nuorti and others 2004 RimhanenFinne and others 2009

a

Outbreaks included have been selected based on location or suspected preharvest contamination. Outbreaks associated with almonds have been included because of recurrent outbreaks suspected of being linked to a common location. b U.S. states are designated by the two-letter abbreviations; Multi = multiple states involved. c SV, Salinas Valley, CA; SJoV, San Joaquin Valley, CA; SJuV, San Juan Valley, CA; CentV, Central Valley, CA. Some location information was provided by California Dept. of Public Health (personal communication). Unknown = traceback not done or incomplete. d Represents three outbreaks (2000, 2001, 2002); the 2000 and 2002 outbreaks were caused by the same strain. e Cases occurred in 43 states, Washington, D.C., and Canada; jalapeño peppers grown in Mexico are suspected as the cause of a majority of cases. Serrano peppers and tomatoes not yet cleared as cause of other illnesses, at the time of preparing this review.

System for 1973–1997, identified 190 outbreaks associated with produce, 16,058 illnesses, 598 hospitalizations and 8 deaths (Sivapalasingam and others 2003). An updated review by CDC of outbreaks associated specifically with leafy greens between 1973 and 2006 determined that 502 outbreaks, >18,000 illnesses, and 15 deaths occurred, with 30 of the outbreaks caused by E. coli O157:H7, 35 by Salmonella, and 196 by Norovirus (Herman and others 2008). Comparison of the numbers in these two studies reflects the fact that produce-associated outbreaks linked with a known food item increased from 0.7% of all foodborne outbreaks in the 1970s to 6% in the 1990s and has increased further to the present. The bacterial, viral, and protozoal pathogens associated with fresh produce outbreaks (number of outbreaks) in the U.S. between 1973 and 1997 include the following: Salmonella (30 outbreaks), E. coli O157:H7 (13), non-O157 E. coli (2), Shigella (10), Campylobacter (4), Bacillus cereus (1), Yersinia enterocolitica (1), Staphylococcus aureus (1), Hepatitis A (12), Norovirus (9), Cyclospora cayetanensis (8), Giardia

10

Section I. Microbial Contamination of Fresh Produce

lamblia (5), and Cryptosporidium parvum (3); an additional 87 outbreaks were documented without any etiology identified (Sivapalasingam and others 2003). The produce items implicated most frequently in outbreaks are “salad” lettuce, seed sprout, melon and cantaloupe (Sivapalasingam and others 2003). Multiple sprout outbreaks of S. enterica and E. coli O157:H7 illness occurring worldwide have been associated usually with sprouts (e.g., alfalfa, mung bean, radish) grown from contaminated seed (Michino and others 1999; Breuer and others 2001; Mahon and others 1997; Proctor and others 2001; Mohle-Boetani and others 2009). The seeds are harvested in different parts of the world (e.g., U.S., Australia, China) under agricultural conditions that in many cases are not controlled well for microbial safety, considering the eventual ready-to-eat product to be produced. The sprouting process involves ideal conditions for enriching even a small concentration of pathogen that may contaminate even a small proportion of the seeds. These conditions emphasize again the importance of the quality of the preharvest environment to produce production at every step of the production cycle, including seed and transplant production, harvesting, and the fields prior to and following harvest (water, fertilizers, crop debris, human and animal visits). Contaminated seeds are not a major risk factor probably in the nonsprout outbreaks to be documented further here; however, seeds should be appreciated as an early preharvest control point in fresh produce production. Preharvest contamination is suspected in numerous outbreaks associated with leafy vegetables (lettuce and spinach), tomatoes, cantaloupes, and possibly other commodities (e.g., jalapeño peppers, April–July, 2008). For U.S.-grown leafy vegetables alone, there have been more than 20 foodborne outbreaks since 1995 linked to contamination by E. coli O157:H7, resulting in at least 600 reported illnesses and 5 deaths. Since 2000, at least 12 outbreaks have been linked to Salmonella contaminated tomatoes (>1600 cases) and 3 outbreaks linked to Salmonella contaminated cantaloupes (72 cases) (Table 1.1). It is worth noting that, during the final preparation of this review, a major ongoing outbreak of Salmonella in St. Paul is associated with jalapeño peppers grown in Mexico and distributed by a company in Texas occurred (CDC 2008b). This was the first reported outbreak associated with this food item; however, additional details will be required to determine whether the contamination occurred on the farm or postharvest (packinghouse). Several outbreaks suspected of being associated with preharvest contamination of tomatoes, lettuce, and carrots by Shigella and Yersinia species also occurred (Table 1.1). These outbreaks have been listed to emphasize some emerging produce-pathogen issues of concern: preharvest contamination, pathogen persistence and/or fitness in the environment, and diversity of pathogens implicated depending upon local growing conditions (Table 1.1; e.g., leafy vegetables—Western U.S./Sweden/Italy, tomatoes—Eastern U.S., cantaloupe—Mexico, Yersinia—Finland). Previous epidemiological studies of fresh produce outbreaks often lacked definitive evidence of the source of contamination and a step within the food production and processing chain where contamination could have occurred. However, traceback investigations of E. coli O157:H7–leafy vegetable outbreaks determined that 12 of them were linked probably to commodity grown on farms in the Salinas Valley, a region located on the Central Coast of California, and the major supplier of fresh produce to the U.S. market (Table 1.1; see references for additional details). Indeed, baby spinach linked to a large multistate outbreak of E. coli O157:H7 in the late spring of 2006 was grown in a valley adjacent to the Salinas Valley (CalFERT 2007b;

Enteric Human Pathogens Associated with Fresh Produce

11

Cooley and others 2007). Similarly, recurrent outbreaks associated with tomatoes were suspected of being grown on farms in Virginia and Florida, and outbreaks with cantaloupes on farms in Mexico (Table 1.1). Produce outbreaks linked to a region where a large amount of fresh produce is grown is logical; however, a number of factors revealed by recent outbreak investigations are relevant to concepts of where, when, and how contamination occurs. As noted, outbreaks have been associated with commodities grown in the same region and with preharvest contamination rather than later in the distribution chain (e.g., transport or restaurant). Also, pathogen strains of the same serovar could be isolated from watersheds in the vicinity of implicated fields, and for the first time in recent outbreak investigations, E. coli O157:H7 and Salmonella strains indistinguishable from the clinical outbreak strains were isolated from environmental samples (CalFERT 2007b, 2007c, 2008; Cooley and others 2007; Greene and others 2008). Therefore, accurate information about the fate and transport processes relevant to contamination processes and the fitness of pathogens near, on, or in produce plants in the field is critical for developing strategies for minimizing preharvest contamination of produce.

Incidence of Human Pathogens on Fresh Produce How often are produce items contaminated with pathogens? The incidence is very low generally, but any amount may be too much considering the low infectious dose for some of the pathogens, especially E. coli O157:H7 on raw produce. The incidence of major foodborne pathogens on different items of fresh produce and in animal hosts has been reported in numerous studies, in addition to data relevant for assessing the survival and fitness of pathogens in agricultural environments such as manure, water, and soil. These data are relevant to consider also for identifying potential point sources and transport processes of pathogens in production environments linked to outbreaks. Beuchat published in 1996 one of the first and best reviews of reported incidence of common foodborne pathogens on ready-to-eat vegetables, and the potential sources of the pathogens and mechanisms of contamination (Beuchat 1996). The incidence, growth, and survival of foodborne pathogens in fresh and processed produce has been reported also in comprehensive reviews by Nguyen-the and Carlin (Nguyen-the and Carlin 2000) and Harris and others (see Tables I-1 to I-7 in Harris and others 2003), and other recent reviews (Johnston and others 2006b; Beuchat 2006; Mandrell and Brandl 2004). Although distinctions between pre- and postharvest contamination are not provided generally, these reviews provide useful summaries of the different methods for isolating pathogens—for example, Salmonella, Listeria, Yersinia, Campylobacter species, E. coli O157:H7, and generic E. coli—from multiple types of produce items that were grown in different regions of the world. The incidence of pathogens reported in these separate studies often was between 0 and <10% of all samples tested, with an occasional incidence of >20% reported (Nguyen-the and Carlin 1994; Harris and others 2003; Mandrell and Brandl 2004). Moreover, in the few studies reporting the concentration of pathogen per gram of produce, the levels were low in most studies, even for generic E. coli, as a measure of possible fecal contamination. For example, the percentages of positives out of 774 total samples tested for Salmonella on leafy vegetables or salad in eight separate studies were 0 (0/151), 0 (0/63), 0.6 (1/159), 0.9 (1/116), 3.5 (2/57), 6.3 (5/80), 7.1

12

Section I. Microbial Contamination of Fresh Produce

(2/28), and 68% (82/120) (Harris and others 2003). In contrast, all 214 samples of lettuce or salad mix tested for E. coli O157:H7 in large U.K. and U.S. studies were negative (Harris and others 2003). Of >3,800 ready-to-eat salad vegetables from retail markets sold in the U.K., only 0.2% were positive for Salmonella; an additional 0.5% were considered of poor quality due to contamination with E. coli or L. monocytogenes at >100 CFU per g of product (Sagoo and others 2003). A survey of “minimally processed” vegetables in Brazil determined that 4 of 181 samples (2.2%) were contaminated with Salmonella (Froder and others 2007). Similarly, 180 fresh vegetable samples surveyed in South Africa identified 4 (2.2%) contaminated with E. coli O157:H7, and reported levels of E. coli O157:H7 as high as 1,600,000 CFU/g of spinach (Abong’o and others 2008). These results reflect the tremendous diversity of produce quality depending upon spatial and temporal factors, and possibly methodological factors. Multiple outbreaks of Salmonella illness associated with tomatoes have occurred recently, but surveys of tomatoes for the incidence of pathogens have been limited. Of 123 samples of domestic (U.S.) tomatoes tested by the U.S. FDA-CFSAN starting in May, 2001, none were positive for Salmonella or E. coli O157:H7 (FDA-CFSAN 2001b); also, 0/20 imported tomato samples collected starting in March, 1999 were negative for both pathogens (FDA-CFSAN 2001a). However, 11 of 151 imported and 4 of 115 domestic cantaloupe samples in the same surveys were positive for Salmonella or Shigella. These results appear consistent with the fact that multiple outbreaks occurred in 1997, 2000, 2001, and 2002 due to Salmonella-contaminated cantaloupe imported from Mexico (Table 1.1). A large survey of cantaloupe and environmental samples from six farms and packing plants in South Texas and three farms in Mexico resulted in 5/950 and 1/300 cantaloupes positive for Salmonella, respectively (Castillo and others 2004). Irrigation-related samples of cantaloupe production (e.g., water source, tank, in field) revealed a higher incidence of Salmonella for both Texas and Mexico farms: 13/140 (9.2%) and 10/45 (22.2%), respectively, compared to the commodity. Moreover, generic E. coli was isolated at significant levels from some of the samples of Texas and Mexico cantaloupe (3.9%, 25.7%) and Texas and Mexico irrigation water (22.8% and 31.1%, respectively) (Castillo and others 2004). It is noteworthy that none of the 150 field and prewash cantaloupes from Mexico were positive for E. coli, compared to 39/75 (52%) and 38/75 (51%) positive samples for the postwash and packed cantaloupe, respectively. Although the concentrations of Salmonella and generic E. coli in these samples were not reported, these results reflect a prevalence of fecal contamination of water sources (well, river, aquifer, canal, dam), suggesting they may be sources of both pre- and postharvest contamination. Fecal contamination of postharvest processing water is an obvious potential source of cross-contamination of cantaloupes (Castillo and others 2004). The fitness characteristics of pathogens in the environment are important for their long-term survival and exposure to produce. The long-term persistence in the environment of some foodborne pathogen strains is exemplified by a strain of S. Enteritidis implicated in at least one major outbreak, and possibly a minor outbreak, associated with raw almonds in 2000/01 (Isaacs and others 2005) and 2005/06 (Ledet Muller and others 2007), respectively. The S. Enteritidis outbreak strain, subtyped as phage type 30, was isolated from a suspect orchard at multiple times over at least a 5-year period,

Enteric Human Pathogens Associated with Fresh Produce

13

and with increasing frequency in samples collected during and following harvests (Aug–Dec) and following rain events (Uesugi and others 2007). Salmonella strains isolated during the 5-year study were all phage type 30 and indistinguishable from the clinical outbreak strains (or one band difference) by two-enzyme pulsed field gel electrophoresis (PFGE) analysis. Although it was probable that almonds became contaminated by pathogens present in soil/dust where almonds were dropped and then harvested by sweepers, the original source of the outbreak-related strain was never identified, nor were any suspect practices (Uesugi and others 2007). The extended persistence of any pathogens in an agricultural environment, especially strains that have the potential to cause an outbreak, raises questions relevant to other produce-related outbreaks. Is contamination periodic and cumulative or due to major isolated contamination events? Do persistent strains reflect selection and evolution of special fitness characteristics in a specific environment (e.g., orchard environment; almond, leafy vegetable, tomato surface)? Is the incidence or concentration of pathogens greater now than in the past? Does pathogen survival at low concentrations in harsh soil conditions (dry, high UV) with subsequent resuscitation/amplification (rain/moisture, low UV) relate to virulence? Do certain wildlife species (e.g., mammalian, avian, amphibian) become colonized and high shedders of pathogen and associated with persistent contamination? These and other questions stimulated by recent outbreaks are difficult to answer, but they assist in focusing on areas for further research.

Incidence of Generic E. coli on Produce Increased concerns in the U.S. and other countries about produce-associated outbreaks (Table 1.1) have stimulated initiation of multiple surveys of fresh produce for selected pathogens, and also surveys of the incidence of generic E. coli as an indicator of fecal, and potential pathogen, contamination. The results from some of these studies, including recent surveys, are presented to indicate the general microbiological quality of different types of produce grown in different regions conventionally or organically, and tested at different stages of the pre- and postharvest cycle. A survey of produce items (e.g., arugula, cantaloupe, cilantro, parsley, spinach) collected between November 2000 to May 2002 from 13 farms in the southeastern U.S. revealed E. coli levels ranging from 0.7 to 1.5 log CFU/g for field or packing-shed produce (Johnston and others 2005). All samples were negative for L. monocytogenes and E. coli O157:H7; however, 3 of 398 samples tested for Salmonella were positive (0.7%). A similar survey by the same investigators comparing produce grown in the southern U.S. and Mexico involved testing 466 produce items obtained from packing sheds between November 2002 and December 2003. Levels of E. coli ranged between 0.7–1.9 and 0.7–4.0 log CFU/g for Mexican and southeastern U.S. produce, respectively (Johnston and others 2006a). All samples were negative for E. coli O157:H7, Salmonella, and Shigella; however, three domestic cabbage samples were positive for L. monocytogenes (0.6% of total produce samples; 7% of cabbage samples). A variety of fresh produce items grown conventionally or organically on farms in Minnesota were picked between May and September 2002 and surveyed for microbiological quality (Mukherjee and others 2004). E. coli incidence was 4.3, 11.4, and 1.6%

14

Section I. Microbial Contamination of Fresh Produce

for 117 certified organic, 359 noncertified organic, and 129 conventional produce items, respectively, and the average E. coli counts for the positive samples was reported as 3.1 log MPN/g. The E. coli incidence was sixfold higher on organic versus conventional produce and 2.4-fold higher on produce from farms using cattle manure compared to farms using other types of manure. Noncertified organic lettuce had the highest incidence (12/39, 30.8%) for any item with more than 10 samples tested (Mukherjee and others 2004). The microbiological quality of ready-to-eat produce has been surveyed in other parts of the world. In a study of leafy salads collected from retail markets in Brazil, >85% of 181 samples were reported to have >4 logs Enterobacteriaceae per g (Froder and others 2007). Leafy vegetable salads collected postpreparation from 16 university restaurants in Spain yielded 26% positive for E. coli (Soriano and others 2001). In contrast, only one (lettuce) of 50 produce items collected from retail and farmers markets in Washington, D.C. was positive for E. coli (Thunberg and others 2002). These results suggest major diversity in E. coli incidence depending upon the size, time, and location of the study, and possibly differences in the sensitivity of methods. A study initiated by the USDA Agricultural Marketing Service in 2002 and coordinated with state and other federal agencies to survey the microbial quality of fresh produce items available at terminal markets and wholesale distribution centers continues as of 2008 (USDA-AMS-MDP 2008). The cumulative results over 6 years, with approximately 65,000 samples analyzed to date, provides a significant data set for analyzing spatial, temporal, and other factors related to produce contamination using E. coli incidence as the measure of fecal contamination. Multiple commodities, both domestic and imported, have been tested during the program (e.g., cantaloupe, leaf and romaine lettuce, tomatoes, green onions, and alfalfa sprouts) for generic E. coli, E. coli “with pathogenic potential” (including E. coli O157:H7), and Salmonella. The results from tests of >59,000 samples from 2002–2007 indicate that low levels of generic E. coli are common on produce items collected at the distribution stage of the postharvest production cycle compared to levels on produce in the field (Table 1.2); however, only 1.5 to 2.7% of the samples by year were positive for E. coli at concentrations >10 MPN/ml (USDA-AMS-MDP 2008). Moreover, E. coli with pathogenic potential based on PCR results for various virulence factors, including shigatoxin 1 and 2 (Stx 1 and 2), ranged from 0.1 to 0.4% of all samples tested each year. Examples of individual produce items having a high percentage of samples positive

Table 1.2. Incidence of E. coli on selected fresh produce items obtained and tested in years 2002–2007, as part of the USDA, Agricultural Marketing Service, Microbial Data Program (USDA-AMS-MDP 2008) Categories

2002

2003

2004a

2005

Total no. produce samples tested

10,319

10,972

11,211

11,508

No. positive for E. coli

b

% positive for E. coli % E. coli samples with virulence trait(s) a b

759 7.4 0.6

730 6.7 0.4

3,226 28.8a 0.4

4,201 36.5 0.4

Generic E. coli method was modified in 2004 and again in 2007. A sample was considered positive if >0.03 MPN/ml rinse solution was determined.

2006

2007a

7,646

5,279

1,569 20.5 0.4

4,420 83.8 0.1

Enteric Human Pathogens Associated with Fresh Produce

15

for E. coli were cantaloupe (2004 and 2005, 26–32%), leaf and/or romaine lettuce (2004 and 2005, 25–44%), cilantro (2004 and 2005, 66–71%), and parsley (2004 and 2005, 72%) (USDA-AMS-MDP 2008); data not shown. A similar survey for E. coli on 1,183 produce items grown in Ontario, Canada, in 2004 resulted in a 0, 1.3, 6.5, 11.6, 4.9, and 13.4% reported incidence for tomato, cantaloupe, conventional leaf lettuce, organic leaf lettuce, cilantro, and parsley, respectively (Arthur and others 2007a). However, the concentrations of E. coli ranged from >5 to 290 CFU/g for leaf lettuce, to <5 to 7,600 and 16,000 CFU/g for cilantro and parsley, respectively. Only two samples yielded a potential pathogen: S. Schwarzengrund in a sample each of Roma tomato and organic leaf lettuce (Table 1.2) (Arthur and others 2007a). Finally, a recent study of 100 domestic bagged cut spinach and lettuce mixes (conventional and organic) for total bacterial, coliform, and E. coli counts reported means of 7.0 to 7.7 log CFU/g, <0.5 to >4.0 log MPN/g and 3 to 9.2 MPN/g (16% of samples), respectively, depending upon the product; 12.1% conventional and 16.6% organic spinach and 23.1% conventional and 6.3% organic lettuce mix samples were positive for E. coli (Valentin-Bon and others 2008). These results for bagged leafy greens from retail markets are consistent with surveys of ready-to-eat produce in the U.S. and other countries noted above, and other surveys reporting relatively high incidences of E. coli in specific produce items such as lettuces, parsley, and cilantro (Soriano and others 2001; Froder and others 2007; USDA-AMS-MDP 2008). Significant correlations between the levels of E. coli contamination of produce and incidences of major bacterial enteric pathogens are lacking. Thus, E. coli incidence can be considered simply an indicator of potential minor or major preharvest contamination, and a risk factor for additional postharvest contamination, cross-contamination during washing, or amplification of bacteria (pathogen) during transport and storage. E. coli incidence serves as a moderately effective measure of changes in fecal microbial flora during the produce production and processing cycle, and for assessing the potential for pathogenic strains, if they were to be present, to survive under the same produce processing conditions. The concentration of E. coli may be a more relevant indicator of the risks associated with human consumption of a contaminated produce item. Evidence of fecal contamination as high as 50–70% on some produce items does not correlate necessarily to a higher incidence of illness, unless undetected sporadic illness is occurring. Although major outbreaks are of concern, it should be emphasized that relative to the number of consumptions of ready-to-eat produce (and tree nuts) (many billions), outbreaks are not frequent, causing an extremely low number of known total cases per total consumptions; however, some cases are sporadic probably and never linked to a food source. Nevertheless, vigilance and research are important to identify what is probably a rare convergence of events and/or specific circumstances that result in a major outbreak of disease, some of it severe, and thus, a noticeable event. The relatively low incidence of pathogens on produce measured in surveys seems consistent with the speculation that incidence is very rare and occurs only after multiple unusual circumstances that result also in an outbreak. Surveys of produce are informative because they provide a measure of the background incidence of indicators of fecal contamination and pathogens related to dynamic spatial, temporal, and geographic factors. Incidence in the absence of illness or outbreaks also is informative.

16

Section I. Microbial Contamination of Fresh Produce

Animal Sources of Enteric Foodborne Pathogens Relevant to Produce Contamination Carriage of pathogens by food animals is a critical factor relevant to many outbreaks associated with produce, meat, milk, and other food products. Evidence for the colonization of cattle (Elder and others 2000; Hussein and Bollinger 2005; Fegan and others 2005; Low and others 2005; Dargatz and others 2003), swine (Chapman and others 1997; Jay and others 2007), sheep (Ogden and others 2005), poultry (Chapman and others 1997; Rose and others 2002; Foley and others 2008; McCrea and others 2006), and multiple species of wild animals (Ejidokun and others 2006; Hernandez and others 2003; Kirk and others 2002; Sargeant and others 1999; Pritchard and others 2001; Wetzel and LeJeune 2006) by E. coli O157:H7, S. enterica, and C. jejuni (Miller and Mandrell 2006) has been documented. Pathogen colonization of livestock and wild animals is a dynamic process depending upon how and when pathogens are encountered in the environment (food, grass, water), pathogen fitness in the environment and animal GI tracts (viability, dose), animal contact/commingling and movement, immunity, and fecal shedding. In addition, there are unknown factors that might enhance or diminish pathogens in particular environments, for example, weather conditions, feed, predation, or antimicrobials. One or more of these factors may be important in initiating or contributing to the size of an outbreak. Studies documenting the incidence of E. coli O157:H7 and Salmonella in animals are summarized in Table 1.3. Details regarding the methods, periods, locations, and samples studied can be obtained from the original papers cited. E. coli O157:H7 and Non-O157 STEC Cattle are major carriers of E. coli O157, non-O157 shigatoxin-positive E. coli (STEC), S. enterica and C. jejuni strains (Table 1.3). Strains of the same serovars as those associated with produce outbreaks have been isolated frequently from cattle. Similarly, sheep, pigs, chickens, and turkeys are common or intermittent carriers of these pathogens, and a variety of wildlife species carry these pathogens or related pathogens (Tables 1.1 and 1.3). For example, E. coli O157:H7 and nonO157 STEC strains have been isolated from deer (Keene and others 1997; Sargeant and others 1999; Fischer and others 2001; Dunn and others 2004; Renter and others 2006), feral swine (Jay and others 2007), pigeons (Morabito and others 2001), seagulls (Makino and others 2000), starlings, horses, dogs (Hancock and others 1998), barn flies (Keen and others 2006), and slugs (Sproston and others 2006). Salmonella has been isolated from deer (Branham and others 2005; Renter and others 2006), badgers (Nielsen and others 1981), wild mice (Tablante and Lane 1989), wild turtles and tortoises (Hidalgo-Vila and others 2007), and a variety of wild birds (Fenlon 1981; Wahlstrom and others 2003; Hughes and others 2008). The concentration of pathogen in wildlife samples is not well documented; thus, the shedding status of wildlife compared to livestock is unclear. Moreover, the quantity of feces shed by different species of wildlife per animal or for a population in a region is unknown, so data relevant to the total amount of pathogen disseminated by a species in any spatial and temporal context also are unknown. The amount of pathogen shed by an animal is extremely relevant epidemiologically for identifying

Table 1.3. Selected studies reporting incidence of E. coli O157, S. enterica, and C. jejuni in livestock and wild animal feces Pathogen

Animals

Incidence in Feces

Reference

E. coli O157

Beef cattle U.S. (multiple states) and multiple countries

Review of 39 separate studies: Hussein and Bollinger 2005

E. coli O157

Beef cattle (Scotland)

E. coli O157:H7

Beef cattle Dairy cattle U.S. (multiple states) Beef cattle, hides U.S. (multiple states) Zebu (“humped cattle”) Beef cattle U.S. (multiple states) and multiple countries

0.3–19.7%, feedlot a 0.7–27.3%, pasture 0.9–6.9%, range 0.2–27.8%, slaughter 3.4%, some high shedders b 3.6% 3.4%

E. coli O157:H7 E. coli O157:H7 Non-O157 E. coli

E. coli O157

Sheep (U.K.)

E. coli O157

Sheep (U.K.)

9–85% 5.4% 2.1–70.1% overall 4.6–55.9%, feedlot 4.7–44.8%, grazing 2.1–70.1%, slaughter 6.5%, some high shedders b 2.2%

E. coli O157

Sheep (U.K.)

0.7%

E. coli O157:H7 E. coli O157:H7 E. coli O157

Sheep (U.S.) Sheep (Spain) Pigs (U.K.)

4.4% 7.3% 0.4%

E. coli O157:H7 E. coli O157:H7

Pigs (U.S.) Pigs (Japan)

2.0% 1.4%

E. coli O157 E. coli O157 E. coli O157:H7 E. coli O157:H7 E. coli O157:H7

Pigs (U.K.) Pigs (U.K.) Pigs (U.S.) Pigs (U.S.) Feral swine (U.S.)

6.7% 0.3% 1.2% 8.9% 14.9%c

E. coli O157:H7

0.9% 7.5%

E. coli O157 E. coli O157 E. coli O157:H7 E. coli O157

Chickens Turkeys (U.S.) Chickens (U.K.) Goats (U.K.) Deer (US) 3/32 pellets Deer (U.S.)

3.8% 28% 9.4% 2.4%d

E. coli O157:H7

Deer (U.S.)

0.5%e

E. coli O157:H7 E. coli O157:H7 E. coli O157

Deer (U.S.) Deer (U.S.) Rabbits

0.25%f 0.3–0.4%g ?

E. coli O157:H7

Ducks

?

Matthews and others 2006 Doane and others 2007

Arthur and others 2007b Tuyet and others 2006 Review of 21 separate studies: Hussein and Bollinger 2005 Ogden and others 2005 Chapman and others 1997 Milnes and others 2008 Keen and others 2006 Oporto and others 2008 Chapman and others 1997 Feder and others 2003 Nakazawa and Akiba 1999 Cooper and others 2007 Milnes and others 2007 Keen and others 2006 Doane and others 2007 Jay and others 2007 Doane and others 2007

Cooper and others 2007 Cooper and others 2007 Keene and others 1997 Sargeant and others 1999 Fischer and others 2001 Renter and others 2001 Dunn and others 2004 Pritchard and others 2001; Leclercq and Mahillon 2003 Leclercq and Mahillon 2003

17

Table 1.3. Continued Pathogen

Animals

Incidence in Feces

Reference

E. coli O157:H7

Fish

4.7%

Tuyet and others 2006

E. coli O157

Rats (Norway)

40%h

Cizek and others 1999

Non-O157 EHEC

Rabbits

9–25%i

S. enterica

Cattle (U.S.)

6.3%

S. enterica

Cattle (U.S.)

4.4%

S. enterica

Cattle (Australia)

S. enterica S. enterica

Cattle (U.K.) Cattle (U.S.)

S. enterica

Dairy (U.S.)

S. enterica

Goats (U.S.)

4.5%, grass-fed 9.0%, feedlot 1.4% 13–72%, cows 20–71%, calves 60–63%, soil 53–67%, water 46–71%, air 13–63%, bird feces 24–85%, insects Feed, 21–92% 3.7%

Garcia and Fox 2003 Dargatz and others 2003 Barkocy-Gallagher and others 2003 Fegan and others 2004

S. enterica

Sheep (U.S.)

7.3%

S. enterica S. enterica S. enterica

Sheep (U.K.) Pigs (U.K.) Poultry (U.S.)

S. enterica

Poultry (U.S.)

S. enterica

Poultry

1.1% 23.4% 50.8%, transport pads 18.7%, flies 14.2%, drag swabs 12%, boot swabs 10.5%, by flocks 1.1%, by row 13.0%, by flocks

S. enterica

Deer (U.S.)

7.7%, rumen

S. enterica S. enterica

Deer (U.S.) Wild tortoises Wild turtles (Spain) Wild birds (U.S.) Wild birds (U.K.)

1.0% 100% 12–15%

S. enterica S. enterica Salmonella C. jejuni/C. coli

a

Seagulls Cattle, chickens (live), geese, ducks, pigs, sheep (Multiple countries)

1.2–3.2% 0.015% 12.9% 0–100%j

Milnes and others 2007 Pangloli and others 2008 Pangloli and others 2008

Branham and others 2005 Branham and others 2005 Milnes and others 2007 Milnes and others 2007 Bailey and others 2001

Kinde and others 2004 Rasschaert and others 2007 Branham and others 2005 Renter and others 2006 Hidalgo-Vila and others 2007 Kirk and others 2002 Hughes and others 2008 Fenlon 1981 Review of >20 studies; Miller and Mandrell 2006

Ranges of incidence reported for multiple studies; majority of isolates were E. coli O157:H7. High shedders, >10,000 CFU/g; majority positive for Stx2. c Collected from one ranch in California. d 5/212 white-tailed deer. e 3/609 individually sampled deer, 1997 and 1998. f 4/1608 mostly white-tailed deer, Nebraska, 1998. g 1 of 338 hunter-harvested deer, 1 of 226 captive herd deer, Louisiana, 2000–01. h 4 of 10 rats; however, negative for H7. i Laboratory rabbits; all EHECs positive for Stx1. j Cattle, 62% average for 14 studies; chicken (live), 64% for 20 studies; geese/ducks, 55% for 6 studies. b

Enteric Human Pathogens Associated with Fresh Produce

19

potential sources of pathogen and relevant risk factors for contamination of produce (Chase-Topping and others 2007). The incidence data listed in Table 1.3 are from selected recent studies; the data reflect the dynamic nature of the incidence associated with different animal hosts, spatial and temporal differences, and a variety of different methods. In a recent review by Hussein and Bollinger, 39 reported studies of the incidence of E. coli O157:H7 in thousands of cattle fecal samples from feedlots, pasture/range, and entering slaughter ranged from 0.2 to 28%, depending upon the study and the cattle feeding or production process (Hussein and Bollinger 2005). A previous review of some of the same studies involving animals in Asia, Australia, Europe, and North America (sampling periods between 1991–1999) reported incidence in fecal samples in the range of 0.1 to 62% (Duffy 2003). Indeed, the common occurrence of E. coli O157:H7 in cattle is consistent with numerous outbreaks of E. coli O157:H7 occurring as a result of direct human contact with animals, feces, or manures at fairs, farms, and other public settings (Duffy 2003; Durso and others 2005; Keen and others 2006, 2007). Similar studies of sheep in the U.K., U.S., and Spain, representing thousands of samples, reported an incidence of E. coli O157 that ranged from 0.7 to 7.3%, and for domestic pigs incidence ranged from 0.3 to 8.9% (Table 1.3). In multiple studies of cattle feedlots and ranches, strains of E. coli O157:H7 persisted for up to 24 months at individual farms, and strains indistinguishable by molecular typing methods were isolated from farms separated by up to 50 km (Rice and others 1999; LeJeune and others 2004; Wetzel and LeJeune 2006). Indeed, a link between livestock and human illness with E. coli O157:H7 and other STEC has been supported by a direct correlation reported between the density of livestock and amount of reported illness in a region of Ontario, Canada (Michel and others 1999). Salmonella enterica Strains of S. enterica were isolated from 1.4 to 9% of beef cow fecal samples (Australia, U.S., U.K.) reported in four studies (Table 1.3). In a recent study of 7,680 animal and environmental samples from a single U.S. dairy, 13–72% of the cattle samples (depending upon period of testing), and >50% of air, soil, water, insect, and bird feces samples yielded S. enterica (Pangloli and others 2008). Similarly, high incidences of S. enterica in pigs were reported in a U.K. study (23.4%), in poultry flocks (10.5 to 13%) in U.S. and Belgium studies, and in poultry production environmental samples (12 to 51%) in a U.S. study (Table 1.3). S. enterica has been isolated from 1 to 7% of deer samples in two studies reported and up to 3% of wild bird samples. A multidrug-resistant S. Newport strain was prevalent on two different farms for months and shed by a cow for at least 190 days (Cobbold and others 2006), and, as noted above, a strain of SE (PT30) has been isolated from almond orchard soil periodically for at least 5 years (Uesugi and others 2007). Campylobacter Species C. jejuni incidence in cattle, poultry, other farm animals, and wild animals has been reported and reviewed (Miller and Mandrell 2006). Although the incidence of C. jejuni reported in >20 studies is comparable or higher than those reported and listed for E. coli O157 and Salmonella in Table 1.3, few major outbreaks of C. jejuni associated

20

Section I. Microbial Contamination of Fresh Produce

with fresh produce have occurred (Mandrell and Brandl 2004). In agreement perhaps, is the absence of any isolation/detection of C. jejuni on >6,800 produce samples in recent studies reported (Sagoo and others 2001; Thunberg and others 2002; Moore and others 2002; Sagoo and others 2003), suggesting that C. jejuni may be of lesser fitness compared to E. coli O157 and Salmonella in environments relevant to fresh produce production and preharvest contamination (Brandl and others 2004). Nevertheless, high numbers of sporadic C. jejuni illnesses compared to E. coli O157 and Salmonella (MMWR 2005b, 2007b) suggest surveillance to identify food sources associated with C. jejuni illness, including produce, should be continued. The results summarized in Table 1.3 confirm there are multiple livestock and wildlife sources of pathogens and suggest modes of transport of pathogens for contamination of fresh produce in fields or orchards. Livestock are located near produce production in many locations, but not close enough usually to be considered a major risk. However, resident wildlife species are potential sources of pathogens also, and commingle with livestock on ranches, dairies, or feedlots, thus increasing exposure of livestock and wildlife to pathogens. Wildlife colonized by pathogens will roam and potentially disseminate them to produce or other locations in the vicinity of produce (Jay and others 2007). This presents problems for controlling wildlife intrusion into fields depending upon the size and roaming capability of the species. Small mammals (e.g., squirrels, mice, raccoons), large mammals (feral swine, deer, elk), and birds illustrate the diversity of population sizes, barriers (fencing height, depth, gage), and habitat that are issues in considering interventions to control exposure of wildlife to fields. Therefore, only obvious risk factors can be addressed until definitive data are obtained about major sources of pathogen in an environment. A few conclusions can be drawn from the selected livestock and wildlife incidence data. First, they reflect the dynamic fluctuations in the incidence of enteric pathogens that can occur and that relatively high incidence of certain pathogens may occur at specific times. Second, there appears to be a general trend in higher incidence of S. enterica strains in surveys of animal and environmental samples compared to E. coli O157:H7, a trend consistent with the general amount of illness reported for these pathogens in the U.S. and U.K. (MMWR 2005b, 2007b; CDR 2006). In contrast, the recurrent outbreaks of E. coli O157:H7, in the absence of any known Salmonella outbreaks, associated with leafy vegetables grown in the same region (Table 1.1) is inconsistent with this trend. Perhaps, a study of the incidence of Salmonella in the environment of leafy vegetable production would provide clues to explain this paradox.

High-level Shedding of E. coli O157:H7 and Salmonella by Some Animals Measuring the prevalence of pathogens in animals and other environmental reservoirs relevant to produce production are informative, but the concentration and total amount of pathogen disseminated is perhaps more relevant to identifying potential risks in a produce production region. However, quantifying pathogen in complex samples is difficult due to the inability to survey livestock and wildlife populations comprehensively and to obtain accurate values with environmental samples containing low concentrations of pathogens in a complex microbial flora.

Enteric Human Pathogens Associated with Fresh Produce

21

Cattle shedding high levels of E. coli O157:H7 in their feces have been identified in some surveys. The majority of cows positive for E. coli O157:H7 in a herd have <100 CFU/g of feces, and this usually is detectable only by preenrichment and immunomagnetic selection methods. However, high-level shedders (“super shedders”) have been identified that shed between 1,000 and 1,000,000 CFU/g of feces (Low and others 2005; Chase-Topping and others 2007). Similarly, mice shedding >108 CFU viable Salmonella cells per gram of feces have been identified in laboratory studies, and high-shedding status appeared linked directly to the health of the intestinal microflora and level of inflammation in the colon (Lawley and others 2008). Indeed, models of prevalence, heterogenous shedding, and human infectious dose data are consistent with the “80/20 rule” suggesting that 80% of the transmission of an infectious agent results from the 20% of the most infectious members of the population (Matthews and others 2006). Therefore, colonized animals shedding large doses of a pathogenic strain (or strains) relative to the majority of a herd, or any population, in a region are relevant epidemiologically because the strains they shed are likely to be predominant in the environment. If predominant strains are virulent members of the species also, they are candidates for outbreaks of foodborne illness or other forms of infectious disease (Matthews and others 2006). Other factors important epidemiologically are the survival of a virulent pathogen in complex environments and its fitness in water, in soil, and on field crops. It is noteworthy then that E. coli O157:H7 strains linked to four outbreaks associated with bagged leafy vegetables in 2005 and 2006 (including the baby spinach outbreak, 2006) appear to be part of a phylogenetically distinct group (“clade 8”) that includes virulent strains associated with outbreaks from patients who had been hospitalized with hemolytic uremic syndrome and strains associated with increased frequency of hospitalization (Manning and others 2008). Increased virulence correlates also with a lower infectious dose required for illness. The estimates of the dose of E. coli O157:H7, for example, capable of causing illness in a population exposed to contaminated food ranges from 4 to <40 CFU/g of food (Strachan and others 2001; Teunis and others 2004). Thus, a more virulent strain capable of causing illness at an even lower infectious dose emphasizes the risks associated with any pathogen contamination of environments near produce production.

Incidence of Potential Pathogens in Municipal and Agricultural Watersheds Pathogens shed onto soil on the range, in feedlots, or in other habitats are dispersed and disseminated further by runoff into watersheds. Table 1.4 summarizes the results of some selected recent studies of the incidence and fitness of E. coli O157:H7 and S. enterica in municipal or agricultural watersheds because they have been the bacterial pathogens linked most frequently with recent outbreaks associated with preharvest contamination of fresh produce (Table 1.1). The incidence of E. coli O157:H7 in watersheds has been reported to be low generally (<2%) compared to Salmonella, reflecting probably the general concentration of the pathogens in the water samples. Strains of E. coli, potentially pathogenic based on the presence of known virulence genes (tir and stx), were isolated frequently in one U.S. study, indicating that specific urban watersheds can be contaminated heavily

22

Section I. Microbial Contamination of Fresh Produce

Table 1.4. Incidence of pathogenic E. coli and S. enterica in municipal and agricultural watersheds Description

Pathogen

Reference

Alberta, Canada; watershed near agriculture; 1999–2000; not associated with manure output

E. coli O157:H7 (13/1483 = 0.9%) Salmonella (88/1429 = 6.2%) E. coli, tir and/or Stx-pos (653/1218 = 53%) E. coli O157:H7 (38/584 = 6.5%)

Johnson and others 2003

Salmonella (574/? = ?)a E. coli O157:H7 (6/260 = 2.3%) E. coli O157:H7 (5/? = ?%) a

Baudart and others 2000

Salmonella (62/83 = 75%)

Meinersmann and others 2008

Baltimore, MD, area, U.S.; 2002–04; potential pathogens California, central coast, U.S.; agricultural and urban watershed; 2005–06 France, near Mediterranean; agricultural and urban; 1996–97 Central African Republic; N’Goila Cornwall, U.K.; freshwater stream crossing beach; 2004; outbreak-associated Georgia, U.S.; single day, 83 sites on river; 2005 a

Higgins and others 2005 Cooley and others 2007

Tuyet and others 2006 Ihekweazu and others 2006

Total number of samples tested was not noted.

with potential pathogens (Higgins and others 2005). However, the lack of any evidence of human illness associated with these strains suggests that they are not highly virulent compared to E. coli O157:H7. Four of the studies listed in Table 1.4 were initiated as a result of high levels of illness and outbreaks of E. coli O157:H7 associated with exposure to water or food (Johnson and others 2003; Ihekweazu and others 2006; Tuyet and others 2006; Cooley and others 2007). One of these studies of a watershed in a major leafy vegetable production region of the U.S. was initiated as a result of three separate outbreaks of E. coli O157:H7 linked to leafy vegetables grown in the Salinas Valley region of California (Table 1.1), and possibly linked to a single farm (Cooley and others 2007). Water samples obtained monthly on average from <20 sites, most within approximately 30 km of one another, revealed that >6% of the samples were positive for E. coli O157:H7. Sites nearby cattle grazing in elevated regions of the watershed were positive more frequently, and samples obtained during or subsequent to heavy rain events with increased water flow correlated with increased incidence at specific sites. Also, strains indistinguishable or highly related by genotype were isolated at the same time up to 30 km apart, or from the same sites months apart (Cooley and others 2007). Similarly, outbreak investigations of farms and ranches in the central coast region of California have provided clues to intriguing fate and transport relationships from assessment of genotypes of strains of E. coli O157:H7 isolated from environmental and wildlife samples obtained at similar times and locations (Cooley and others 2007; Jay and others 2007). Predominant strains may be persistent in some environments and transported by the commingling of wildlife and livestock into watersheds and possibly fields where produce is grown. However, the amounts of pathogen, their

Enteric Human Pathogens Associated with Fresh Produce

23

persistence in soil, and the processes involved in exposure of seedlings or mature plants to pathogen are unclear. In the absence of an effective “kill step” for postharvest produce, it remains important to identify sources of pathogens and their fate and transport in produce environments; this may assist in development of strategies for preventing contamination of produce destined for the ready-to-eat market.

Fate and Transport of Human Pathogens in the Environment It has been difficult to determine the primary source of preharvest produce contamination; however, nearby livestock, poultry, or other farm animals are obvious potential point sources for further dissemination in the environment, and linked possibly to produce (Table 1.3). Potential mechanisms for dissemination of pathogens from contained farms or feedlots are movement of livestock to new locations, wildlife intrusion, water runoff/flooding (Table 1.4), dust/bioaerosols, manure/compost/compost-tea fertilizers, and possibly other intra- and interfarm human activities (farm vehicles and equipment). For pathogens to be transported outside an animal host, they must remain fit enough to survive (and possibly grow) until they encounter an environment favorable for growth. Findings from previous studies measuring the survival of pathogenic E. coli and Salmonella in manure, soil, and water are relevant to hypotheses about how preharvest contamination occurs. Table 1.5 is a list of selected studies that provide a comparison of measured fitness characteristics of E. coli O157, E. coli O157:H7, and Salmonella in environments relevant to fresh produce contamination, including manure, soil, manure-amended soil, and water. It is worth noting that some of these studies report the incidence of pathogens in their natural state in relevant environmental samples, whereas others involved spiking samples with marked strains and then monitoring their incidence over a period of time. Each study listed in Table 1.5 involved different locations and experimental conditions; however, it is noteworthy that outcomes generally were consistent. For example, in nearly all studies, E. coli O157 or E. coli O157:H7 remains detectable in some samples for >30 days, but longer than 6 months in other samples (Table 1.5; cow water trough, sheep manure, manure-amended soil). Salmonella cells were detectable for similar periods of time (e.g., soil, manure-amended soil), but an outbreak strain was detectable for >1500 days in soil samples from an almond orchard linked to the outbreak (see below). Similarly, multiple strains of E. coli O157 were isolated for months from biofilms on flint shingles immersed in stream beds exposed to runoff from farm animals positive for the pathogen (Cooper and others 2007). These studies support the persistence theory and possible mechanisms of periodic reintroduction of pathogens in agricultural environments. Conversely, a recent study of potential pathogens isolated from livestock and then inoculated onto spinach and lettuce in field plots reported rapid die-off of a shigatoxin-negative strain of E. coli O157:H7; this was in contrast to the survival of a strain of S. Enteritidis for at least 14 days (Hutchison and others 2008). These contrasting results emphasize again the variability of pathogen survival in complex environments, and the dependence of results probably upon pathogen fitness, experimental design (field versus microcosm), and other factors (spatial, temporal, indigenous flora, disease, etc.), any of which might

Table 1.5. Selected studies of the fitness of E. coli O157, E. coli O157:H7, and Salmonella in environmental samples or microcosms Pathogen

Environment

E. coli O157:H7

Water, 8 °C Water, 25 °C Water trough, sediment Water, <15 °C Water + feces, <15 °C Water, biofilms Water: lake, river, drinking trough microcosms Soil Soil, manure-amended (child illness) Soil, manure-amended

E. coli O157 E. coli O157:H7 E. coli O157 E. coli O157:H7

E. coli O157 E. coli O157 E. coli O157:H7 E. coli O157:H7 (Stx-neg) E. coli O157:H7 E. coli O157:H7 E. coli O157 (Stx-neg) E. coli O157:H7 E. coli O157:H7 E. coli O157:H7 (Stx-neg)

E. coli O157:H7 (Stx-neg) S. enterica S. enterica S. enterica S. Newport S. Enteritidis S. enterica S. Enteritidis a

Maximum Survival (Days) >91 <84 245 14 24 >30 6 to >60 Lake > river

Reference

105 69

Wang and Doyle 1998 LeJeune and others 2001 McGee and others 2002 Cooper and others 2007 Avery and others 2008

>35

Ogden and others 2002 Mukherjee and others 2006 Williams and others 2007

Soil, 36 types

54–105

Franz and others 2008

Soil, cover crops Manure, cow Manure, sheep Feces, cow Water Feces, cow, turned Feces, cow, unturned Manure, cow Manure, slurry Soil, manure-amended Lettuce Parsley Onions Carrots Lettuce and spinach

40–96 47 >600 97 109 42 90 21 35 154–217 77 177 74 168 <7

Gagliardi and Karns 2002 Kudva and others 1998

Water, river

>45

Soil, chicken farm Soil Soil, manure-amended Manure, cow Soil, almond orchard Soil, tomato crop debris (microcosm) a Lettuce and spinach

240 >120 107–332 49–184 >1500

Santo Domingo and others 2000 Davies and Breslin 2003 Holley and others 2006 You and others 2006

Scott and others 2006 Fremaux and others 2007 Himathongkham and others 1999 Islam and others 2004, 2005

Hutchison and others 2008

Uesugi and others 2007

56

Barak and Liang 2008

>14 to <21

Hutchison and others 2008

Some soils included crop debris from tomato plants infected with the pathogen Xanthomanas campestris and colonized with S. enterica.

24

Enteric Human Pathogens Associated with Fresh Produce

25

in some combination be conducive to pathogen survival, growth, and, in some instances, increased virulence in a leafy vegetable–associated outbreak (Table 1.1). These results reflect a “snapshot” of the pathogen under the selected test or environmental conditions, in addition to a spectrum of fitness characteristics of the pathogen assessed. Two studies relevant to concepts of persistence of specific pathogen strains in a preharvest environment and direct links to human illness are worth noting. A survey of a family garden subsequent to the O157:H7 illness of a child playing in the raw manure–amended garden revealed that strains indistinguishable from the child’s strain were detectable in soil samples from the garden for >69 days, and that incidence was much higher in soil sampled during ambient temperatures compared to 4 °C (Mukherjee and others 2006). Similarly, strains of S. Enteritidis Phage Type 30 associated with at least one outbreak linked to raw almonds, and possibly a second (Table 1.1), were isolated over at least a 5-year period from soil drag swab samples obtained in an orchard linked to the outbreak (Uesugi and others 2007). The Salmonella strain, indistinguishable from outbreak strains, was isolated from soil more frequently during and after harvests (average 20–42% of samples, Aug–Dec), and in >50% of soil samples following a heavy rain event. Although the virulence and infectiousness of an environmental pathogen strain cannot be compared to related human clinical strains, the sets of E. coli O157:H7 and S. Enteritidis PT 30 environmental strains noted above are closely related epidemiologically to the corresponding clinical strains. It can be speculated that persistence of these pathogen strains in the garden and orchard environments may relate directly to the evolution of fitness characteristics that correlate also with virulence (Manning and others 2008). Manure-amended soil, plants and plant debris appear to be beneficial to the survival of E. coli O157:H7 and Salmonella (Table 1.5). Ruminant-digested grasses and feeds and crop debris have nutrients supporting survival and possibly growth of enteric pathogens under the appropriate environmental conditions, including temperature, moisture, and atmosphere (Brandl 2006). For example, E. coli cells present naturally in cow feces placed in shaded and nonshaded fields increased 1.5 log after 6 to 8 days, declining fast in nonshaded fecal samples and then rebounding >1 log in nonshaded samples after rain events (Van Kessel and others 2007). In contrast, E. coli in air-dried sandy and silty soils amended with municipal sludge (biosolids) declined more slowly than in moist soils; up to 3 log differences were noted after 35 compared to 91 days in the field (Lang and Smith 2007). These studies are monitoring generic rather than pathogenic E. coli; however, the results are informative about different feces (cow, human), exposure to sun (UV) or moisture, and rates of resuscitation in rain— important environmental factors affecting pathogens in the environment. E. coli O157 and S. enterica, and generic E. coli as fecal indicator bacteria, appear capable of surviving months or even years under the appropriate environmental conditions and, under optimal conditions, they grow 1 to 3 logs (Table 1.5). Indeed, in a recent study of Salmonella in tomato crop debris, it appears this may be another aspect of the preharvest environment worth considering as a site conducive to survival or growth of pathogen for extended periods of time (Barak and Liang 2008). Tomato seeds planted in soil with Salmonella-contaminated tomato crop debris resulted in plants contaminated with Salmonella in the rhizoplane > phyllosphere. Salmonella survived

26

Section I. Microbial Contamination of Fresh Produce

well in the tomato phyllosphere of plants from seeds inoculated with the tomato plant pathogen, Xanthomonas campestris pv. vesicatoria and planted in low Salmonella inoculum soil, indicating the potential importance of debris, plant disease, and fallow periods in the preharvest produce production cycle (Barak and Liang 2008). Thus, breakdown of tomato crop debris by plant pathogens may enhance the conditions for even better survival or growth of a human pathogen (Barak and Liang 2008; Brandl 2006; Brandl and Amundson 2008). Pathogen reservoirs where tenfold or more growth of pathogen may occur are critical risk factors relevant to food contamination. Highshedding animals; manure; crop and/or ground cover debris; and produce plant seedlings, leaves, and roots are candidate sites for amplification. Unidentified reservoirs of amplification, such as wild animals, microorganisms, and plants, may exist also.

Source-Tracking Pathogens and Fecal Indicators of Contamination in Watersheds The epidemiology of major produce-associated outbreaks occurring in the last decade has revealed that preharvest contamination occurs (Table 1.1). However, surveys of fresh produce at different stages in the production and processing cycle indicate that bacterial pathogens are at low incidence generally (Beuchat 1996; Harris and others 2003; Nguyen-the and Carlin 1994, 2000), even though fecal indicator bacteria (E. coli) present appear to increase in prevalence during transport and distribution (Table 1.2) to wholesale and retail markets (Valentin-Bon and others 2008). Therefore, specific events following preharvest contamination are important to identify also since they may provide clues to amplification sites resulting in a high incidence or concentration. An important stage in preharvest contamination is movement onto fields, and more importantly, onto or into seedlings or the mature plants. Water (Table 1.4; irrigation, flooding), intrusion by animals either directly (Table 1.3; wildlife, domestic, humans) or indirectly (fertilizer, compost), and dust are potential mechanisms of contamination. Water quality is a primary factor in production of safe fresh produce, and irrigation water comes from a variety of sources dependent upon the type of produce and location. The majority of leafy vegetable production in the region of the U.S. implicated in outbreaks involves irrigation with well water of high quality relative to surface water that may be nearby. Indeed, well water was reported to be the source of irrigation of leafy vegetables associated with recent outbreaks (CalFERT 2007b, 2008). It is noteworthy also that U.S. winter produce production occurs mainly in the Imperial Valley of California and the Yuma region of Arizona, where irrigation water is sourced often from surface water. In contrast, outbreaks associated with produce from these locations have not occurred or have been rare (Table 1.1). Obviously, the quality of water in lakes, ponds, reservoirs, and watersheds is critical to produce production even when it is not used directly for irrigation. Surface water could be a major source of pathogens affecting aquifer recharging, exposure of animals to colonization, and/or transport to produce fields by irrigation, or processes as yet unidentified. Watersheds are impaired by the presence of fecal bacteria from livestock, wildlife, and humans. Any fecal contamination increases the probability of enteric pathogen contamination of produce either directly or indirectly. The level of impairment is

Enteric Human Pathogens Associated with Fresh Produce

27

dependent upon many factors related to the geography and ecology within and surrounding the watershed, including the density of animals, hydrology, elevation/runoff, meteorological conditions (e.g., rainfall and temperature), pathogen fitness (Table 1.5), water composition (salinity, nutrients), predation, and vegetation. Waterborne disease outbreaks in the U.S. (1948–1994) and Canada (1975–2001) occur more frequently following heavy rain events, indicating transport of pathogens from human, domestic animal, livestock, or wildlife sources through runoff, and, ultimately, contamination of drinking water supplies (Curriero and others 2001; Thomas and others 2006). Although no definitive links between heavy rain events and human illness have been reported, flood contamination of fields or irrigation water sources intended for growing produce is a potential risk factor for illness (CDHS 2005). Watershed hydrology may be crucial to understanding pathogen transport within an environment. Hydrological processes are relevant to transport of pathogens in the environment, including fecal disintegration and dispersion, resuscitation of pathogens in arid environments, trapping of pathogens in wetlands, concentration of pathogens on or in sediment particles, land-to-watershed-to-land movement, and exposure of wildlife to pathogens (Ferguson and others 2003). Similarly, the soil and sediment particles present in flowing or static water bodies can interact and bind with microorganisms by mechanisms that are not well defined, and likely vary depending upon variations in soil, fecal and water composition, weather, and other factors (Gagliardi and Karns 2000; Brookes and others 2004; Ferguson and others 2003). Transport of pathogens in dust, on harvest equipment, in manure/compost and pesticide and herbicide sprays diluted with surface water should be considered also. Pathogens and microbial species as indicators of fecal contamination can be prevalent in environments near produce production (Tables 1.3 and 1.4). Sensitive and accurate detection of specific pathogens in the environment to track the fate and transport of pathogens to fields requires intensive sampling, successful isolation of pathogens or fecal indicator microorganisms, and efficient molecular genotyping methods for microbial source tracking pathogens in relevant and complex environments (Field and Samadpour 2007; Meays and others 2004). A variety of different source tracking methods have been developed to identify sources of fecal contamination, sometimes yielding mixed results and accuracy (Field and Samadpour 2007; Stoeckel and others 2004). Microbial source tracking methods have evolved to include modern genetic methods that involve fingerprinting isolates from the environment and different animal hosts to create a database for comparing fingerprints of new strains to those in the database and thus identify putative sources of fecal contamination (Field and Samadpour 2007). Pulsed field gel electrophoresis (PFGE) remains a common method for fingerprinting foodborne pathogens, mainly because of CDC’s PulseNet database, which stores PFGE profiles submitted by public health labs representing tens of thousands of sporadic and outbreak strains for comparison (Swaminathan and others 2001). However, sequence-based typing methods, such as MultiLocus Variable number tandem repeat Analysis (MLVA), MultiLocus Sequence Typing (MLST), and Single Nucleotide Polymorphism (SNP) microarrays, are gaining in acceptance due to ease of use, speed, and high-resolution data for comparisons. MLVA is an effective method for genotyping E. coli O157:H7 (Hyytia-Trees and others 2006) and is being evaluated also for S. Enteritidis. MLVA proved effective in

28

Section I. Microbial Contamination of Fresh Produce

environmental studies involving tracking E. coli O157:H7 strains in produce production environments, watersheds, and cattle feedlots (Cooley and others 2007; Murphy and others 2008). An intriguing finding in the 2006 investigation of the E. coli O157:H7 multistate outbreak linked to bagged baby spinach was the isolation of multiple strains of E. coli O157:H7 from the feces of multiple feral swine trapped in the vicinity of the suspected spinach field; some of these isolates, and isolates from cow fecal, river. and dirt samples also collected within a mile of the field, were indistinguishable from the clinical outbreak strains (Jay and others 2007; Cooley and others 2007). Similarly, evidence of transport of E. coli O157:H7 strains between dairy farms by wild birds has been reported (Wetzel and LeJeune 2006).

How Do Pathogens Get onto Preharvest Produce and Survive? Hypotheses from Recent Outbreaks The transient incidence of pathogens in livestock, wildlife (Table 1.3), and watersheds (Table 1.4), the environmental fitness characteristics of foodborne pathogens (Table 1.5), and recurring outbreaks of foodborne illness associated with ready-to-eat produce (Table 1.1) are consistent with the findings of low-level, but significant, incidence of generic E. coli on fresh produce obtained from distribution centers and retail markets (Table 1.2). Although some of this E. coli could be present at harvest, postharvest contamination also could occur in a variety of ways, such as rodents, contaminated bins or transport vehicles, commingling of food at retail markets or restaurants, or ill workers. Postharvest cross-contamination could exacerbate what might have been a limited contamination event initially. Preharvest contamination of produce occurs by obvious processes, but perhaps also by unknown, or less well understood, processes. Although no definitive conclusions have been offered about the sources of preharvest contamination of leafy vegetables and tomatoes associated with recent outbreaks (Table 1.1), reasonable hypotheses involve transport of pathogen in animal fecal waste by 1) watershed to flooded fields (CDHS 2005), 2) feral swine intrusion (Jay and others 2007), 3) irrigation by pipes used previously to remove dairy holding pond waste (CalFERT 2008), and 4) amphibian or other wild animals emerging from contaminated surface water to intrude into fields (MMWR 2005a). Water is a central factor in hypotheses of contamination, so studies of the dispersion and dissemination of microbes in water and the use of microbes as tracers of water movement are relevant to understanding dissemination of enteric pathogens in water. Heavy rainfall is associated with rapid dispersal of pathogens from fecal matter on the ground into surface and groundwater (Ferguson and others 2003). Pathogen incidence and survival in feces, water, soil, and other matrices (Table 1.3, 1.4, 1.5) are relevant for modeling environmental contamination of preharvest produce, identifying sources, and controlling contamination, but details are lacking about how different species of bacteria, including pathogens, disperse and survive in water and other sites in the production environment and how this might relate to preharvest contamination. Bacteria, yeasts, and bacteriophage have been used as tracers by dosing a large number of laboratory-grown cells (approximately 1014 cells) into a river and monitoring movement (Wimpenny and others 1972). The bacterial strain traced, S. marcescens

Enteric Human Pathogens Associated with Fresh Produce

29

(distinctive red colonies), for example, moved in the river at approximately 2.5 km/hr over the 2.9 km between the dosing and detection points. The dosed strain was detected at a maximum of 500 cells/ml, which reflected a significant dilution (>1.7 × 108-fold) of the bacteria during transport (Wimpenny and others 1972). To achieve a comparable amount of E. coli O157:H7 from “high-shedder” cattle feces (e.g., 106 cells/g), for example, would require >200,000 kg of feces. In a separate study in an elevated region within miles of leafy vegetable production, transport of E. coli O157:H7 strains was tracked from a point source (small corral with a few head of cattle) into a small stream (Cooley and others 2007). Indistinguishable or related pathogen strains identified by MLVA genotyping were isolated at the point source and up to 135 m downstream (3 m lower altitude) from the point source. However, water flow was relatively low prior to and at the time of sampling (Cooley and others 2007). Isolation and/or detection of pathogens in water at distant sites from a suspected point source, therefore, might involve one or more of the following: large volumes of feces and/or high-shedding animals, very sensitive detection of few pathogen cells, multiple point sources with related strains, or transport mechanisms (e.g., cell-cell or cell-particulate aggregates, mats, flotation) different than those reflected by laboratory cultured microorganisms in tracer studies. Accurate tracer studies of pathogens in the environment would be advantageous for understanding fate and transport mechanisms relevant to produce contamination. Pathogens in animal feces deposited on rangeland, feedlots, or dairy alleys, and into storage ponds are exposed to dispersion, transport, and inactivation that could be affected by soil and fecal matrices, particle sizes, buoyancy, microbial competitors/ predators or cooperators, and even climate (rainfall, temperature, UV exposure). It is noteworthy that during the 2006 outbreak of E. coli O157:H7 associated with bagged baby spinach, unusually high daily temperatures occurred at the time of planting: July 22–25, 2006: max. daily 100–110 °F (37.7–43.3 °C); ave. daily 77–85 °F (25–29.4 °C), and approximately 5–6 days prior to harvest (CalFERT 2007b,c). This unusual condition stimulates questions regarding when contamination occurred in the crop cycle and whether high temperatures may have enhanced survival or growth of pathogen in the preharvest environment. For example, E. coli O157 has been shown to survive and increase in number with increasing temperature (10–30 °C) in natural freshwater microcosms containing low concentrations of organic carbon (Vital and others 2008). The direct correlation between pathogen growth and water temperature is consistent with enteric bacteria that have evolved to grow optimally at body temperatures. Survival of Human Pathogens on Preharvest Plants Outbreaks associated with preharvest contaminated produce confirm that enteric bacteria are capable of attaching somewhere on the plants and remaining viable (Tables 1.1 and 1.3). Field studies with nonpathogenic varieties of E. coli O157:H7 and other pathogens on plants under field conditions confirm that they can survive for weeks and months depending upon the amount of bacteria applied and the treatment conditions (Tables 1.2 and 1.3). Laboratory studies indicate that E. coli O157:H7 and Salmonella applied to a variety of plant roots, leaves, and seeds can attach tenaciously (resisting sanitization) and survive, but also in some instances grow when conditions

30

Section I. Microbial Contamination of Fresh Produce

are ideal for a pathogen (warm temperature, high humidity, adequate nutrients) (Brandl 2006). Sophisticated fluorescence microscopy experiments have revealed specific locations on leaves and roots where subcuticlar cells, root hairs, or breaks in the tissue (e.g., lateral root formation) provide sites and nutrients for harboring opportunistic pathogen cells. Aggregation of enteric pathogen cells with one another and with plant epiphytic or plant pathogen microflora suggest that active and complex interactions may occur on plants in the field, resulting possibly in interactions/contamination very difficult to remove by normal washing or sanitizing methods (Brandl 2006). In addition, there appears to be emerging support for the hypothesis that some human pathogen cells on plants may become internalized through different routes of entry on roots, shoots, and flowers (Guo and others 2001; Solomon and others 2002; Warriner and others 2003; Dong and others 2003; Franz and others 2007; Doyle and Erickson 2008; Schikora and others 2008). Indeed, recent reports examining the plant response to potential human pathogens in model plant systems (Arabidopsis thaliana mutants and gene expression arrays) indicate that genes and gene pathways are upregulated similarly to plant resistance responses to plant pathogens (Dong and others 2003; Thilmony and others 2006; Schikora and others 2008). Thus, the potential for some human pathogens to be endopathogenic for some plant hosts in a preharvest environment raises obvious concerns regarding postharvest treatments for decontamination. Reviews of different mechanisms that plant epiphytes and pathogens and human enteric pathogens use to attach to plants (Mandrell and others 2006; Solomon and others 2006) and an excellent review of the general biology, ecology, and fitness characteristics of human enteric pathogens on plants have been published previously (Brandl 2006). Further details about the molecular interactions that can occur between bacterial human pathogens (e.g., flagellin, fimbriae, pili, curli, outer membrane proteins) and plants (generally undefined), and the microbial ecology on plants that may enhance or control pathogen survival are provided in these reviews and also chapters elsewhere in this book.

Conclusions The increased incidence of produce-related outbreaks tracked to specific regions, and E. coli O157:H7 outbreaks in particular, has stimulated questions about what might have changed over the last decade to explain this increase. Is it related to growing (fertilization, water, shallow tilling, seeds, cultivars) or production practices (cutting, transport, bagging, atmosphere), changes in the pathogens (increased fitness in animals, water), livestock (transport, incidence of pathogens), or better detection (methods, public health system, media)? Clearly, some of these questions raise issues that would be considered higher risk factors than others and worthy of prioritizing for research. Most people can appreciate that animals or feces on or near fresh produce fields are major potential risk factors, probably worthy of attempts to prevent continued intrusion. Lacking convincing evidence of pathogen carriage by a suspect animal species, however, becomes problematic for making informed decisions about mitigation approaches (predation, fencing, testing). Indeed, lack of definitive proof of sources of pathogens has created a significant conflict between conservationists, environmentalists, and growers on one side versus those in the produce industry responsible for

Enteric Human Pathogens Associated with Fresh Produce

31

addressing preharvest produce food safety issues. The conflict involves a contradiction between creation of vegetative zones for filtering runoff from fields and wildlife habitat, and the perceived risks of attracting to this habitat wildlife colonized possibly with pathogens (Berreti and Stuart 2008). Some compromise between these competing interests will be necessary for sustaining the valuable locations where produce is grown and improving the quality and safety of produce. As noted above, a convergence of multiple events probably is required to cause a major outbreak, implying that each event alone may be insufficient. The changes in pathogen incidence and virulence in a preharvest food production environment can be speculated to be associated with corresponding and dynamic changes in the biology, ecology, hydrology, meteorology, and agricultural practices in an environment. Considering the impossibility of controlling certain aspects of the ready-to-eat produce production environment, it is logical to assume that additional outbreaks will occur. Intensive practices leading to exposure of pathogens to complex environments, or significant replication of microorganisms, will increase the rates of new mutations and fitness in environments where mutations are beneficial. Modern molecular biology techniques (genomics) are facilitating the fingerprinting of outbreak-related pathogen strains for purposes of high-resolution tracking of the possible sources of contamination in preharvest environments. Also, comparative genomics of these data reveal insights about pathogen evolution and emergence of virulence-related factors that raise questions about whether produce outbreak-related pathogens are more virulent and have special fitness characteristics (Zhang and others 2006; Manning and others 2008). The rapid changes possible in bacterial genomes by mutations, phage insertions and deletions, and recombination, as examples, predict the emergence from high-intensity environments (food production) of organisms with selected fitness characteristics that reflect the environment. If some of these fitness characteristics are virulence traits in humans (i.e., pathogens), pathogens will be identified through studies of human illness. Considering the known potential risk factors in the preharvest environment documented above, some approaches for preventing contamination of food can be offered. Common sense approaches include maintaining water quality and minimizing exposure of fields to wild animals, surface water (flooding), and dust from agricultural activity. Other less obvious approaches requiring more resources are identifying highshedding livestock or wildlife, treatment of livestock with effective vaccines or other antimicrobials, checking and maintaining feed quality, observing field conditions (wildlife intrusions), redirecting or destroying suspect produce, and controlling wild animal habitat. Postharvest approaches involve sample testing (test and hold), clean water, novel sanitizers (chemical or biological), and irradiation, to name a few. More details regarding interventions will be discussed in other sections of this book. Finally, it should be noted again that the incidence of illness linked to contaminated produce is quite low relative to the total number of produce consumptions. Nevertheless, the increased incidence of outbreaks and the apparent hypervirulence of pathogen strains associated with some of these outbreaks (Manning and others 2008), emphasize that continued vigilance is necessary to minimize the severity of any outbreaks that might occur. Until a highly effective and nontoxic “kill step” is developed for eliminating pathogens from postharvest fresh produce, pathogens in the preharvest environment deserve our serious attention and continuing research efforts.

32

Section I. Microbial Contamination of Fresh Produce

Acknowledgments The author thanks representatives of the USDA Agricultural Marketing Service for providing data collected for the “Microbial Data Program,” M. Jay-Russell for source information regarding E. coli O157:H7 leafy vegetable outbreaks, and his colleagues and collaborators in USDA Cooperative State Research, Education, and Extension Service (CSREES), Epidemiological Approaches to Food Safety Program, projects 2006-01240 and 2007-02029.

References Abong’o BO, Momba MNB and Mwambakana JN. 2008. Prevalence and antimicrobial susceptibility of Escherichia coli O157:H7 in vegetables sold in the Amathole District, Eastern Cape Province of South Africa. J Food Prot 71(4):816–819. Ackers ML, Mahon BE, Leahy E, Goode B, Damrow T, Hayes PS, Bibb WF, Rice DH, Barrett TJ, Hutwagner L, Griffin PM and Slutsker L. 1998. An outbreak of Escherichia coli O157:H7 infections associated with leaf lettuce consumption. J Infect Dis 177(6):1588–1593. Arthur L, Jones S, Fabri M and Odumeru J. 2007a. Microbial survey of selected Ontario-grown fresh fruits and vegetables. J Food Prot 70(12):2864–2867. Arthur TM, Bosilevac JM, Nou X, Shackelford SD, Wheeler TL and Koohmaraie M. 2007b. Comparison of the molecular genotypes of Escherichia coli O157:H7 from the hides of beef cattle in different regions of North America. J Food Prot 70(7):1622–1626. Avery LM, Williams AP, Killham K and Jones DL. 2008. Survival of Escherichia coli O157:H7 in waters from lakes, rivers, puddles and animal-drinking troughs. Sci Total Environ 389(2–3):378–385. Bailey JS, Stern NJ, Fedorka-Cray P, Craven SE, Cox NA, Cosby DE, Ladely S and Musgrove MT. 2001. Sources and movement of Salmonella through integrated poultry operations: a multistate epidemiological investigation. J Food Prot 64(11):1690–1697. Barak JD and Liang AS. 2008. Role of soil, crop debris, and a plant pathogen in Salmonella enterica contamination of tomato plants. PLoS ONE 3(2):e1657. Barkocy-Gallagher GA, Arthur TM, Rivera-Betancourt M, Nou X, Shackelford SD, Wheeler TL and Koohmaraie M. 2003. Seasonal prevalence of Shiga toxin-producing Escherichia coli, including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. J Food Prot 66(11):1978–1986. Baudart J, Lemarchand K, Brisabois A and Lebaron P. 2000. Diversity of Salmonella strains isolated from the aquatic environment as determined by serotyping and amplification of the ribosomal DNA spacer regions. Appl Environ Microbiol 66(4):1544–1552. Berreti M and Stuart D. 2008. Food safety and environmental quality impose conflicting demands on Central Coast growers. Calif Agric 62:68–73. Beuchat LR. 1996. Pathogenic microorganisms associated with fresh produce. J Food Prot 59(2):204–216. ———. 2006. Vectors and conditions for preharvest contamination of fruits and vegetables with pathogens capable of causing enteric diseases. Brit Food J 108(1):38–53. Brandl MT. 2006. Fitness of human enteric pathogens on plants and implications for food safety. Annu Rev Phytopathol 44:367–392. Brandl MT and Amundson R. 2008. Leaf age as a risk factor in the contamination of lettuce with Escherichia coli O157:H7 and Salmonella enterica. Appl Environ Microbiol 74(8):2298–2306. Brandl MT, Haxo AF, Bates AH and Mandrell RE. 2004. Comparison of survival of Campylobacter jejuni in the phyllosphere with that in the rhizosphere of spinach and radish plants. Appl Environ Microbiol 70(2):1182–1189. Branham LA, Carr MA, Scott CB and Callaway TR. 2005. E. coli O157 and Salmonella spp. in white-tailed deer and livestock. Curr Issues Intest Microbiol 6(2):25–29. Breuer T, Benkel DH, Shapiro RL, Hall WN, Winnett MM, Linn MJ, Neimann J, Barrett TJ, Dietrich S, Downes FP, Toney DM, Pearson JL, Rolka H, Slutsker L and Griffin PM. 2001. A multistate outbreak of Escherichia coli O157:H7 infections linked to alfalfa sprouts grown from contaminated seeds. Emerg Infect Dis 7(6):977–982.

Enteric Human Pathogens Associated with Fresh Produce

33

Brookes JD, Antenucci J, Hipsey M, Burch MD, Ashbolt NJ and Ferguson C. 2004. Fate and transport of pathogens in lakes and reservoirs. Environ Int 30(5):741–759. CalFERT. 2007a. Environmental investigation of Escherichia coli O157: H7 outbreak associated with Taco Bell restaurants in Northeastern United States. California Food Emergency Response Team. Sacramento, CA. http://www.dhs.ca.gov/fdb/local/PDF/Taco_Bell_final_report_redacted_11_19_2007.pdf. ———. 2007b. Investigation of an Escherichia coli O157:H7 outbreak associated with Dole pre-packaged spinach. California Food Emergency Response Team. Sacramento, CA. http://www.dhs.ca.gov/ps/fdb/ HTML/Food/EnvInvRpt.htm. ———. 2007c. Investigation of an Escherichia coli O157:H7 outbreak associated with Dole pre-packaged spinach. Attachment 10: Environmental samples from farms and watersheds. California Food Emergency Response Team. Sacramento, CA. http://www.dhs.ca.gov/ps/fdb/HTML/Food/EnvInvRpt.htm. ———. 2008. Investigation of the Taco John’s Escherichia coli O157:H7 outbreak associated with iceberg lettuce. California Food Emergency Response Team. Sacramento, CA. http://www.dhs.ca.gov/fdb/ HTML/Food/EnvInvRpt.htm. Campbell JV, Mohle-Boetani J, Reporter R, Abbott S, Farrar J, Brandl M, Mandrell R, and Werner SB. 2001. An outbreak of Salmonella serotype Thompson associated with fresh cilantro. J Infect Dis 183(6):984–987. Castillo A, Mercado I, Lucia LM, Martinez-Ruiz Y, Ponce de Leon J, Murano EA and Acuff GR. 2004. Salmonella contamination during production of cantaloupe: a binational study. J Food Prot 67(4):713–720. CDC. Outbreak Surveillance Data, Annual Listing of Foodborne Disease Outbreaks, 1995 Foodborne Disease Outbreak Line Listing. January 4. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm. Accessed February 25, 2008. ———. Outbreak Surveillance Data, Annual Listing of Foodborne Disease Outbreaks, 1996 Foodborne Disease Outbreak Line Listing. January 4. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm. Accessed February 25, 2008. ———. Outbreak Surveillance Data, Annual Listing of Foodborne Disease Outbreaks, 1998 Foodborne Disease Outbreak Line Listing. January 4. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm. Accessed February 25, 2008. ———. Outbreak Surveillance Data, Annual Listing of Foodborne Disease Outbreaks, 1999 Foodborne Disease Outbreak Line Listing. January 4. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm. Accessed February 25, 2008. ———. Outbreak Surveillance Data, Annual Listing of Foodborne Disease Outbreaks, 2002 Foodborne Disease Outbreak Line Listing. January 4. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm. Accessed February 25, 2008. ———. Outbreak Surveillance Data, Annual Listing of Foodborne Disease Outbreaks, 2003 Foodborne Disease Outbreak Line Listing. January 4. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm. Accessed February 25, 2008. ———. Outbreak Surveillance Data, Annual Listing of Foodborne Disease Outbreaks, 2004 Foodborne Disease Outbreak Line Listing. January 4. Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm. Accessed February 25, 2008. ———. 2008a. Investigation of outbreak of infections caused by Salmonella Litchfield. March 22. Centers for Disease Control and Prevention. http://www.cdc.gov/salmonella/litchfield/. Accessed. ———. 2008b. Investigation of outbreak of infections caused by Salmonella Saintpaul. June 2. Centers for Disease Control and Prevention. http://www.cdc.gov/salmonella/saintpaul/. Accessed June 4, 2008. CDHS. 2004a. Investigation of E. coli O157:H7 Illnesses in San Diego and Orange Counties California Department of Health Services. Sacramento, CA. http://www.dhs.ca.gov/ps/fdb/HTML/Food/EnvInvRpt.htm. ———. 2004b. Investigation of E. coli O157:H7 Outbreak in San Mateo Retirement Facility. California Department of Health Services. Sacramento, CA. http://www.dhs.ca.gov/ps/fdb/HTML/Food/EnvInvRpt.htm. ———. 2005. Addendum Report to Investigation of Pre-washed Mixed Bagged Salad following an Outbreak of Escherichia coli O157:H7 in San Diego and Orange County. California Department of Health Services. Sacramento, CA. http://www.dhs.ca.gov/ps/fdb/HTML/Food/EnvInvRpt.htm. CDR. 2006. Enteric illness in the United Kingdom, 2005. Health Protection Agency. London. http://www. hpa.org.uk/CDR/archives/2006/cdr1506.pdf. Chapman PA, Siddons CA, Gerdan Malo AT and Harkin MA. 1997. A 1-year study of Escherichia coli O157 in cattle, sheep, pigs and poultry. Epidemiol Infect 119(2):245–250.

34

Section I. Microbial Contamination of Fresh Produce

Chase-Topping ME, McKendrick IJ, Pearce MC, MacDonald P, Matthews L, Halliday J, Allison L, Fenlon D, Low JC, Gunn G and Woolhouse ME. 2007. Risk factors for the presence of high-level shedders of Escherichia coli O157 on Scottish farms. J Clin Microbiol 45(5):1594–1603. Cizek A, Alexa P, Literak I, Hamrik J, Novak P and Smola J. 1999. Shiga toxin-producing Escherichia coli O157 in feedlot cattle and Norwegian rats from a large-scale farm. Lett Appl Microbiol 28(6): 435–439. Clemens R. 2004. The expanding U.S. market for fresh produce. Iowa Ag Rev 10:Winter. http://www.card. iastate.edu/iowa_ag_review/winter_04/article4.aspx. Cobbold RN, Rice DH, Davis MA, Besser TE and Hancock DD. 2006. Long-term persistence of multidrug-resistant Salmonella enterica serovar Newport in two dairy herds. J Am Vet Med Assoc 228(4):585–591. Cooley M, Carychao D, Crawford-Miksza L, Jay MT, Myers C, Rose C, Keys C, Farrar J and Mandrell RE. 2007. Incidence and tracking of Escherichia coli O157:H7 in a major produce production region in California. PLoS ONE 2(11):e1159. Cooper IR, Taylor HD and Hanlon GW. 2007. Virulence traits associated with verocytotoxigenic Escherichia coli O157 recovered from freshwater biofilms. J Appl Microbiol 102(5):1293–1299. CSPI. 2007. Outbreak Alert: Closing the gaps in our federal food safety net. Center for Science in the Public Interest, Washington, D.C. http://www.cspinet.org/foodsafety/outbreak_alert.pdf. Accessed March 3, 2008. ———. 2008. Outbreak Alert! Database, 1990–2005. Center for Science in the Public Interest, Washington, D.C. http://www.cspinet.org/foodsafety/outbreak/pathogen.php. Accessed March 22, 2008. Cummings K, Barrett E, Mohle-Boetani JC, Brooks JT, Farrar J, Hunt T, Fiore A, Komatsu K, Werner SB and Slutsker L. 2001. A multistate outbreak of Salmonella enterica serotype Baildon associated with domestic raw tomatoes. Emerg Infect Dis 7(6):1046–1048. Curriero FC, Patz JA, Rose JB and Lele S. 2001. The association between extreme precipitation and waterborne disease outbreaks in the United States, 1948–1994. Am J Public Health 91(8):1194–1199. Dargatz DA, Fedorka-Cray PJ, Ladely SR, Kopral CA, Ferris KE and Headrick ML. 2003. Prevalence and antimicrobial susceptibility of Salmonella spp. isolates from US cattle in feedlots in 1999 and 2000. J Appl Microbiol 95(4):753–761. Davies RH and Breslin M. 2003. Persistence of Salmonella enteritidis phage type 4 in the environment and arthropod vectors on an empty free-range chicken farm. Environ Microbiol 5(2):79–84. Doane CA, Pangloli P, Richards HA, Mount JR, Golden DA and Draughon FA. 2007. Occurrence of Escherichia coli O157:H7 in diverse farm environments. J Food Prot 70(1):6–10. Dong Y, Iniguez AL, Triplett EW and Ahmer BMM. 2003. Kinetics and strain specificity of rhizosphere and endophytic colonization by enteric bacteria on seedlings of Medicago sativa and Medicago truncatula. Appl Environ Microbiol 69(3):1783–1790. Doyle MP and Erickson MC. 2008. Summer meeting 2007—the problems with fresh produce: an overview. J Appl Microbiol 105:317–330. Duffy G. 2003. Verocytoxigenic Escherichia coli in animal faeces, manures and slurries. J Appl Microbiol 94Suppl:94S–103S. Dunn JR, Keen JE, Moreland D and Thompson RA. 2004. Prevalence of Escherichia coli O157:H7 in white-tailed deer from Louisiana. J Wildl Dis 40(2):361–365. Durso LM, Reynolds K, Bauer N, Jr. and Keen JE. 2005. Shiga-toxigenic Escherichia coli O157:H7 infections among livestock exhibitors and visitors at a Texas County Fair. Vector Borne Zoonotic Dis 5(2):193–201. Ejidokun OO, Walsh A, Barnett J, Hope Y, Ellis S, Sharp MW, Paiba GA, Logan M, Willshaw GA and Cheasty T. 2006. Human Vero cytotoxigenic Escherichia coli (VTEC) O157 infection linked to birds. Epidemiol Infect 134(2):421–423. Elder RO, Keen JE, Siragusa GR, Barkocy-Gallagher GA, Koohmaraie M and Laegreid WW. 2000. Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing [see comments]. Proc Natl Acad Sci U S A 97(7):2999–3003. FDA-CFSAN. 2001a. FDA survey of imported fresh produce, FY 1999 field assignment. January 30, 2001. http://www.cfsan.fda.gov/∼dms/prodsur6.html. Accessed Mar. 3, 2008, 2008. ———. 2001b. Survey of domestic fresh produce: interim results (July 31, 2001). July 31, 2001. http:// vm.cfsan.fda.gov/∼dms/prodsur9.html. Accessed Oct. 16, 2002, 2002.

Enteric Human Pathogens Associated with Fresh Produce

35

Feder I, Wallace FM, Gray JT, Fratamico P, Fedorka-Cray PJ, Pearce RA, Call JE, Perrine R and Luchansky JB. 2003. Isolation of Escherichia coli O157:H7 from intact colon fecal samples of swine. Emerg Infect Dis 9(3):380–383. Fegan N, Vanderlinde P, Higgs G and Desmarchelier P. 2004. Quantification and prevalence of Salmonella in beef cattle presenting at slaughter. J Appl Microbiol 97(5):892–898. ———. 2005. A study of the prevalence and enumeration of Salmonella enterica in cattle and on carcasses during processing. J Food Prot 68(6):1147–1153. Fenlon DR. 1981. Seagulls (Larus spp.) as vectors of salmonellae: an investigation into the range of serotypes and numbers of salmonellae in gull faeces. J Hyg (Lond) 86(2):195–202. Ferguson C, De Roda Husman AM, Altavilla N, Deere D and Ashbolt N. 2003. Fate and transport of surface water pathogens in watersheds. Crit Rev Environ Sci Technol 33(3):299–361. Field KG and Samadpour M. 2007. Fecal source tracking, the indicator paradigm, and managing water quality. Wat Res 41(16):3517–3538. Fischer JR, Zhao T, Doyle MP, Goldberg MR, Brown CA, Sewell CT, Kavanaugh DM and Bauman CD. 2001. Experimental and field studies of Escherichia coli O157:H7 in white-tailed deer. Appl Environ Microbiol 67(3):1218–1224. Foley SL, Lynne AM and Nayak R. 2008. Salmonella challenges: prevalence in swine and poultry and potential pathogenicity of such isolates. J Anim Sci 86:E149–162. Franz E, Semenov AV, Termorshuizen AJ, de Vos OJ, Bokhorst JG and van Bruggen AH. 2008. Manureamended soil characteristics affecting the survival of E. coli O157:H7 in 36 Dutch soils. Environ Microbiol 10(2):313–327. Franz E, Visser AA, Van Diepeningen AD, Klerks MM, Termorshuizen AJ and van Bruggen AH. 2007. Quantification of contamination of lettuce by GFP-expressing Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium. Food Microbiol 24(1):106–112. Fremaux B, Delignette-Muller ML, Prigent-Combaret C, Gleizal A and Vernozy-Rozand C. 2007. Growth and survival of non-O157:H7 Shiga-toxin-producing Escherichia coli in cow manure. J Appl Microbiol 102(1):89–99. Froder H, Martins CG, De Souza KL, Landgraf M, Franco BD and Destro MT. 2007. Minimally processed vegetable salads: microbial quality evaluation. J Food Prot 70(5):1277–1280. Gagliardi JV and Karns JS. 2000. Leaching of Escherichia coli O157:H7 in diverse soils under various agricultural management practices. Appl Environ Microbiol 66(3):877–883. ———. 2002. Persistence of Escherichia coli O157:H7 in soil and on plant roots. Environ Microbiol 4(2):89–96. Garcia A and Fox JG. 2003. The rabbit as a new reservoir host of enterohemorrhagic Escherichia coli. Emerg Infect Dis 9(12):1592–1597. Gaynor K, Park SY, Kanenaka R, Colindres R, Mintz E, Ram PK, Kitsutani P, Nakata M, Wedel S, Boxrud D, Jennings D, Yoshida H, Tosaka N, He H, Ching-Lee M and Effler PV. 2009. International foodborne outbreak of Shigella sonnei infection in airline passengers. Epidemiol Infect 137:335–341. Greene SK, Daly ER, Talbot EA, Demma LJ, Holzbauer S, Patel NJ, Hill TA, Walderhaug MO, Hoekstra RM, Lynch MF and Painter JA. 2008. Recurrent multistate outbreak of Salmonella Newport associated with tomatoes from contaminated fields, 2005. Epidemiol Infect 136(2):157–165. Guo X, Chen J, Brackett RE and Beuchat LR. 2001. Survival of salmonellae on and in tomato plants from the time of inoculation at flowering and early stages of fruit development through fruit ripening. Appl Environ Microbiol 67(10):4760–4764. Gupta SK, Nalluswami K, Snider C, Perch M, Balasegaram M, Burmeister D, Lockett J, Sandt C, Hoekstra RM and Montgomery S. 2007. Outbreak of Salmonella Braenderup infections associated with Roma tomatoes, northeastern United States, 2004: a useful method for subtyping exposures in field investigations. Epidemiol Infect 135(7):1165–1173. Hancock DD, Besser TE, Rice DH, Ebel ED, Herriott DE and Carpenter LV. 1998. Multiple sources of Escherichia coli O157 in feedlots and dairy farms in the Northwestern USA. Prev Vet Med 35:11–19. Harris LJ, Farber JN, Beuchat LR, Parish ME, Suslow TV, Garret EH and Busta FF. 2003. Outbreaks associated with fresh produce: Incidence, growth, and survival of pathogens in fresh and fresh-cut produce. Comp Rev Food Sci Food Safety. pp. 78–89. Hedberg CW, Angulo FJ, White KE, Langkop CW, Schell WL, Stobierski MG, Schuchat A, Besser JM, Dietrich S, Helsel L, Griffin PM, McFarland JW and Osterholm MT. 1999. Outbreaks of salmonellosis

36

Section I. Microbial Contamination of Fresh Produce

associated with eating uncooked tomatoes: implications for public health. The Investigation Team. Epidemiol Infect 122(3):385–393. Herman KM, Ayers TL and Lynch M. 2008. Foodborne disease outbreaks associated with leafy greens, 1973–2006. International Conference on Emerging Infectious Diseases. Atlanta, GA: Center for Disease Control and Prevention. p. 81. Hernandez J, Bonnedahl J, Waldenstrom J, Palmgren H and Olsen B. 2003. Salmonella in birds migrating through Sweden. Emerg Infect Dis 9(6):753–755. Hidalgo-Vila J, Diaz-Paniagua C, de Frutos-Escobar C, Jimenez-Martinez C and Perez-Santigosa N. 2007. Salmonella in free living terrestrial and aquatic turtles. Vet Microbiol 119(2–4):311–315. Higgins JA, Belt KT, Karns JS, Russell-Anelli J and Shelton DR. 2005. tir- and stx-positive Escherichia coli in stream waters in a metropolitan area. Appl Environ Microbiol 71(5):2511–2519. Hilborn ED, Mermin JH, Mshar PA, Hadler JL, Voetsch A, Wojtkunski C, Swartz M, Mshar R, LambertFair MA, Farrar JA, Glynn MK and Slutsker L. 1999. A multistate outbreak of Escherichia coli O157:H7 infections associated with consumption of mesclun lettuce. Arch Intern Med 159(15):1758–1764. Himathongkham S, Bahari S, Riemann H and Cliver D. 1999. Survival of Escherichia coli O157:H7 and Salmonella typhimurium in cow manure and cow manure slurry. FEMS Microbiol Lett 178(2):251–257. Holley RA, Arrus KM, Ominski KH, Tenuta M and Blank G. 2006. Salmonella survival in manure-treated soils during simulated seasonal temperature exposure. J Environ Qual 35(4):1170–1180. Hughes LA, Shopland S, Wigley P, Bradon H, Leatherbarrow AH, Williams NJ, Bennett M, de Pinna E, Lawson B, Cunningham AA and Chantrey J. 2008. Characterisation of Salmonella enterica serotype Typhimurium isolates from wild birds in northern England from 2005 to 2006. BMC Vet Res 4(1):4. Hussein HS and Bollinger LM. 2005. Prevalence of Shiga toxin-producing Escherichia coli in beef cattle. J Food Prot 68(10):2224–2241. Hutchison ML, Avery SM and Monaghan JM. 2008. The air-borne distribution of zoonotic agents from livestock waste spreading and microbiological risk to fresh produce from contaminated irrigation sources. J Appl Microbiol 105:848–857. Hyytia-Trees E, Smole SC, Fields PA, Swaminathan B and Ribot EM. 2006. Second generation subtyping: a proposed PulseNet protocol for multiple-locus variable-number tandem repeat analysis of Shiga toxinproducing Escherichia coli O157 (STEC O157). Foodborne Pathog Dis 3(1):118–131. Ihekweazu C, Barlow M, Roberts S, Christensen H, Guttridge B, Lewis D and Paynter S. 2006. Outbreak of E. coli O157 infection in the south west of the UK: risks from streams crossing seaside beaches. Euro Surveill 11(4):128–130. Isaacs S, Aramini J, Ciebin B, Farrar JA, Ahmed R, Middleton D, Chandran AU, Harris LJ, Howes M, Chan E, Pichette AS, Campbell K, Gupta A, Lior LY, Pearce M, Clark C, Rodgers F, Jamieson F, Brophy I and Ellis A. 2005. An international outbreak of salmonellosis associated with raw almonds contaminated with a rare phage type of Salmonella enteritidis. J Food Prot 68(1):191–198. Islam M, Doyle MP, Phatak SC, Millner P and Jiang X. 2004. Persistence of enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. J Food Prot 67(7):1365–1370. ———. 2005. Survival of Escherichia coli O157:H7 in soil and on carrots and onions grown in fields treated with contaminated manure composts or irrigation water. Food Microbiol 22(1):63–70. Jay MT, Cooley M, Carychao D, Wiscomb GW, Sweitzer RA, Crawford-Miksza L, Farrar JA, Lau DK, O’Connell J, Millington A, Asmundson RV, Atwill ER and Mandrell RE. 2007. Escherichia coli O157:H7 in feral swine near spinach fields and cattle, central California coast. Emerg Infect Dis 13(12):1908–1911. Johnson JY, Thomas JE, Graham TA, Townshend I, Byrne J, Selinger LB and Gannon VP. 2003. Prevalence of Escherichia coli O157:H7 and Salmonella spp. in surface waters of southern Alberta and its relation to manure sources. Can J Microbiol 49(5):326–335. Johnston LM, Jaykus LA, Moll D, Anciso J, Mora B and Moe CL. 2006a. A field study of the microbiological quality of fresh produce of domestic and Mexican origin. Int J Food Microbiol 112(2):83–95. Johnston LM, Jaykus LA, Moll D, Martinez MC, Anciso J, Mora B and Moe CL. 2005. A field study of the microbiological quality of fresh produce. J Food Prot 68(9):1840–1847. Johnston LM, Moe CL, Moll D and Jaykus L. 2006b. The epidemiology of produce-associated outbreaks of foodborne disease. In: James J, editor. Microbial hazard identification in fresh fruit and vegetables. Hoboken, NJ: John Wiley and Sons. pp. 37–72.

Enteric Human Pathogens Associated with Fresh Produce

37

Keen JE, Durso LM and Meehan TP. 2007. Isolation of Salmonella enterica and Shiga-toxigenic Escherichia coli O157 from feces of animals in public contact areas of United States zoological parks. Appl Environ Microbiol 73(1):362–365. Keen JE, Wittum TE, Dunn JR, Bono JL and Durso LM. 2006. Shiga-toxigenic Escherichia coli O157 in agricultural fair livestock, United States. Emerg Infect Dis 12(5):780–786. Keene WE, Sazie E, Kok J, Rice DH, Hancock DD, Balan VK, Zhao T and Doyle MP. 1997. An outbreak of Escherichia coli O157:H7 infections traced to jerky made from deer meat. JAMA 277: 1229–1231. Kinde H, Castellan DM, Kass PH, Ardans A, Cutler G, Breitmeyer RE, Bell DD, Ernst RA, Kerr DC, Little HE, Willoughby D, Riemann HP, Snowdon JA and Kuney DR. 2004. The occurrence and distribution of Salmonella enteritidis and other serovars on California egg laying premises: a comparison of two sampling methods and two culturing techniques. Avian Dis 48(3):590–594. Kirk JH, Holmberg CA and Jeffrey JS. 2002. Prevalence of Salmonella spp in selected birds captured on California dairies. J Am Vet Med Assoc 220(3):359–362. Kudva IT, Blanch K and Hovde CJ. 1998. Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Appl Environ Microbiol 64(9):3166–3174. Lang NL and Smith SR. 2007. Influence of soil type, moisture content and biosolids application on the fate of Escherichia coli in agricultural soil under controlled laboratory conditions. J Appl Microbiol 103(6):2122–2131. Lawley TD, Bouley DM, Hoy YE, Gerke C, Relman DA and Monack DM. 2008. Host transmission of Salmonella enterica serovar Typhimurium is controlled by virulence factors and indigenous intestinal microbiota. Infect Immun 76(1):403–416. Leclercq A and Mahillon J. 2003. Farmed rabbits and ducks as vectors for VTEC O157:H7. Vet Rec 152(23):723–724. Ledet Muller L, Hjertqvist M, Payne L, Pettersson H, Olsson A, Plym Forshell L and Andersson Y. 2007. Cluster of Salmonella Enteritidis in Sweden 2005–2006—suspected source: almonds. Euro Surveill 12(6):E9–10. LeJeune JT, Besser TE and Hancock DD. 2001. Cattle water troughs as reservoirs of Escherichia coli O157. Appl Environ Microbiol 67(7):3053–3057. LeJeune JT, Besser TE, Rice DH, Berg JL, Stilborn RP and Hancock DD. 2004. Longitudinal study of fecal shedding of Escherichia coli O157:H7 in feedlot cattle: predominance and persistence of specific clonal types despite massive cattle population turnover. Appl Environ Microbiol 70(1):377–384. Low JC, McKendrick IJ, McKechnie C, Fenlon D, Naylor SW, Currie C, Smith DG, Allison L and Gally DL. 2005. Rectal carriage of enterohemorrhagic Escherichia coli O157 in slaughtered cattle. Appl Environ Microbiol 71(1):93–97. Mahon BE, Ponka A, Hall WN, Komatsu K, Dietrich SE, Siitonen A, Cage G, Hayes PS, Lambert-Fair MA, Bean NH, Griffin PM and Slutsker L. 1997. An international outbreak of Salmonella infections caused by alfalfa sprouts grown from contaminated seeds. J Infect Dis 175(4):876–882. Makino S, Kobori H, Asakura H, Watarai M, Shirahata T, Ikeda T, Takeshi K and Tsukamoto T. 2000. Detection and characterization of Shiga toxin-producing Escherichia coli from seagulls. Epidemiol Infect 125(1):55–61. Mandrell RE and Brandl MT. 2004. Campylobacter species and fresh produce: outbreaks, incidence and biology. In: Beier RC, Pillai SD, Phillips TD and Ziprin RL, editors. Preharvest and Postharvest Food Safety: Contemporary Issues and Future Directions. Ames, IA: IFT Press and Blackwell Publishing. pp. 59–72. Mandrell RE, Gorski L and Brandl MT. 2006. Attachment of microorganisms to fresh produce. In: Sapers GM, Gorny JR and Yousef AE, editors. Microbiology of Fruits and Vegetables. Boca Raton, FL: CRC, Taylor and Francis. pp. 33–74. Manning SD, Motiwala AS, Springman AC, Qi W, Lacher DW, Ouellette LM, Mladonicky JM, Somsel P, Rudrik JT, Dietrich SE, Zhang W, Swaminathan B, Alland D and Whittam TS. 2008. Variation in virulence among clades of Escherichia coli O157:H7 associated with disease outbreaks. Proc Natl Acad Sci U S A. Matthews L, Low JC, Gally DL, Pearce MC, Mellor DJ, Heesterbeek JA, Chase-Topping M, Naylor SW, Shaw DJ, Reid SW, Gunn GJ and Woolhouse ME. 2006. Heterogeneous shedding of Escherichia coli O157 in cattle and its implications for control. Proc Natl Acad Sci U S A 103(3):547–552.

38

Section I. Microbial Contamination of Fresh Produce

McCrea BA, Tonooka KH, VanWorth C, Boggs CL, Atwill ER and Schrader JS. 2006. Prevalence of Campylobacter and Salmonella species on farm, after transport, and at processing in specialty market poultry. Poult Sci 85(1):136–143. McGee P, Bolton DJ, Sheridan JJ, Earley B, Kelly G and Leonard N. 2002. Survival of Escherichia coli O157:H7 in farm water: its role as a vector in the transmission of the organism within herds. J Appl Microbiol 93(4):706–713. MDPH. Health officials investigate E. coli O157:H7 cases related to Dole prepackaged lettuce mixes sold at Rainbow Foods. Sept. 28, 2006. http://www.health.state.mn.us/news/pressrel/ecoli093005.html. Accessed. Mead PS, Slutsker L, Dietz V, McCaig LF, Bresee JS, Shapiro C, Griffin PM and Tauxe RV. 1999. Food -related illness and death in the United States. Emerg Infect Dis 5(5):607–625. Meays CL, Broersma K, Nordin R and Mazumder A. 2004. Source tracking fecal bacteria in water: A critical review of current methods. J Environ Manag 73(1):71–79. Meinersmann RJ, Berrang ME, Jackson CR, Fedorka-Cray P, Ladely S, Little E, Frye JG and Mattsson B. 2008. Salmonella, Campylobacter and Enterococcus spp.: their antimicrobial resistance profiles and their spatial relationships in a synoptic study of the Upper Oconee River basin. Microb Ecol 55(3):444–452. Michel P, Wilson JB, Martin SW, Clarke RC, McEwen SA and Gyles CL. 1999. Temporal and geographical distributions of reported cases of Escherichia coli O157:H7 infection in Ontario. Epidemiol Infect 122(2):193–200. Michino H, Araki K, Minami S, Takaya S, Sakai N, Miyazaki M, Ono A and Yanagawa H. 1999. Massive outbreak of Escherichia coli O157:H7 infection in school children in Sakai City, Japan, associated with consumption of white radish sprouts. Am J Epidemiol 150(8):787–796. Miller WG and Mandrell RE. 2006. Campylobacter in the food and water supply: prevalence, outbreaks, isolation, and detection. In: Ketley J and Konkel ME, editors. Campylobacter jejuni: New Perspectives in Molecular and Cellular Biology. Norfolk, UK: Horizon Scientific Press. pp. 101–163. Milnes AS, Stewart I, Clifton-Hadley FA, Davies RH, Newell DG, Sayers AR, Cheasty T, Cassar C, Ridley A, Cook AJ, Evans SJ, Teale CJ, Smith RP, McNally A, Toszeghy M, Futter R, Kay A and Paiba GA. 2008. Intestinal carriage of verocytotoxigenic Escherichia coli O157, Salmonella, thermophilic Campylobacter and Yersinia enterocolitica, in cattle, sheep and pigs at slaughter in Great Britain during 2003. Epidemiol Infect 136:739–751. MMWR. 2002. Multistate outbreaks of Salmonella serotype Poona infections associated with eating cantaloupe from Mexico—United States and Canada, 2000–2002. MMWR Morb Mortal Wkly Rep 51(46):1044–1047. ———. 2004. Outbreak of Salmonella serotype Enteritidis infections associated with raw almonds—United States and Canada, 2003–2004. MMWR Morb Mortal Wkly Rep 53(22):484–487. ———. 2005a. Outbreaks of Salmonella infections associated with eating Roma tomatoes—United States and Canada, 2004. MMWR Morb Mortal Wkly Rep 54(13):325–328. ———. 2005b. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food—10 sites, United States, 2004. MMWR Morb Mortal Wkly Rep 54(14):352–356. ———. 2007a. Multistate outbreaks of Salmonella infections associated with raw tomatoes eaten in restaurants—United States, 2005–2006. MMWR Morb Mortal Wkly Rep 56(35):909–911. ———. 2007b. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food—10 states, 2006. MMWR Morb Mortal Wkly Rep 56(14):336–339. Mohle-Boetani JC, Farrar J, Bradley P, Barak JD, Miller M, Mandrell R, Mead P, Keene WE, Cummings K, Abbott S and Werner SB. 2009. Salmonella infections associated with mung bean sprouts: epidemiological and environmental investigations. Epidemiol Infect 137:357–366. Mohle-Boetani JC, Reporter R, Werner SB, Abbott S, Farrar J, Waterman SH and Vugia DJ. 1999. An outbreak of Salmonella serogroup Saphra due to cantaloupes from Mexico. J Infect Dis 180(4):1361–1364. Moore JE, Wilson TS, Wareing DR, Humphrey TJ and Murphy PG. 2002. Prevalence of thermophilic Campylobacter spp. in ready-to-eat foods and raw poultry in Northern Ireland. J Food Prot 65(8):1326–1328. Morabito S, Dell’Omo G, Agrimi U, Schmidt H, Karch H, Cheasty T and Caprioli A. 2001. Detection and characterization of Shiga toxin-producing Escherichia coli in feral pigeons. Vet Microbiol 82(3):275–283. Mukherjee A, Cho S, Scheftel J, Jawahir S, Smith K and Diez-Gonzalez F. 2006. Soil survival of Escherichia coli O157:H7 acquired by a child from garden soil recently fertilized with cattle manure. J Appl Microbiol 101(2):429–436.

Enteric Human Pathogens Associated with Fresh Produce

39

Mukherjee A, Speh D, Dyck E and Diez-Gonzalez F. 2004. Preharvest evaluation of coliforms, Escherichia coli, Salmonella, and Escherichia coli O157:H7 in organic and conventional produce grown by Minnesota farmers. J Food Prot 67(5):894–900. Murphy M, Minihan D, Buckley JF, O’Mahony M, Whyte P and Fanning S. 2008. Multiple-locus variable number of tandem repeat analysis (MLVA) of Irish verocytotoxigenic Escherichia coli O157 from feedlot cattle: uncovering strain dissemination routes. BMC Vet Res 4:2. Nakazawa M and Akiba M. 1999. Swine as a potential reservoir of shiga toxin-producing Escherichia coli O157:H7 in Japan. Emerg Infect Dis 5(6):833–834. Nguyen-the C and Carlin F. 1994. The microbiology of minimally processed fresh fruits and vegetables. Crit Rev Food Sci Nutr 34(4):371–401. ———. 2000. Fresh and processed vegetables. In: Lund B, Baird-Parker T and Gould G, editors. The microbiological safety and quality of food. Gaithersburg, MD: Aspen. pp. 620–684. Nielsen BB, Clausen B and Elvestad K. 1981. The incidence of Salmonella bacteria in wild-living animals from Denmark and in imported animals. Nord Vet Med 33(9–11):427–433. Nuorti JP, Niskanen T, Hallanvuo S, Mikkola J, Kela E, Hatakka M, Fredriksson-Ahomaa M, Lyytikainen O, Siitonen A, Korkeala H and Ruutu P. 2004. A widespread outbreak of Yersinia pseudotuberculosis O:3 infection from iceberg lettuce. J Infect Dis 189(5):766–774. Nygard K, Lassen J, Vold L, Andersson Y, Fisher I, Lofdahl S, Threlfall J, Luzzi I, Peters T, Hampton M, Torpdahl M, Kapperud G and Aavitsland P. 2008. Outbreak of Salmonella Thompson infections linked to imported rucola lettuce. Foodborne Pathog Dis. In press: Online. Ogden ID, Hepburn NF, MacRae M, Strachan NJ, Fenlon DR, Rusbridge SM and Pennington TH. 2002. Long-term survival of Escherichia coli O157 on pasture following an outbreak associated with sheep at a scout camp. Lett Appl Microbiol 34(2):100–104. Ogden ID, MacRae M and Strachan NJ. 2005. Concentration and prevalence of Escherichia coli O157 in sheep faeces at pasture in Scotland. J Appl Microbiol 98(3):646–651. Oporto B, Esteban JI, Aduriz G, Juste RA and Hurtado A. 2008. Escherichia coli O157:H7 and non-O157 shiga toxin-producing E. coli in healthy cattle, sheep and swine herds in Northern Spain. Zoonoses Pub Health 55(2):73–81. Pangloli P, Dje Y, Ahmed O, Doane CA, Oliver SP and Draughon FA. 2008. Seasonal incidence and molecular characterization of Salmonella from dairy cows, calves, and farm environment. Foodborne Pathog Dis 5(1):87–96. Pritchard GC, Williamson S, Carson T, Bailey JR, Warner L, Willshaw G and Cheasty T. 2001. Wild rabbits— a novel vector for verocytotoxigenic Escherichia coli O157. Vet Rec 149(18):567. Proctor ME, Hamacher M, Tortorello ML, Archer JR and Davis JP. 2001. Multistate outbreak of Salmonella serovar Muenchen infections associated with alfalfa sprouts grown from seeds pretreated with calcium hypochlorite. J Clin Microbiol 39(10):3461–3465. Rasschaert G, Houf K, Van Hende J and De Zutter L. 2007. Investigation of the concurrent colonization with Campylobacter and Salmonella in poultry flocks and assessment of the sampling site for status determination at slaughter. Vet Microbiol 123(1–3):104–109. Reller ME, Nelson JM, Molbak K, Ackman DM, Schoonmaker-Bopp DJ, Root TP and Mintz ED. 2006. A large, multiple-restaurant outbreak of infection with Shigella flexneri serotype 2a traced to tomatoes. Clin Infect Dis 42(2):163–169. Renter DG, Gnad DP, Sargeant JM and Hygnstrom SE. 2006. Prevalence and serovars of Salmonella in the feces of free-ranging white-tailed deer (Odocoileus virginianus) in Nebraska. J Wildl Dis 42(3):699–703. Renter DG, Sargeant JM, Hygnstorm SE, Hoffman JD and Gillespie JR. 2001. Escherichia coli O157:H7 in free-ranging deer in Nebraska. J Wildl Dis 37(4):755–760. Rice DH, McMenamin KM, Pritchett LC, Hancock DD and Besser TE. 1999. Genetic subtyping of Escherichia coli O157 isolates from 41 Pacific Northwest USA cattle farms. Epidemiol Infect 122(3):479–484. Rimhanen-Finne R, Niskanen T, Hallanvuo S, Makary P, Haukka K, Pajunen S, Siitonen A, Ristolainen R, Poyry H, Ollgren J and Kuusi M. 2009. Yersinia pseudotuberculosis causing a large outbreak associated with carrots in Finland, 2006. Epidemiol Infect 137:342–347. Rose BE, Hill WE, Umholtz R, Ransom GM and James WO. 2002. Testing for Salmonella in raw meat and poultry products collected at federally inspected establishments in the United States, 1998 through 2000. J Food Prot 65(6):937–947.

40

Section I. Microbial Contamination of Fresh Produce

Sagoo SK, Little CL and Mitchell RT. 2001. The microbiological examination of ready-to-eat organic vegetables from retail establishments in the United Kingdom. Lett App Microbiol 33(6):434–439. Sagoo SK, Little CL, Ward L, Gillespie IA and Mitchell RT. 2003. Microbiological study of ready-to-eat salad vegetables from retail establishments uncovers a national outbreak of salmonellosis. J Food Prot 66(3):403–409. Santo Domingo JW, Harmon S and Bennett J. 2000. Survival of Salmonella species in river water. Curr Microbiol 40(6):409–417. Sargeant JM, Hafer DJ, Gillespie JR, Oberst RD and Flood SJ. 1999. Prevalence of Escherichia coli O157:H7 in white-tailed deer sharing rangeland with cattle. J Am Vet Med Assoc 215(6):792–794. Schikora A, Carreri A, Charpentier E and Hirt H. 2008. The dark side of the salad: Salmonella typhimurium overcomes the innate immune response of Arabidopsis thaliana and shows an endopathogenic lifestyle. PLoS ONE 3(5):e2279. Scott L, McGee P, Sheridan JJ, Earley B and Leonard N. 2006. A comparison of the survival in feces and water of Escherichia coli O157:H7 grown under laboratory conditions or obtained from cattle feces. J Food Prot 69(1):6–11. Sewell AM and Farber JM. 2001. Foodborne outbreaks in Canada linked to produce. J Food Prot 64(11):1863–1877. Seymour IJ and Appleton H. 2001. Foodborne viruses and fresh produce. J Appl Microbiol 91(5): 759–773. Sivapalasingam S, Barrett E, Kimura A, Van Duyne S, De Witt W, Ying M, Frisch A, Phan Q, Gould E, Shillam P, Reddy V, Cooper T, Hoekstra M, Higgins C, Sanders JP, Tauxe RV and Slutsker L. 2003. A multistate outbreak of Salmonella enterica serotype Newport infection linked to mango consumption: impact of water-dip disinfestation technology. Clin Infect Dis 37(12):1585–1590. Sivapalasingam S, Friedman CR, Cohen L and Tauxe RV. 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J Food Prot 67(10):2342–2353. Soderstrom A, Osterberg P, Lindqvist A, Jonsson B, Lindberg A, Blide Ulander S, Welinder-Olsson C, Lofdahl S, Kaijser B, De Jong B, Kuhlmann-Berenzon S, Boqvist S, Eriksson E, Szanto E, Andersson S, Allestam G, Hedenstrom I, Ledet Muller L and Andersson Y. 2008. A large Escherichia coli O157 outbreak in Sweden associated with locally produced lettuce. Foodborne Pathog Dis 5(3):339–349. Solomon EB, Brandl MT and Mandrell RE. 2006. Biology of foodborne pathogens on produce. In: Matthews KR, editor. Emerging Issues in Food Safety: Microbiology of Fresh Produce. Washington, D.C.: ASM Press. pp. 55–83. Solomon EB, Yaron S and Matthews KR. 2002. Transmission of Escherichia coli O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalization. Appl Environ Microbiol 68(1):397–400. Soriano JM, Rico H, Molto JC and Manes J. 2001. Listeria species in raw and ready-to-eat foods from restaurants. J Food Prot 64(4):551–553. Sproston EL, Macrae M, Ogden ID, Wilson MJ and Strachan NJ. 2006. Slugs: potential novel vectors of Escherichia coli O157. Appl Environ Microbiol 72(1):144–149. Srikantiah P. 2002. Outbreak of Salmonella Javiana infections among participants of the 2002 US Transplant Games—Orlando, Florida, June 2002. Centers for Disease Control and Prevention. Atlanta, GA. Stoeckel DM, Mathes MV, Hyer KE, Hagedorn C, Kator H, Lukasik J, O’Brien TL, Fenger TW, Samadpour M, Strickler KM and Wiggins BA. 2004. Comparison of seven protocols to identify fecal contamination sources using Escherichia coli. Environ Sci Technol 38(22):6109–6117. Strachan NJ, Fenlon DR and Ogden ID. 2001. Modelling the vector pathway and infection of humans in an environmental outbreak of Escherichia coli O157. FEMS Microbiol Lett 203(1):69–73. Swaminathan B, Barrett TJ, Hunter SB and Tauxe RV. 2001. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg Infect Dis 7(3):382–389. Tablante NL and Lane VM. 1989. Wild mice as potential reservoirs of Salmonella dublin in a closed dairy herd. Can Vet J 30(7):590–592. Teunis P, Takumi K and Shinagawa K. 2004. Dose response for infection by Escherichia coli O157:H7 from outbreak data. Risk Anal 24(2):401–407. Thilmony R, Underwood W and He SY. 2006. Genome-wide transcriptional analysis of the Arabidopsis thaliana interaction with the plant pathogen Pseudomonas syringae pv. tomato DC3000 and the human pathogen Escherichia coli O157:H7. Plant J 46(1):34–53.

Enteric Human Pathogens Associated with Fresh Produce

41

Thomas KM, Charron DF, Waltner-Toews D, Schuster C, Maarouf AR and Holt JD. 2006. A role of high impact weather events in waterborne disease outbreaks in Canada, 1975–2001. Int J Environ Health Res 16(3):167–180. Thunberg RL, Tran TT, Bennett RW, Matthews RN and Belay N. 2002. Microbial evaluation of selected fresh produce obtained at retail markets. J Food Prot 65(4):677–682. Tuyet DT, Yassibanda S, Nguyen Thi PL, Koyenede MR, Gouali M, Bekondi C, Mazzi J and Germani Y. 2006. Enteropathogenic Escherichia coli O157 in Bangui and N’Goila, Central African Republic: A brief report. Am J Trop Med Hyg 75(3):513–515. Uesugi AR, Danyluk MD, Mandrell RE and Harris LJ. 2007. Isolation of Salmonella Enteritidis phage type 30 from a single almond orchard over a 5-year period. J Food Prot 70(8):1784–1789. USDA-AMS-MDP. U.S. Department of Agriculture, Agricultural Marketing Service, Microbial Data Program. Data summaries, 2002–2006. July 9, 2008, Washington, DC. http://www.ams.usda.gov/ science/mpo/Mdp.htm. Accessed July 23, 2008. USDA-ERS. 2001. Lettuce: In & Out of the Bag. Washington, DC. http://www.ers.usda.gov/publications/ AgOutlook/April2001/AO280d.pdf. Valentin-Bon I, Jacobson A, Monday SR and Feng PC. 2008. Microbiological quality of bagged cut spinach and lettuce mixes. Appl Environ Microbiol 74(4):1240–1242. Van Kessel JS, Pachepsky YA, Shelton DR and Karns JS. 2007. Survival of Escherichia coli in cowpats in pasture and in laboratory conditions. J Appl Microbiol 103(4):1122–1127. Vital M, Hammes F and Egli T. 2008. Escherichia coli O157 can grow in natural freshwater at low carbon concentrations. Environ Microbiol 10:2387–2396. Wahlstrom H, Tysen E, Olsson Engvall E, Brandstrom B, Eriksson E, Morner T and Vagsholm I. 2003. Survey of Campylobacter species, VTEC O157 and Salmonella species in Swedish wildlife. Vet Rec 153(3):74–80. Wang G and Doyle MP. 1998. Survival of enterohemorrhagic Escherichia coli O157:H7 in water. J Food Prot 61(6):662–667. Warriner K, Ibrahim F, Dickinson M, Wright C and Waites WM. 2003. Internalization of human pathogens within growing salad vegetables. Biotechnol Genet Eng Rev 20:117–134. WDOH. E. coli infections detected in two Puget Sound counties. June 5, 2008. Health WSDo, Seattle, WA. http://www.doh.wa.gov/Publicat/2008_news/08-092.htm. Accessed June 12, 2008. Wetzel AN and LeJeune JT. 2006. Clonal dissemination of Escherichia coli O157:H7 subtypes among dairy farms in northeast Ohio. Appl Environ Microbiol 72(4):2621–2626. Williams AP, Avery LM, Killham K and Jones DL. 2007. Survival of Escherichia coli O157:H7 in the rhizosphere of maize grown in waste-amended soil. J Appl Microbiol 102(2):319–326. Wimpenny JWT, Cotton N and Statham M. 1972. Microbes as tracers of water movement. Wat Res 6:731–773. You Y, Rankin SC, Aceto HW, Benson CE, Toth JD and Dou Z. 2006. Survival of Salmonella enterica serovar Newport in manure and manure-amended soils. Appl Environ Microbiol 72(9):5777–5783. Zhang W, Qi W, Albert TJ, Motiwala AS, Alland D, Hyytia-Trees EK, Ribot EM, Fields PI, Whittam TS and Swaminathan B. 2006. Probing genomic diversity and evolution of Escherichia coli O157 by single nucleotide polymorphisms. Genome Res 16(6):757–767.

2

The Origin and Spread of Human Pathogens in Fruit Production Systems Susan Bach and Pascal Delaquis

Introduction Botanists define fruit as the seed-bearing structures of angiosperms formed from ovaries after flowering. In culinary tradition the term refers to fleshy, succulent plant parts that are harvested seasonally when the degree of ripeness is conducive to preparation for immediate consumption or for processing and preservation. Fruit has long been an integral part of the human diet. There is clear evidence of communal harvest by ancient hunter gatherers, and recent archeological finds suggest that purposeful cultivation of figs was practiced 11,400 years ago in the Near East (Gibbons 2006). Fruit consumption rates and patterns in premodern times varied widely with geographic location and cultural practice. At present, many countries in the developed world are reporting higher intakes, presumably in response to growing awareness of the benefits of fruit for the maintenance of health and the prevention of chronic diseases (Ness and Powles 1997; Takachi and others 2008). Overall production capacity and international trade are expanding to meet increasing demand, particularly for fresh whole or minimally processed (“fresh-cut”) products. Unfortunately, epidemiological data reveal a corresponding increase in foodborne illnesses caused by consumption of fresh fruit or fruit products contaminated with infectious microorganisms. Hence there is an urgent need to recognize potential sources of such hazards, to establish their fate at all stages along the production-to-consumption continuum, and to develop effective risk-mitigation strategies.

Role of Fresh Fruit in Foodborne Illness Consumption of fresh fruit is associated with a range of foodborne illnesses caused by viruses, bacteria, and single-celled parasites. Table 2.1 was gleaned from several references provided in this work and is intended to illustrate the diversity of pathogens linked to known outbreaks together with the range of implicated products. An exhaustive compilation of U.S. data showed that the number of reported outbreaks linked to fresh produce including fruit rose between 1973 and 2001 (Sivapalasingam and others 2004). Similar patterns are evident in epidemiological data from other jurisdictions (De Roever 1999; Sewell and Farber 2001). While the role of fresh fruit in documented outbreaks is now well established, the relative contribution to sporadic foodborne illness is unknown. Analysis of available data hints that Salmonella serovars are the single most common cause of fruit-associated foodborne illness and that infections implicating verocytotoxigenic E. coli (including serovar O157:H7) are occurring with increasing 43

44

Section I. Microbial Contamination of Fresh Produce

Table 2.1. Fresh fruit commodities or fruit products implicated in outbreaks of foodborne illness and associated etiological agents Etiological Agent

Implicated Fruit or Fruit Product

Hepatitis A

Frozen strawberries

Norovirus Calicivirus Salmonella E. coli O157:H7 E. coli O11:H43 Campylobacter jejuni Staphylococcus aureus Clostridium botulinum

Mixed fresh-cut fruit, raspberries, melon Frozen mixed fruit Melon, apple juice, orange juice, fruit salad Cantaloupe, apple cider Pineapple Melon, strawberries, fruit salad, orange juice Strawberries Carrot juice

Cryptosporidium parvum Giardia lamblia Cyclospora cayetanensis

Apple juice Fruit salad Raspberries, fruit salad

frequency. However, it should be stressed that detection of enteric viruses in clinical samples and foods remains technically challenging, and illnesses caused by these pathogens are likely underreported. Although Campylobacter spp. are recognized as a major cause of foodborne infections in the Western world, relatively few examples of transmission through fresh produce have been recorded to date. Consumption of raw fruit is nevertheless identified as a risk factor for sporadic campylobacteriosis (Neimann and others 2003). Microbiological surveys indicate that pathogenic species of this genus are rarely found in crops grown in Western countries but that they frequently contaminate salad vegetables or fruit in developing countries (GarciaVillanova Ruiz and others 1987; Chai and others 2007; Hussain and others 2007). The latter is a concern given expanding international trade and reports of outbreaks associated with imported products. A notable example occurred in the 1990s when several infections caused by the single-celled parasite Cyclospora cayetanensis were reported in Canada and the U.S., countries well outside the geographical range for the species. Raspberries imported from Central America were implicated in these events (Herwaldt and Ackers 1997). Although unsanitary handling or processing was responsible for some documented incidents, the majority could be ascribed to contamination during production. Analysis of available data also suggests that some commodities are more frequently implicated in such events, including cantaloupe melons. Fundamental differences imparted by the nature of the fruit, the characteristics of specific production systems, and postharvest handling schemes prohibit strict comparisons across commodities. Nevertheless, observations derived from outbreak investigations and ensuing research have provided important data about the ecology of foodborne pathogens in fruit production systems and the means by which they contaminate crops. The following discussion therefore draws heavily on information derived from this body of work to summarize the state of knowledge about the origin and spread of human pathogens in fruit production systems.

The Origin and Spread of Human Pathogens in Fruit Production Systems

45

Preharvest Contamination of Crops The majority of pathogens implicated in fruit-associated outbreaks are enteric microorganisms that originate from the gut of warm-blooded animals. Hence infectious agents acquired from the consumption of contaminated fruit are transmitted via the fecal-oral route. Transmission of pathogens in agricultural settings is distinctive with respect to time elapsed subsequent to shedding, the nature of implicated vectors, and distance from source. A schematic representation of potential sources of infectious microorganisms and their mode of transmission in fruit production systems is provided in Figure 2.1. The risk of contamination with enteric microorganisms introduced to the production environment is clearly influenced by the nature of the plant. Melons, for example, are generally grown in intimate contact with the ground where the risk of direct adsorption of soil microorganisms is high (Materon and others 2007). Evidence of biofilm formation by Salmonella spp. in melon rinds suggests that some enteric microorganisms may be capable of growth in the field (Annous and others 2005). In contrast, aerial fruit such as apples develop in an environment where direct contact with concentrated sources of microorganisms is less common and opportunities for growth are likely rare. Contamination via Animal Wastes Human pathogens can be introduced to fruit production systems through animal feces from wildlife or livestock or by the application of manures or sewage sludges

Feces from livestock, wildlife

Wind Dust

Insects

Harvesting equipment

Irrigation water

Human contact

Urban, agricultural pollution Animal fertilizers

Postharvest handling

Figure 2.1. Potential sources of human pathogens and their transmission in fruit production systems.

46

Section I. Microbial Contamination of Fresh Produce

as fertilizers. The latter practice is increasingly controversial in the developed world since research has shown that viral and bacterial pathogens are capable of extended survival or persistence in agricultural soils (Bagdasaryan 1964; Natvig and others 2002; Islam and others 2004, 2005; Nicholson and others 2005). Research on the behavior of pathogens under field conditions has focused on vegetable production systems, and comparatively less is known about their persistence or the risk of transfer in fruit production systems. In spite of deficiencies in data for specific crops there is a clear trend toward reducing or eliminating the use of untreated animal wastes as soil amendments or fertilizers for all horticultural crops destined for the fresh or fresh-cut market, including fruit. Industry-based associations or regulatory bodies in many jurisdictions now either encourage or enforce treatment of animal wastes by physical (heat, irradiation) or biological (composting) means before field application to reduce or eliminate the associated risk. Direct crop contamination with animal feces is more difficult to manage. Although livestock access to production sites can be restricted—by fencing, for example— some species of wildlife are difficult or impossible to deter. Bird control is an ongoing, significant source of economic losses in fruit or berry production despite the deployment of a variety of passive or lethal measures. The feces of many avian species have been shown to contain pathogens such as Salmonella or E. coli O157:H7 (Wallace and others 1997). None of the fruit-associated outbreaks reported to date have been conclusively linked to contamination by wildlife feces. However, there is convincing evidence that apples used in the manufacture of unpasteurized apple juice responsible for some outbreaks caused by E. coli O157:H7 may have been contaminated by contact with the feces of infected deer (Riordan and others 2001; Garcia and others 2006). Invertebrates have been shown to acquire human pathogens through contact with animal waste or environments polluted by animal feces. Insects can carry and excrete human pathogens (Xu and others 2003; Alam and Zurek 2004) and insect-mediated transfer to fruit has been demonstrated experimentally. For example, Janisiewicz and others (1999) showed that infected fruit flies can transfer E. coli O157:H7 to exposed apple tissues. Nematodes (Gibbs and others 2005) and slugs (Sproston and others 2006) have also been shown to excrete human pathogens and may serve as reservoirs in soil or the production environment. However, the role of invertebrates as vectors for the transmission of human pathogens in commercial production systems and the inherent risk are presently unknown due to a dearth of field data. Contamination through Irrigation Water Water is a well-recognized vector in disease transmission. Enteric viral, bacterial, or single-celled pathogens in animal waste can be carried in large numbers over great distances through natural or man-made waterways. The strong association between irrigation water quality and produce safety is supported by epidemiological data from several incidents involving vegetables wherein the implicated etiological agents were recovered from infected individuals and irrigation water (Takkinen and others 2005). There are no similar reports for fruit-associated outbreaks, although field or experimental evidence points to the potential role of irrigation water as a vector for contamination (Espinoza-Medina and others 2006). For example, Materon and others

The Origin and Spread of Human Pathogens in Fruit Production Systems

47

(2007) showed that Salmonella was introduced to melon fields in Texas via irrigation water and that fruit grown in the fields were contaminated with the bacterium. Raspberries responsible for several outbreaks were grown in Guatemala where contaminated water is a significant factor in the spread of Cyclospora cayetanensis (Mota and others 2000; Mansfield and Gajadhar 2004). Growing demand and competition for water to irrigate food crops and to satisfy public or industrial uses has put increasing pressure on surface and groundwater supplies in many parts of the world. Commercial-scale production of fruit crops is a water-intensive activity, and natural supplies are used where the availability or cost of treated water is restrictive. The microbiological quality of surface waters is changeable, particularly near urban areas or where mixed agriculture is practiced. Subsurface sources generally present less risk, although distant transport of human pathogens through groundwater has been demonstrated and wells can become contaminated (Bitton and others 1983; Powell and others 2003). Notwithstanding the source of irrigation water, frequent testing is required to reduce the risk of crop contamination. Microbiological standards based on fecal coliforms or E. coli tend to vary, but tolerances are increasingly restrictive in recognition of the associated safety risks. The World Health Organization recommends that water applied to crops eaten raw contain <1,000 fecal coliform bacteria/100 ml (WHO 1989). Canadian standards, for example, include the additional provision that samples contain <100 E. coli/100 ml (CCME 1987). Although it is not possible to correlate the level of disease risk associated with a given titer of indicator microorganisms, these guidelines will undoubtedly continue to be used until economical technologies for the rapid, direct detection and identification of human pathogens in water become available. Furthermore, it is difficult to meet guidelines consistently in many producing regions, particularly where use of reclaimed waters from sewage treatment plants is increasingly common (Steele and Odumeru 2004). To date, little progress has been reported in the development of practical, field-deployable water-disinfection units that can accommodate volumes needed for large-scale commercial fruit production. Research in vegetable production systems suggests that overhead irrigation leads to higher contamination levels on the crop compared with furrow or drip irrigation (Keraita and others 2007). The effect of irrigation practice on the transfer of pathogens to fruit crops has not been examined in detail. Overhead irrigation would appear to increase the risk due to the higher probability of direct contact with fruit. Microorganisms deposited on plant surfaces are subjected to considerable stresses, however, and some reports suggest that postirrigation die-off is rapid due to UV radiation or dehydration (Brandl 2006). There is clearly a need to better understand the role of different irrigation methods on the fate of human pathogens in fruit production systems. Airborne Contamination Research carried out in urban environments has shown that enteric bacteria such as E. coli can be present in air or dust disseminated by surface winds (Rosas and others 1997). Microbiological analysis of air at high altitudes and the historical record indicate that transcontinental spread of human and plant pathogenic species is clearly possible (Mohr 1997). The relationship between air quality within confined processing

48

Section I. Microbial Contamination of Fresh Produce

environments and the microbiology of foods is recognized (al-Dagal and Fung 1990), and lateral transmission by the airborne route has been demonstrated in animal production (Holt and others 1998). In contrast, the role of air in the dissemination of human pathogens in agricultural environments remains undefined. It is currently difficult to predict the associated health risks due to a lack of information on the fate and transport of bioaerosols or contaminated dusts in open environments (Pillai and Ricke 2002). Given the frequent association between proximity to potential sources of fecal contamination and produce-associated outbreaks, there is a manifest need to determine the risk of airborne transmission in horticultural production systems.

Contamination at Harvest Harvest provides many opportunities for the introduction to or dissemination of human pathogens in fruit crops. Fruit destined for the fresh market is generally picked by hand. The role of hands in the transmission of human pathogens during food handling or food preparation is well established (Bidawid and others 2000; Courtenay and others 2005). Prevention strategies based on physical barriers such as gloves or disinfection using soaps and sanitizers can effectively reduce such risks in the foodprocessing or food-service environment (Kramer and others 2002; Montville and others 2002; Michaels and others 2003, 2004). Although there is evidence that the hands of field workers can become contaminated with human pathogens during harvest (Materon and others 2007), little is known about their role in the transfer of contaminants to fruit. In addition, sanitation schemes designed for food handlers may not be adaptable to the agricultural environment where constant soiling of hands and contact with potential sources of contamination are unavoidable. In some cases the wearing of gloves may be too restrictive to permit efficient harvest. Contact between fruit or with surfaces in bins or conveying equipment increases throughout harvest. Research carried out with melons (Gagliardi and others 2003), apples (Abadias and others 2006) and persimmons (Izumi and others 2008) indicates that such contact results in higher microbial loads on the external surfaces of fruit, and that soil is likely the main source of microorganisms acquired at this stage. In some cases water may be used to convey, wash, or cool fruit in the field. The microbiological quality of water used in these operations can be maintained by the use of a sanitizer. Chlorine is widely used for this purpose, although chlorine demand tends to increase rapidly in field units due to soil or detritus accumulation, and effective antimicrobial levels can be difficult to maintain.

A Case Study: Dissemination of Enteric Bacteria in Sweet Cherry Orchards Considerable research has been carried out to identify sources of human pathogens, the means by which they are disseminated, and their fate in vegetable production systems. Comparatively less effort has been directed at understanding these occurrences in fruit production systems. Evidently, most fruit-bearing plants are difficult to grow in the laboratory, and biosafety issues prohibit field inoculation with actual human pathogens. Hence different approaches are needed to establish the behavior of human pathogens in fruit production systems. The bacterium E. coli has long been

The Origin and Spread of Human Pathogens in Fruit Production Systems

49

used as an indicator of fecal contamination in food or environmental samples. Selective cultural methods combined with molecular genotyping techniques can be used to track the movement of the species in natural environments. We used this approach to determine sources of enteric bacteria in sweet cherry production systems and to examine their fate during harvest. Tree fruit samples collected before harvest or after hydrocooling and cecal swabs from downed birds were analyzed for the presence of generic E. coli. Hand swabs were also obtained from pickers and workers on sorting lines. Table 2.2 shows that E. coli was never recovered from fruit collected before harvest, but 23.2% of an equal number of samples of harvested, hydro-cooled fruit samples were positive. Hydro-cooling water in all the test orchards was chlorinated and E. coli was never found in water samples, although one operation employed an unchlorinated dump tank from which E. coli was recovered. The bacterium was also present in 35.7% of swabs performed on hands of pickers, and 15% from line sorters. Genotyping by the enterobacterial repetitive intergenic consensus (ERIC)-PCR reaction was used to establish genetic relatedness among isolates from all the samples. As shown in Figure 2.2, genotypes associated with birds were found on the hands of pickers and sorters and in finished fruit. This investigation confirmed that wildlife is a significant source of enteric bacteria in sweet cherry production systems. The inability to isolate E. coli from preharvest fruit suggests that fecal contamination is likely heterogeneously distributed in the orchard environment. Harvest and postharvest handling lead to the dissemination of enteric bacteria through harvested fruit, principally by contact with hands. Cherries destined for the fresh market are always harvested manually to maintain desirable quality attributes, notably the presence of an intact stem. The motion required to remove intact fruit from the tree cannot be accomplished with a gloved hand, and irritation or injuries to index fingers are frequent, as shown in Figure 2.3. Hands quickly become heavily soiled, dry, and chapped over the course of a typical harvest day. Work schedules preclude frequent stops for washing with soap and water; and available options for hand disinfection in the field, such as disposable alcohol-based pads, are either impractical or ineffective. Although worker sanitation has vastly improved in response to an increasing awareness of food safety issues, there remains a need to develop practical solutions to hand disinfection under field conditions.

Table 2.2. Proportions of pre- and postharvest fruit, bird, and hand samples positive for the presence of E. coli in cherry orchards Orchard

Birds

1

Fruit on the Tree 0/30

2 3 4 %

0/30 0/5 0/5 0

NA = no samples available.

15/30

Pickers’ Hands 3/10

Sorters’ Hands 1/10

Fruit from Packing Line 0/10

Dump Tank NA

11/30 NA NA 86.6

12/30 7/20 3/20 35.7

3/30 5/20 3/20 15.0

1/20 7/19 8/20 23.2

NA NA 2/2 100

40

45

50

55

60

65

70

75

80

85

90

95

100 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. bird . bird picker's hands . picker's hands . picker's hands . picker's hands . picker's hands . picker's hands . picker's hands . bird . bird . bird . . bird bird . bird . sorter's hands . . picker's hands sorter's hands . picker's hands . picker's hands . picker's hands . picker's hands . picker's hands . sorter's hands . sorter's hands . bird . bird . fruit from packing line . fruit from packing line . picker's hands . picker's hands . picker's hands . bird . bird . picker's hands . picker's hands . bird . bird . bird . picker's hands . sorter's hands . sorter's hands . bird . picker's hands . picker's hands . bird . picker's hands . bird . bird . bird . picker's hands . bird .

Figure 2.2. Dendrogram showing the relationship between E. coli isolates recovered from pre- and postharvest fruit, birds, and workers’ hands in cherry orchards. Each isolate was genotyped using the ERIC-PCR.

Figure 2.3. Photograph of a cherry picker’s hands taken approximately 10 minutes after the start of harvest. Note the accumulation of debris and soiling, and the bandaged left index finger.

50

The Origin and Spread of Human Pathogens in Fruit Production Systems

51

Conclusions Complete control cannot be achieved over the numerous environmental variables that influence the microbiological quality of fruit produced in open agricultural environments. It is also unlikely that a single risk-based control strategy can accommodate all fruit varieties, given the variability in fruit production systems. The risk of contamination with human pathogens must be addressed through an understanding of salient sources of such hazards, by the identification of means for their transport to the production site and the recognition of factors that favor their dissemination through a crop. Deficiencies in knowledge about the mechanisms that underlie these events clearly require further study.

References Abadias M, Cañamás TP, Asensio A, Anguera M, Viñas I. 2006. Microbial quality of commercial “Golden Delicious” apples throughout production and shelf-life in Lleida (Catalonia, Spain). International Journal of Food Microbiology 108:404–409. Alam MJ, Zurek L. 2004. Association of Escherichia coli O157:H7 with houseflies on a cattle farm. Applied and Environmental Microbiology 70(12):7578–7580. al-Dagal M, Fung DY. 1990. Aeromicrobiology—a review. Critical Reviews in Food Science and Nutrition 29(5):333–340. Annous BA, Solomon EB, Cooke BH, Burke A. 2005. Biofilm formation by Salmonella spp. on cantaloupe melons. Journal of Food Safety 25(4):276–287. Bagdasaryan G. 1964. Survival of viruses of the entero-virus group (poliomyelitis, echo, cocksakie) in soil and on vegetation. Journal of Hygiene, Epidemiology, Microbiology and Immunity (8):497–505. Bidawid S, Farber JM, Sattar SA. 2000. Contamination of foods by food handlers: Experiments on hepatitis A virus transfer to food and its interruption. Applied and Environmental Microbiology 66:2759–2763. Bitton G, Farrah SR, Ruskin RH, Butner J, Chou YJ. 1983. Survival of pathogenic and indicator organisms in groundwater. Ground Water 21:405–410. Brandl, M. 2006. Fitness of human enteric pathogens on plants and implications for food safety. Annual Review of Phytopathology 44:17–26. CCME (Canadian Council of Ministers of the Environment). 1987. Canadian Water Quality Guidelines. Water Quality Branch, Inland Waters Directorate, Environment Canada, Ottawa. Chai LC, Robin T, Ragavan UM, Gunsalam JW, Abu Bakar F, Ghazali FM, Radu S, Kumar MP. 2007. Thermophilic Campylobacter spp. in salad vegetables in Malaysia. International Journal of Food Microbiology 117(1):106–111. Courtenay M, Ramirez L, Cox B, Han I, Jiang X, Dawson P. 2005. Effects of various hand hygiene regimes on removal and/or destruction of Escherichia coli on hands. Food Science and Technology 5:77–84. De Roever C. 1999. Microbiological safety evaluations and recommendations on fresh produce. Food Control 9(6):321–347. Espinoza-Medina IE, Rodrigues-Leyva FJ, Vargas-Arispuro I, Islas-Osuna MA, Acedo-Felix E, MartinezTellez MA. 2006. PCR identification of Salmonella: Potential contamination sources from production and postharvest handling of cantaloupes. Journal of Food Protection 69(6):1422–1425. Gagliardi JV, Millner PD, Lester G, Ingram D. 2003. On-farm and postharvest processing sources of bacterial contamination to melon rinds. Journal of Food Protection 66(1):82–87. Garcia, L, Henderson, J, Fabri, M, Oke, M. 2006. Potential sources of microbial contamination in unpasteurized apple cider. Journal of Food Protection 69(1):137–144. Garcia-Villanova Ruiz B, Galvez Vargas R, Garcia-Villanova R. 1987. Contamination on fresh vegetables during cultivation and marketing. International Journal of Food Microbiology 4:285–291. Gibbons, A. 2006. Ancient figs push back origin of plant cultivation. Science 2 312(5778):1292. Gibbs DS, Anderson GL, Beuchat LR, Carta LA, Williams PL. 2005. Potential role of Diploscapter sp. strain LKC25, a bacterivorous nematode from soil, as a vector of food-borne pathogenic bacteria to preharvest fruits and vegetables. Applied and Environmental Microbiology 71(5):2433–2437.

52

Section I. Microbial Contamination of Fresh Produce

Herwaldt BL, Ackers ML. 1997. An outbreak in 1996 of cyclosporiasis associated with imported raspberries. New England Journal of Medicine 336:1548–1556. Holt PS, Mitchell BW, Gast RK. 1998. Airborne horizontal transmission of Salmonella enteritidis in molted laying chickens. Avian Diseases 42(1):45–52. Hussain I, Mahmoud MS, Akhtar M, Khan A. 2007. Prevalence of Campylobacter species in meat, milk and other food commodities in Pakistan. Food Microbiology 24(3):219–222. Islam M, Doyle MP, Phatak SC, Millner P, Jiang X. 2004. Persistence of enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. Journal of Food Protection 67:1365–1370. ———. Survival of Escherichia coli O157:H7 in soil and on carrots and onions grown in fields treated with contaminated manure composts or irrigation water. Food Microbiology 22:63–70. Izumi H, Tsukada Y, Poubol J, Hisa K. 2008. On-farm sources of microbial contamination of persimmon fruit in Japan. Journal of Food Protection 71(1):52–59. Janisiewicz WJ, Conway WS, Brown MW, Sapers GM, Fratamico P, Buchanan RL. 1999. Fate of Escherichia coli O157:H7 on fresh cut apple tissue and its potential for transmission by fruit flies. Applied and Environmental Microbiology 65:1–5. Keraita B, Konradsen F, Dreschsel P, Abaidoo RC. 2007. Effect of low-cost irrigation methods on microbial contamination of lettuce irrigated with untreated wastewater. Tropical Medicine and International Health 12:15–22. Kramer A, Rudolph P, Kampf G, Pittet D. 2002. Limited efficacy of alcohol based hand gels. The Lancet 359:1489–1490. Mansfield A, Gajadhar A. 2004. Cyclospora cayetanensis, a food- and waterborne coccidian parasite. Veterinary Parasitology 126:73–90. Materon LA, Martinez-Garcia M, McDonald V. 2007. Identification of sources of microbial pathogens on cantaloupe rinds from pre-harvest to post-harvest operations. World Journal of Microbiology and Biotechnology 23:1281–1287. Michaels B, Gangar V, Lin C-M, Doyle M. 2003. Use limitations of alcoholic instant hand sanitizer as part of a food service hand hygiene program. Food Service Technology 3(2):71–80. Michaels B, Keller C, Blevins M, Paoli G, Ruthman T, Todd E, Griffith CJ. 2004. Prevention of food worker transmission of foodborne pathogens: risk assessment and evaluation of effective hygiene intervention strategies. Food Service Technology 4(1):31–49. Mohr AJ. 1997. Fate and transport of microorganisms in air. ASM Press, Washington, D.C. Montville R, Chen YH, Schaffner DW. 2002. Risk assessment of hand washing efficacy using literature and experimental data. International Journal of Food Microbiology 73(2–3):305–313. Mota P, Rauch CA, Edberg SC. 2000. Microsporidia and Cyclospora: epidemiology and assessment of risk from the environment. Critical Reviews in Microbiology 26:69–90. Natvig E, Ingham SC, Ingham BH, Cooperband LR, Roper TR. 2002. Salmonella enterica serovar Typhimurium and Escherichia coli contamination of root and leaf vegetables grown in soils with incorporated bovine manure. Applied and Environmental Microbiology 68:2737–2744. Ness AR, Powles JW. 1997. Fruit and vegetables, and cardiovascular disease: A review. International Journal of Epidemiology 26 (1):1–13. Neimann J, Engberg J, Mølback K, Wegener HC. 2003. A case-control study of risk factors for sporadic campylobacter infections in Denmark. Epidemiology and Infection 130:353–366. Nicholson FA, Groves SJ, Chambers BJ. 2005. Pathogen survival during livestock manure storage and following land application. Bioresearch Technology 96:135–143. Pillai SD, Ricke SC. 2002. Bioaerosols from municipal and animal wastes: background and contemporary issues. Canadian Journal of Microbiology 48:681–696. Powell KL, Taylor RG, Cronin AA, Barrett MH, Pedley S, Sellwood J. 2003. Microbial contamination of two urban sandstone aquifers in the U.K. Water Research 37(2):339–352. Riordan DCR, Sapers GM, Hankinsson TR, Magee M, Mattrazzo AM, Annous BA. 2001. A study of U.S. orchards to identify potential sources of Escherichia coli 0157:H7. Journal of Food Protection 64(9):1320–1327. Rosas I, Salinas E, Yela A, Calva E, Eslave C, Cravioto A. 1997. Escherichia coli in settled-dust and air samples collected in residential environments in Mexico City. Applied and Environmental Microbiology 63(10):4093–4095.

The Origin and Spread of Human Pathogens in Fruit Production Systems

53

Sewell AM, Farber JM. 2001. Foodborne outbreaks in Canada linked to produce. Journal of Food Protection 64:1863–1877. Sivapalasingam S, Friedman CR, Cohen LA, Tauxe RV. 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. Journal of Food Protection 67(10):2342–2353. Sproston EL, Macrae M, Ogden ID, Wilson MJ, Strachan NJC. 2006. Slugs: Potential Novel Vectors of Escherichia coli O157. Applied and Environmental Microbiology 72(1):144–149. Steele M, Odumeru J. 2004. Irrigation water as a source of foodborne pathogens on fruit and vegetables. Journal of Food Protection 67:2839–2849. Takachi R, Inoue M, Ishihara J, Kurahashi N, Iwasaki M, Sasazuki S, Iso H, Tsubono Y, Tsugane S. 2008. Fruit and vegetable intake and risk of total cancer and cardiovascular disease. American Journal of Epidemiology 167(1):59–70. Takkinen J, Nakari U-M, Johansson T, Niskanen T, Siitonen A, Kuusi M. 2005. A nationwide outbreak of multiresistant Salmonella Typhimurium var. Copenhagen DT104B infection in Finland due to contaminated lettuce from Spain, May 2005. Eurosurveillance Weekly 10, http://www.eurosurveillance.org/ ew/2005/050630.asp (accessed on 1/20/08). Wallace JS, Cheasty T, Jones K. 1997. Isolation of verocytotin-producing Escherichia coli O157:H7 from wild birds. Journal of Applied Microbiology 82:399–404. WHO (World Health Organization). 1989. Health guidelines for the use of wastewater in agriculture and aquaculture. Technical Report Series No.778. World Health Organisation. Xu J, Liu Q, Jing H, Pang P, Yang J, Zhao G, Li H. 2003. Isolation of Escherichia coli O157:H7 from Dung Beetles Catharsius molossus. Microbiology and Immunology 47(1):45–49.

3 Internalization of Pathogens in Produce Elliot T. Ryser, Jianjun Hao, and Zhinong Yan

Introduction Washing fresh produce in various sanitizers is only marginally effective in reducing populations of pathogenic and spoilage microorganisms, with complete elimination impossible without compromising product quality (Aruscavage and others 2006; Beuchat 2004, 2006; Beuchat and Ryu 1997). This failure is attributed to a wide range of topographical surface characteristics inherent to fresh produce, including 1) waterrepelling waxes (e.g., apples, green peppers), cracks or crevices (e.g., cantaloupe, various seeds), and natural openings (e.g., stomata, hydathodes, nectarthodes, lenticels, stem scars, calyx) (Beuchat 2002; Takeuchi and Frank 2000); 2) biofilm formation on leaves and roots (Warriner and others 2003b; Lapidot and others 2006); and 3) bacterial infiltration through cut edges or wounds on produce (Beuchat 2002; Seo and Frank 1999; Mendonca 2005), all of which minimize pathogen exposure to sanitizers (Beuchat 2002, 2006; Aruscavage and others 2006). The more than 600 outbreaks traced to fresh produce since 1990 (DeWaal and others 2006) have collectively increased the attention now given to internalization of bacterial foodborne pathogens, particularly Salmonella and Escherichia coli O157:H7, in fresh produce. Internalization of endophytic bacteria important in plant pathology and microbial ecology has been recognized since the 1870s and well studied since 1940 (Hallmann and others 1997; Sturz and others 2000; Rosenblueth and MartínezRomero 2006; Ryan and others 2008) with several authoritative reviews having been written on internalization of various foodborne pathogens, including Salmonella and E. coli O157:H7 into fruits and vegetables by various routes and their ability to persist internally (Aruscavage and others 2006; Beuchat 2006; Brandl 2006; Mendonca 2005; Solomon and others 2006; Stone and others 2000; Warriner 2005). This internalization as well as the aforementioned factors leads to enhanced survival of foodborne pathogens after treatment with chemical sanitizers. An understanding of the plant rhizosphere, phyllosphere, and plant-microbe interactions is key to the development of potential strategies to minimize internalization of human foodborne pathogens in plants. These topics are discussed in greater detail, beginning with the definition of internalization.

Bacterial Endophytes Bacteria residing internally in plant tissue are known as endophytes, with these organisms found in almost all plant species. Endophytic bacteria were first recognized during the 1870s when Pasteur and others identified bacteria within asymptomatic 55

56

Section I. Microbial Contamination of Fresh Produce

plants (Hollis 1951). Since the 1940s, our knowledge of indigenous endophytes in various plant tissues has greatly increased (Hallmann and others 1997b), with endophytic bacteria important to the field of plant pathology identified in tomatoes, cucumbers, and beans as early as the 1960s (Samish and others 1961, 1963a,b; Leben 1972; Meneley and Stanghellini 1974). These bacterial endophytes, which belong to many different genera including Agrobacteria, Pseudomonas, Bacillus, Enterobacter, Burkholderia, Erwinia, Klebsiella, and Microbacteria, have the potential to improve plant growth, induce systemic resistance against other plant pathogens, enhance phytoremediation, and improve soil fertility (Hallmann and others 1997b; Rosenblueth and Martinez-Romero 2006; Sturz and others 2000; Ryan and others 2008). Conceptually, Kado (1992) considered bacterial endophytes as “bacteria that reside within living plant tissue without doing substantive harm or gaining benefit other than securing residency”. Although shared by several others (Wilson 1995; Schulz and Boyle 2006), this definition excludes any symbiotic relationship with the host plant or other beneficial effects such as growth promotion and induced systemic resistance against plant pathogens (Zehnder and others 2001; Sturz and others 2000). Functionally, Hallmann and others (1997b) considered those bacteria that could be isolated from surface-sterilized plant tissue or extracted from inner plant tissue as endophytic. Now the widespread use of various fluorescent biomarkers for labeling bacteria allows their direct visualization within plant tissue using various advanced microscopy techniques, with such internalized bacteria now also being functionally identified as endophytic (Rosenblueth and Martinez-Romero 2006). Based on these definitions, Salmonella and E. coli O157:H7 can also be considered endophytic because these foodborne pathogens can become internalized in certain types of produce without harming the plant.

Sources of Human Endophytic Pathogens Produce is prone to contamination from microorganisms, including human bacterial pathogens, throughout the entire field-to-fork continuum (Fig. 3.1). Major sources of human endophytic pathogens include seed, soil, manure, and irrigation water (Beuchat 2006; Brandl 2006; see also chapters 3 and 4). Irrigation water can be contaminated by field flooding, a river passing through dairy farms, and recycled irrigation water. During postharvest processing, flume tank water can also serve as an important source of potentially endophytic bacteria.

Bacterial Attachment and Colonization Many human bacterial pathogens are well adapted for survival in both soil and water and can persist in the plant spermosphere (germinated seed), rhizosphere (roots), and phyllosphere (leaves) together with the inherent microflora that include mutualists and commensals (Danhorn and Fuqua 2007). Once attached, these bacteria can form biofilms comprised of assemblages or aggregates of microorganisms embedded in a matrix of exopolymers that adhere to surfaces, including those found on fruits and vegetables (Annous and others 2005; Morris and Monier 2003; Morris and others 1997). Bacteria within these biofilms interact with plant tissues through various adhesions including polysaccharides and surface proteins that can in turn enhance attachment and colonization. Recognition between lectins and their cognate carbohydrates

Internalization of Pathogens in Produce

Contamination Sources

57

Portals of Entry

Spray irrigation Drip irrigation Manure

Preharvest Fungi Protozoa Insects Nematodes

Leaves Seeds Stomatas Trichomes Wounds Hydathodes Fungi Protozoa Flowers Insects Nectarthods Nematodes Floral tubes Fruits Lenticels

Mechanical handling Cutting Trimming Coring

Wounds Bruises

Harvest Cooling Ice Vacuum

Stem scar Calyx

Human handling

Processing Conveying Fluming Shredding Dicing Drying

Cut surfaces Bruises

Postharvest

Stem scar Calyx

Mechanical damage Human handling

Figure 3.1. Modes of transmission and portals of entry for bacterial pathogens in fruits and vegetables.

is a common means of specificity. Colonizing bacteria develop biofilms and promote plant-bacteria interactions via cell-cell communication. When present on fresh produce, these biofilms that can harbor foodborne pathogens are typically more difficult to remove (Fig. 3.2) (Fett 2000) and provide a protective barrier, thereby decreasing the efficacy of sanitizers (Morris and Monier 2003; Zottola 1994).

Impact of Plant Stress on Internalization Plants stressed by drought, temperature extremes, insect infestations, and other adverse growing/environmental conditions become weakened (Marcais and Breda 2006) and

Figure 3.2. Scanning electron micrographs of commercially grown alfalfa sprouts. (A) An incipient biofilm on the surface of a cotyledon, note the strands (arrows) of material attaching the bacteria to each other and to the plant surface; (B) two biofilms (arrows) on a cotyledon surface, note the continuous matrix of material between the covering and the bacterial cell; (C) a biofilm on the surface of a hypocotyl, note the mixed bacterial cell morphologies; (D) a biofilm on the surface of a primary root.

58

Internalization of Pathogens in Produce

59

are therefore more prone to invasion by both plant and human pathogens. Some of the best examples come from the field of plant mycology with stressed corn, wheat, and other crops being most susceptible to infections from a wide range of mycotoxigenic molds including Aspergillus, Fusarium, Penicillium, and Claviceps (Papendic and others 1971; Payne and others 1982). When Wilson and others (1999) assessed the population dynamics of plant pathogenic and nonpathogenic strains of Pseudomonas syringae as well as the nonpathogenic species Pantoea agglomerans, Stenotrophomonas maltophilia, and Methylobacterium organophilum in the bean phyllosphere, the plant pathogens survived better than the nonpathogens on leaves as well as inside these plants during environmentally stressful conditions. Once invaded, various enzymes such as cellulase, amylase, and pectinase commonly produced by plant pathogens can degrade the cell wall and other tissues of plants. Syringomycin, one such enzyme produced by Pseudomonas syringae, forms pores in the plant cell membrane leading to the release of cell contents (Alfano and Collmer 1996). During the degradation process, new carbon sources are released that can foster bacterial growth (Aruscavage and others 2006).

Portals of Entry Bacteria can become internalized throughout the farm-to-fork continuum (Fig. 3.1). The most common portals of entry include the stomata, lenticels, lateral roots, germinating radicles, trichomes, wounds, areas of decay, and stem scars (Aruscavage and others 2006; Hallmann and others 1997b; Sturz and others 1999, 2000), with similar invasive routes shown for human pathogens (Kenny and others 2001). Stomata, lenticels, and stem scars are among the most important portals of entry for both plant endophytic bacteria (Hallman and others 1997b; Lamb and others 1996) and foodborne pathogens, including Salmonella (Aruscavage and others 2006) and E. coli O157:H7 (Itoh and others 1998; Seo and Frank 1999; Takeuchi and Frank 2000) with the calyx, stem end, and floral tubes of fruits being most vulnerable to infiltration (Buchanan and others 1999; Burnett and others 2000; Kenney and others 2001). Preharvest Endophytic bacteria frequently penetrate plant tissue via root hair cells or junctions between the root hairs and adjacent epidermal cells (Hallmann and others 1997b; Sturz and others 2000) followed by transport through the vascular system (Lamb and others 1996). Enterobacter asburiae (a plant endophytic bacterium) reportedly entered and spread in cotton through active uptake (Quadt-Hallmann and others 1997). In this case, penetration was not dependent on any wounds. The authors suggested that this organism may have degraded cellulose and then migrated through the plant via intercellular spaces. Wax cutin, which fills the lenticels and seals any cracks, minimizes bacterial attachment to the phyllosphere and provides another natural barrier to bacterial penetration (Aruscavage and others 2006; Burnett and others 2000; Kenney and others 2001). However, wounds resulting from insect, fungal, or nematode infestations, cuts, bruises, or other types of physical damage to the waxy cuticle can easily lead to bacterial harborage sites along with infiltration of the underlying tissue.

60

Section I. Microbial Contamination of Fresh Produce

Postharvest Immersing warm produce in cooler water during fluming or washing is another common means for bacterial internalization. The decrease in produce temperature that follows allows the atmospheric and hydrostatic forces on the immersed produce to equilibrate with the internal pressure, thus facilitating ingress of water along with uptake of any bacteria present (Bartz and Showalter 1981; Buchanan and others 1999). Such internalization of E. coli O157:H7 has been shown for apples, with much of the penetration occurring through the blossom end when the aqueous bacterial suspension was cooler than the fruit (Buchanan and others 1999; Burnett and others 2000). Given its close association with the plant vascular system, stem scar tissue is also highly prone to infiltration by endophytic bacteria as well as E. coli O157:H7 and Salmonella (Bartz and Showalter 1981; Eblen and others 2004). The extent of internalization is influenced by age, with fresh stem scars on tomatoes more vulnerable to infiltration than older stem scars, and green and pink tomatoes more susceptible to water uptake than red fruit (Bartz and Showalter 1981).

Bacterial Movement in Plants As early as the 1980s, internalized plant endophytes such as growth-promoting rhizobacteria and various plant pathogens were shown to migrate through the plant vascular system (Hallmann and others 1997b; Kluepfel 1993; Lamb and others 1996; Yan and others 2003). Lamb and others (1996) detected Rhizobacterium spp. and Pseudomonas aureofaciens in interior aerial tissue from 16 types of monocotyledonous (e.g., corn, wheat) and dicotyledonous plants (e.g., broccoli) grown from inoculated seed with direct vascular transport from the roots suggested. However, passive uptake of E. asburiae was not reported in cotton plants (Quadt-Hallmann and others 1997). Human pathogens that infiltrate plant tissue can either remain localized or move systemically (Solomon and others 2002b; Wachtel and others 2002a). Pseudomonas aeruginosa strain PA14, a human opportunistic pathogen that infects Arabidopsis, reportedly moved long distances through the vascular parenchyma (Plotnikova and others 2000). Salmonella enterica and E. coli O157:H7 were also shown to internalize and move long distances through the Arabidopsis vascular system, resulting in whole plant contamination in the absence of microbial competitors (Cooley and others 2003). However, both pathogen populations significantly decreased when Arabidopsis was grown in nonsterile soil containing E. asburiae, with invasion through the lateral root junctions seen using fluorescently labeled strains of S. enterica and E. coli O157:H7. Movement was eliminated and invasion decreased when nonmotile mutants of S. enterica were used (Cooley and others 2003), suggesting that internalization is an active process based partially on flagellar motility.

Methods for Examining Bacterial Internalization Proper methodology is critical in confirming bacterial internalization. Hallmann and others (1997b) provided a thorough review of methods to study bacterial endophytes. The key contents from this review paper and some newly developed methods reported from literature are now summarized below.

Internalization of Pathogens in Produce

61

Surface Disinfection and Maceration As reviewed by Hallmann and others (1997b), the first methods to identify endophytes in plants were based on various surface decontamination strategies. Some commonly used chemical sterilants include sodium hypochlorite (NaOCl, 2–10%), ethanol (70– 90%), hydrogen peroxide (H2O2, 3%), peracetic acid (0.03%), and mercuric chloride (HgCl2, 0.1%), with two or more of these sterilants sometimes used in combination. In order to increase disinfection efficacy, these aqueous solutions can be supplemented with detergents such as Tween 20 or 80 to reduce the surface tension on plant surfaces. After a treatment period of 10 sec to 10 min—depending on the chemical disinfectant used—the plant tissue is repeatedly rinsed in sterile water or buffer to remove any residual disinfectant and then dried with a sterile tissue or in air under ambient conditions. Alternatively, plants with tap roots and stems can be dipped in ethanol and flamed for surface disinfection (Yan and others 2003). Selection of disinfectants and treatment times depends on the plant species (e.g., woody, root crop, leafy green, fruit, berry), maturity (e.g., seedlings, young or old plants), portion of the plant (e.g., roots, stems, leaves, seeds) and surface topography (e.g., waxy, smooth, wrinkled). Therefore, surface disinfection needs to be optimized for each specific situation. Ideally, surface disinfection will eliminate all epiphytic bacteria as confirmed by sterility testing (Hallmann and others 1997b), with the entire endophytic population remaining both viable and recoverable after treatment. However, this end result is seldom, if ever, achieved, with all of these treatments yielding varying percentages of surviving epiphytes and inactivated endophytes. Two reviews (Lodewyckx and others 2002; Stone and others 2000) highlighted the methods used to isolate and characterize endophytic bacteria from different plant species. With respect to foodborne pathogens, internalization of Salmonella and E. coli O157:H7 in fresh produce has been demonstrated using various surface sterilization methods including sodium hypochlorite for apples (Buchanan and others 1999), ethanol flaming for tomatoes (Ibarra-Sánchez and others 2004), HgCl2 and ethanol for sprouts (Dong and others 2003; Itoh and others 1998), and sodium hypochlorite for lettuce and spinach (Jablasone and others 2005; Johannessen and others 2005; Solomon and others 2002). To quantify endophytic bacteria, surface-sterilized plant tissue is macerated either manually (e.g., mortar and pestle) or mechanically (e.g., blender, stomacher, tissue pulverizer) in a buffer solution, with serial dilutions and then plated to appropriate media. To avoid the disadvantage from strong surface disinfectants that may penetrate and kill endophytes, additional alternatives are needed. Interestingly, some potentially useful methods routinely used to recover bacteria from food contact surfaces including agitation with or without glass beads, votexing, sonication, and pulsification have not been widely examined and warrant further investigation. Extraction of Endophytes Alternatives to surface disinfection include vacuum and pressure extraction. The vacuum technique used in plant pathology to extract fastidious bacteria from xylem is also useful for recovering endophytes from grapevines and citrus fruits. The Scholander pressure bomb method (used to measure plant respiration in aliquots forced out from plant tissue under high pressure) was used to study endophytes in cotton, soybeans, and beans (Hallmann and others 1997a). Compared to other

62

Section I. Microbial Contamination of Fresh Produce

maceration techniques, the pressure bomb method detected bacteria colonizing the vascular system, while maceration mainly recovered endophytes residing in the root cortex. Hence, a combination of methods is recommended to best understand the full colonization pattern of endophytic bacteria. Due to the high pressures involved, the pressure bomb technique favors hard plant tissue and is not suitable for fruits such as tomatoes and cucumbers (Hallmann and others 1997b). It has not yet been tested on root crops or leafy greens. Centrifugation after surface decontamination—another alternative to maceration (Hallmann and others 1997b)—is often used to collect intercellular fluid from plant tissue. Compared to vacuum and pressure extraction, the centrifugation method is suitable for soft plant tissue. Direct tissue printing is also useful in isolating endophytic bacteria. Radish sprout hypocotyls, grown from seeds artificially inoculated with E. coli O157:H7, were surface-sterilized by treating with 80% ethanol for 4–5 sec and 0.1% HgCl2 for 4 or 10 min. The 1.5 cm sections were split and pressed onto agar plates for 10 min and incubated thereafter. Presence of colonies on the agar surface indicated that E. coli O157:H7 infiltrated radish sprouts (Itoh and others 1998). In another study, red bean seedlings were stem-inoculated at the soil line with a fluorescently labeled strain of E. asburiae JM22, a systemic endophyte isolated from cotton. After cutting the stems at different distances from the root, the cut surfaces were squeezed and pressed onto trypticase soy agar containing 100 ppm ampicillin to recover JM22 from stem tissue 10 cm above the soil line (Yan unpublished data). Endophytic bacteria have been mainly studied by culturing. However, several noncultural molecular approaches also have been used to examine the interactions between bacterial endophytes and plants (Hallmann and others 1997b; Kluepfel 1993; Rosenblueth and Martinez-Romero 2006; Ryan and others 2008). For example, Araújo and others (2002) studied the interaction between endophytic bacterial communities and Xylella fastidiosa (the causative agent of citrus variegated chlorosis) using cultivation-based plating techniques and a cultivation-independent method involving PCRgenerated 16S rRNA gene (rDNA) fragments and denaturing gradient gel electrophoresis (DGGE). DGGE analysis of 16S rRNA gene fragments amplified from total plant DNA resulted in several bands that matched those from the bacterial isolates, indicating that DGGE profiles can be used to detect some endophytic bacteria in citrus plants. However, some bands had no match with any isolate, suggesting the occurrence of other nonculturable or undetected endophytic bacteria.

Labeling of Endophytic Bacteria Endophytes can be detected and quantified using the isolation procedures described above. However, to understand the mechanisms of entry, localization, survival, and interactions with other indigenous microorganisms, the specific endophytic organism of interest must be introduced into its host plant or other plant species. Several labeling methods, which are based on antibiotic resistance or fluorescent tagging used to differentiate the target endophyte from other background microorganisms, are described next. The oldest and most common marker used in conjunction with conventional plating is based on the selection of spontaneous antibiotic-resistant derivatives of the wild-type bacterial strain. However, two or more antibiotics may be needed to

Internalization of Pathogens in Produce

63

differentiate the endophyte from a more highly resistant background flora. After demonstrating that the derived antibiotic resistant strain is similar to the wild type strain in terms of plant surface adhesion, biofilm formation, growth rate, survivability, and other important characteristics, the labeled strain is introduced into the plant or environment and recovered using a selective medium that contains the appropriate antibiotic(s) (Kluepfel 1993; Yan and others 2003). This simple, rapid, sensitive, and inexpensive technique has been widely successful. Alternatively, serological techniques based on antibodies raised against bacterial cell wall protein components or whole bacterial cells have been used to detect endophytes. After fluorescently tagging the specific antibody with fluorescein isothiocyanate (FITC) (Mahaffee and others 1997) or colloidal gold (Quadt-Hallmann and others 1997), the endophyte can be viewed using fluorescence or electron microscopy. This method was further improved by combining immunofluorescence with pour plating in a technique termed immunofluorescent colony staining (IFC) (Kluepfel 1993). With respect to foodborne pathogens, Itoh and others (1998) used a fluorescein-conjugated goat antirabbit immunoglobulin to observe surface colonization and internalization of E. coli O157:H7 in radish sprouts. Currently, the most common method is to introduce a reporter gene into the organism of interest. The gene encoding the green fluorescence protein (GFP, originally isolated from the jellyfish Aequoria victoria) has become extremely useful with this fluorescent protein tag having been introduced into bacteria, yeast, slime molds, plants, drosophila, zebra fish, and mammalian cells. When used in combination with confocal laser scanning microscopy (CLSM), this GFP marker allows direct in vivo observation of the fluorescently tagged organism both on and in a wide range of plant tissues, including apples, tomatoes, sprouts, lettuce, and spinach. Glucuronidase (GUS)—another useful marker to visualize bacteria in plants, is based on the discovery that gene fusions comprising the β-glucuronidase gene can be effectively expressed in many organisms, leading to production of active β-glucuronidase. This enzyme, in turn, cleaves a chromagenic substrate that can be directly observed as a green/blue precipitate in bacteria. Warriner and others (2003a,b) used a GUS marker to confirm internalization of E. coli in spinach and bean sprouts. Other gene markers—such as LacZ (β-galactosidase gene), inaZ (ice nucleation gene), and Lux—have been used in plant pathology as described by Kluepfel (1993). However, these methods generally have not been used to study foodborne bacteria in plants. Autoradiography has also proven useful in assessing plant-microbe interactions (Dumont and others 2006; Hallmann and others 1997b). After growing in a broth medium amended with isotopes such as 14C, 15N, or 32P, endophytic bacteria can be introduced into plants and detected using autoradiography. Again, this isotopic labeling method has not been used to study foodborne pathogens in plants.

Methods for Introducing Endophytic Bacteria into Plants After proper labeling, various approaches can be used to introduce the target organism into plants. One study by Musson and others (1995) compared 12 different methods to introduce 21 bacterial endophytes into cotton seedlings. The endophytes were delivered by vacuum infiltration, soaking, methylcellulose coating, soil drenching, root tip pruning, stem injection, stem stab, and foliar spray, with four methods also

64

Section I. Microbial Contamination of Fresh Produce

used in combination. Overall, the delivery rate was strain-method–specific with some strains yielding higher endophyte populations than others using the same delivery method. Internalization was also strain-dependent meaning that the endophytes varied in their ability to infiltrate the plant. In a similar study comparing five different delivery methods—seed inoculation, soil drenching, foliar spray, pruned-root dip, and a combination of seed inoculation and soil drenching—the pruned-root dip method was most effective in introducing endophytic bacteria into maize (Bressan and Borges 2004). Using seed inoculation as the delivery method, Bacillus subtilis—a potential biological control agent against black rot disease in cabbage—infiltrated the roots and moved into the stems and leaves (Wulff and others 2003). With respect to transplanting systems, mixing bacteria into a soil-free mix before seed sprouting provided an ideal means to deliver endophytes into plants (Yan and others 2003). The above methods developed in plant pathology can be similarly adapted to study internalization of human pathogens in produce. Foodborne bacteria can invade plant tissue at many points along the farm-to-fork continuum. Methods used to introduce foodborne endophytic bacteria into plants during preharvest are based on the most plausible routes of entry, including the seed, soil, irrigation water, and sewage. Contamination of sprouts typically involves submerging or spot-inoculating seeds with the target organism, leading to internal colonization during sprouting (Dong and others 2003; Itoh and others 1998; Warriner and others 2003b). For lettuce, inoculated soil and water are commonly used (Solomon and others 2002; Wachtel and others 2002a). Stem injection and flower brushing have also been used to introduce Salmonella and E. coli O157:H7 into tomatoes (Ibarra-Sánchez and others 2004; Guo and others 2002b). Root pruning, together with root-knot nematode damage, can also serve as an inoculation method (Hora and others 2005). In contrast, other methods have been used to introduce these same organisms into harvested plant material during processing. Niemira (2007) successfully used vacuum perfusion to force E. coli O157:H7 into the leaf vasculature and apoplast of spinach and lettuce, whereas Hajdock and Warriner (2007) introduced Salmonella and E. coli O157:H7 into lettuce via vacuum filtration. Infiltration by submerging produce in contaminated flume water that is colder than the product will allow bacteria to infiltrate damaged tissue and natural openings (Burnett and others 2000; Han and others 2000; Penteado and others 2004; Seo and Frank 1999). In a thorough study by Lang and others (2004), an attempt was made to standardize the dip, spot, and spray methods used to inoculate lettuce and parsley with E. coli O157:H7, Salmonella, and Listeria monocytogenes. Significantly higher populations of E. coli O157:H7 and Salmonella were recovered using dip as compared to spot or spray inoculation. Although E. coli O157:H7 and Salmonella populations recovered from spot- and spray-inoculated lettuce were not significantly different, spot-inoculation did yield significantly higher populations than spray inoculation for parsley. Significantly different numbers of L. monocytogenes were also recovered from inoculated lettuce (dip > spray > spot), indicating that these inoculation methods are both produce- and bacteria-dependent. However, bacterial internalization was not assessed in this study. Finally, two nonmicrobial approaches—namely aqueous dye solutions and fluorescent microbeads—have been used as indicators of bacterial infiltration into produce. Immersing the blossom end of intact apples in a red dye solution and a bacterial

Internalization of Pathogens in Produce

65

suspension demonstrated uptake of both the red dye and bacteria (Buchanan and others 1999) with similar results reported for intact oranges and grapefruit (Merker and others 1999). In other work, Solomon and Matthews (2005) used FluoSpheres (fluorescent microbeads) as an indicator to study E. coli O157:H7 internalization in lettuce. When mature lettuce plants were surface-irrigated with FluoSpheres, microscopic examination of the root, stem, and leaf tissue revealed internalized FluoSpheres within these portions of growing plants. No significant difference was found between the numbers of beads and E. coli O157:H7 cells internalized. Niemira (2007) subsequently confirmed these microbeads as an appropriate surrogate for uptake of E. coli O157:H7 in lettuce through the plant vascular system (Figs. 3.3, 3.4).

Figure 3.3. Romaine lettuce leaf pieces following perfusion with E. coli O157:H7 inoculum, SEM micrographs. In left image, arrows indicate bacterial cells in intercellular spaces; letter A indicates area of magnification in right image.

Figure 3.4. Romaine lettuce leaf pieces following perfusion with microbeads, SEM micrographs. Letter A indicates area of magnification in right image, showing microbeads in vascular elements.

66

Section I. Microbial Contamination of Fresh Produce

Foodborne Pathogen Interactions with Plant and Soil Microflora The soil and plant phyllosphere contain a vast array of naturally occurring bacterial epiphytes and endophytes as well as fungi and other microorganisms that directly affect the survival and persistence of human pathogens. For example, Enterobacter asburiae, an epiphyte from lettuce leaves, reportedly decreased the survival of E. coli O157:H7 on lettuce leaves, whereas Wausteria paucula allowed E. coli O157:H7 populations to increase twenty- to thirtyfold (Cooley and others 2006). In other work, L. monocytogenes grew in apple wounds infected with Glomerella cingulata, but decreased in similar wounds colonized by Penicillium expansum (Conway and others 2000). Microbial competition readily occurs in the rhizosphere and plant exudates, which provide various carbon and nitrogen substrates for bacterial growth (Cooley and others 2006). According to Cooley and others (2003), populations of S. enterica and E. coli O157:H7 decreased when Arabdopsis thaliana plants were grown in both unautoclaved and amended soil, suggesting competition from indigenous micoflora. This inhibition was further attributed to a strain of E. asburiae isolated from A. thaliana that suppressed the growth of both pathogens. In other work, one Pseudomonas fluorescens isolate from the lettuce rhizosphere inhibited E. coli O157:H7 in vitro (Johannessen and others 2005). Endophytic fungi can also serve as a vector for internalization of enteric pathogens (Artursson and Jansson 2003; Lee and Cooksey 2000). A primary means by which plant pathogens aid in the internalization of human pathogens is by disrupting plant tissues, which in turn release nutrients and water with a concomitant increase in pH to values near neutrality that favor bacterial growth (Aruscavage and others 2006; Leverentz and others 2001; Riordan and others 2000). Greater colonization and infiltration is seen in higher pH wounds and bruises that have released additional nutrients for bacterial growth (Wachtel and others 2002a) with deep mechanical wounds, cuts, and punctures also leading to greater internalization. Although the concentration and type of nutrients leached varies among plant species (Dingman 2000), the released nutrients will not always enhance the growth of foodborne pathogens. Reinders and others (2001) showed that caffeic acid released from apple wounds inhibited E. coli O157. Growth of enteric bacterial pathogens in fresh produce is often greater in the presence of certain fungal (e.g., Botrytis, Rhizopus— cause of grey mold and black soft rot, respectively) or bacterial plant pathogens (Erwinia carotovora—cause of bacterial soft rot) (Wells and Butterfield 1997, 1999), which indirectly provide important nutrients for foodborne pathogens to survive and internalize. According to Brandl (2006) Salmonella populations were tenfold higher when potato, carrot, and bell pepper disks were coinoculated with several soft-rot pathogens than when inoculated with Salmonella alone. Although Glomerella cingulata (a pathogen causing anthracnose disease in strawberries) increased E. coli O157:H7 populations 3 logs in wounded apple tissue (Riordan and others 2000), this response is by no means universal for all plant pathogens with E. coli O157:H7 growth not enhanced in the presence of Pseudomonas syringae (Hora and others 2005). Many insects and nematodes can serve as internalization vectors for foodborne pathogens (Janisiewicz and others 1999; Sasaki and others 2000; Sela and others 2005). Caenorhabditis elegans (Alfano and Collmer 1996) and Meloidobyne

Internalization of Pathogens in Produce

67

incognita, a plant nematode, were associated with internalization of Salmonella in cantaloupe, with Salmonella protected after being ingested by the nematode (Caldwell and others 2003). Insects can also harbor human pathogens both externally and internally (Waterfield and others 2004). However, according to Hora and others (2005), the root-knot nematode, Meloidogyne hapla, did not enhance internalization of E. coli O157:H7 in spinach.

Evidence for Internalization of Foodborne Pathogens in Produce Compared to the internalization of plant pathogens, studies addressing the infiltration of fruits and vegetables by foodborne pathogens are of relatively recent origin, with much of this work having been conducted in response to the growing number of high profile outbreaks involving fresh produce, the first of which received widespread attention in the mid-1990s. Since this time, well over 50 studies addressing internalization of primarily Salmonella and E. coli O157:H7 in various fruits and vegetables have been reported, the results from which are summarized in the next sections. Apples A series of E. coli O157:H7 outbreaks during the mid-1990s involving fresh apple cider prompted concern over the use of dropped apples and potential internalization of pathogens. In response, Buchanan and others (1999) were first to show that E. coli O157:H7 could penetrate apples during washing. Initial dye immersion tests demonstrated dye uptake into the inner core of warm (∼22 °C) but not cold (4 °C) apples. When warm or cold apples were subsequently immersed for 20 min in a cold (4 °C) aqueous suspension containing ∼7.5 log E. coli O157:H7 CFU/ml, highest average populations were found in the outer core near the calyx and blossom ends (5.43 log CFU/g) followed by the skin (3.91 log CFU/g), inner core (3.40 CFU/g) and pulp (3.08 CFU/g), with similar results obtained for four different varieties: Golden Delicious, McIntosh, Red Delicious, and Braeburn. These findings were subsequently confirmed by Burnett and others (2000) who used CLSM to visualize a GFP-labeled strain of E. coli O157:H7 inside the floral tubes of Red Delicious apples. Once internalized, Buchanan and others (1999) also showed that a 1-min rinse in a 2000 mg/l sodium hypochlorite solution followed by a 1-min water rinse decreased E. coli populations only ten- to thousandfold, with highest numbers present in the outer core. When apples strike the ground, bruising and other forms of damage also increase the likelihood for internalization, with bacteria more readily infiltrating cracks in the wax layer, lenticels, and stomata. Using a GFP-labeled strain of E. coli O157:H7, Kenney and others (2001) reported that bruising increased pathogen migration into lenticels and decreased penetration into wax platelets compared to unbruised apples when apples at 30 °C were immersed in a 4 °C bacterial suspension. Rubbing these apples forced the pathogen from the edges of the wax platelets into crevices. Although subsequent washing decreased E. coli populations about 2 logs, those cells embedded in the wax layer were not removed. After infiltrating bruised tissue or wounds, the pH and °Brix within this microenvironment as influenced by the native microflora partially dictates the ability of E. coli O157:H7 to survive and/or grow (Dingman 2000; Janisiewicz and others 1999; Riordan and others 2001). When Seeman and others

68

Section I. Microbial Contamination of Fresh Produce

(2002) placed apples on E. coli–inoculated topsoil in a controlled outdoor setting, the organism entered the outer and inner core as well as the skin and pulp within 1 day. Apple orientation dictated the extent of internalization with highest E. coli populations found in the outer core when the apple calyx directly contacted the soil. Cantaloupe Salmonella is a frequent contaminant of cantaloupe with at least 11 such multistate outbreaks reported to the Centers for Disease Control and Prevention since 1973, 5 of which were traced to Salmonella Poona, a serovar infrequently associated with human foodborne illness (Bowen and others 2006). Although most often confined to the outer rind of cantaloupes in biofilms (Annous and others 2005; Beuchat and Scouten 2004; Richards and Beuchat 2004), Salmonella can also penetrate the stem scar tissue and wounds, particularly when these areas become colonized by certain phytopathogenic molds (Richards and Beuchat 2005a,b). In the first of two studies by Richards and Beuchat (2004), Eastern cantaloupes at 30 °C were significantly heavier after immersion in a S. Poona suspension at 4 °C compared to cantaloupes at 4 or 30 °C that were immersed in a 30 °C inoculum. Although these findings were expected based on pressure differences, Western cantaloupes were similar in weight regardless of the treatment with these findings perhaps related to their far denser surface netting leading to greater retention of surface water. When aqueous suspensions of S. Poona (∼9 log CFU/ml) at 4 and 30 °C were used for immersion, greater infiltration of S. Poona into the rind was seen using Eastern cantaloupes at 4 (5.0 log CFU/cm2) rather than 30 °C (4.56–4.74 log CFU/cm2) regardless of the water temperature. In contrast, higher numbers of salmonellae were recovered from the rinds of Western cantaloupes immersed in a 4 °C suspension, regardless of the cantaloupe temperature. However, all treatment combinations yielded similar penetration of S. Poona into the stem scar tissue of both cantaloupe varieties. The fact that these bacterial infiltration findings contradict those previously reported for apples and tomatoes is likely again related to the complex surface topography of cantaloupes. Richards and Beuchat (2005a) subsequently reported that S. Poona could also migrate to a depth of 3–4 cm in cantaloupes that had been wound-inoculated at a depth of 4 mm, with infiltration enhanced in the presence of two phytopathogenic molds: Cladosporium cladosporioides and, to a lesser extent, Penicillium expansum. Hence, consumption of damaged or decaying cantaloupe many pose an increased risk of illness. Mangos Being prone to fly larvae infestations, United States regulations require that all imported mangos be disinfested by hot water immersion, irradiation, or other proven means. Immersing mangos in 46.1 °C water for 65 to 90 min as required by the regulation leads to a positive temperature differential during subsequent cooling along with potential introduction of foodborne pathogens. In response to a 1999 salmonellosis outbreak in the United States, mangos heat-treated at 46.1 °C for 90 min were subjected to a 22 °C/10 min immersion treatment in water inoculated to contain 7 logs S. Enteritidis CFU/ml (Penteado and others 2004). Immediately after treatment, the stem ends from 5 of 6 mangos yielded internalized Salmonella. Less infiltration was seen

Internalization of Pathogens in Produce

69

at the midside and blossom end, with fruit maturity having no pronounced effect on the rate of internalization. A virtually identical study reported by Brazilian investigators three years later (Branquinoho Bordini and others 2007) confirmed the previous findings with populations of internalized Salmonella at least tenfold higher at the stem end compared to the midside or blossom end, with these findings supporting a shift toward nonthermal disinfestation methods such as ionizing irradiation for mangos. Oranges Following several widely publicized salmonellosis outbreaks traced to unpasteurized orange juice in the late 1990s, Eblen and others (2004) conducted a series of experiments to assess the ability of Salmonella and E. coli O157:H7 to infiltrate both intact and punctured oranges. When intact Florida and California oranges at 21 °C were initially immersed in a dye solution at 4 °C, 3 to 4% of the oranges took up the dye through the stem scar with no dye infiltration observed when the two temperatures were reversed, again indicating the importance of a positive temperature differential for both water and pathogen uptake. Subsequently, oranges tempered to 37 °C were spot inoculated at the stem scar with an E. coli O157:H7 or S. Enteritidis dye solution containing 107 CFU/ml and then equilibrated for 3 h at 4 and 24 °C, respectively. Overall, 2.5 to 3.0% of these oranges yielded the target pathogen, with 0.1% of the E. coli O157:H7 and 0.01% of the Salmonella inoculum internalized, giving populations of 3.82 and 2.34 log CFU/g, respectively. Introducing three evenly spaced 0.91 mm diameter/1.5 cm deep holes opposite the stem scar increased Salmonella uptake to 31% with pathogen growth observed during subsequent storage at 24 °C, presumably between the segments and pulp vesicles due to a more favorable pH. Hence, these findings confirm the potential public health hazard of processing warm, damaged, or otherwise punctured fruit for juice. Tomatoes Problems regarding tomatoes with premature bacterial soft rot of tomatoes exiting commercial dump tanks during the 1970s prompted several studies that confirmed the ability of water and subsequently spoilage bacteria to infiltrate tomatoes through cracks that formed when warm fruit was immersed in colder water (Segall and others 1977). Bartz and Showalter (1981) clearly demonstrated that an increase in tomato weight resulting from exposure to a negative fruit/water temperature differential led to increased infiltration by several spoilage bacteria, particularly in the scar stem area of green tomatoes as was first reported 28 years earlier by Samish and others (1963a), with increased bacterial penetration seen in more deeply submerged tomatoes as a result of increased hydrostatic pressure. At least 12 multistate outbreaks of salmonellosis, including approximately 2,000 culture-confirmed cases, have been traced to tomatoes since 1990 (Bidol and others 2007), raising concerns regarding growth, survival, and inactivation of salmonellae. When an outbreak strain of Salmonella Montevideo was introduced into the core tissue of mature green tomatoes using a −15 °C temperature differential between the fruit and immersion water, Zhuang and others (1995) showed that populations in the core tissue increased >tenfold during 18 days of storage at 20 °C, but remained unchanged or declined when stored at 10 °C. However, certain proteolytic yeasts and molds in

70

Section I. Microbial Contamination of Fresh Produce

damaged tomatoes can reportedly raise the core tissue pH to levels more conducive for Salmonella growth (Wade and Beuchat 2003a,b; Wade and others 2003). Submerging 25 °C tomatoes in a 320 ppm free chlorine solution at 37 °C for 2 min decreased the core tissue populations only 1 log, whereas Ibarra-Sánchez and others (2004) reportedly reduced S. Typhimurium populations in tomato core tissue from 2 logs to undetectable levels using a 15-sec 2% lactic acid spray treatment at 5 or 55 °C. The salmonellosis outbreaks just mentioned have also raised a number of preharvest food safety issues related to the microbiological safety of irrigation and pesticide water, animal manure, and sewage sludge (Guan and others 2005), any of which could contaminate soil and potentially lead to internalization of Salmonella or other foodborne pathogens through the root system. When placed in contact with inoculated soil, Salmonella can migrate from the soil directly into the stem scar tissue of green tomatoes (Guo and others 2002b) as was mentioned earlier in regard to E. coli O157 migration into apples (Seeman and others 2002). Uptake of Salmonella through the root system was demonstrated by Guo and others (2002a) using hydroponically grown tomato plants. When the roots were exposed to a hydroponic solution containing S. Montevideo, S. Enteritidis, and three other serovars, S. Montevideo migrated through the root system into the hypocotyls and cotyledons, stems, and leaves, whereas S. Enteritidis was absent. Stem injection and flower brushing of tomato plants with Salmonella cocktails also led to recovery of the pathogen both internally and externally. These findings help explain why Jablasone and others (2004) failed to detect Salmonella in the leaves, stems, or fruit of potted tomato plants that were irrigated with S. Enteritidis–inoculated water at 105 CFU/ml. Sprouts Since 1995 a series of widely publicized Salmonella and to a lesser extent E. coli O157:H7 outbreaks in the United States and elsewhere were traced to various types of sprouted seeds that were contaminated prior to germination. Commercially, these seeds are sprouted at 25–35 °C under very high relative humidity—conditions that promote rapid microbial growth and subsequent biofilm formation, as confirmed on 2-day-old alfalfa sprouts by both direct bacterial enumeration and scanning electron microscopy as previously shown in Figure 3.2 (Fett 2000). Following a major 1996 radish sprout outbreak in Japan that was traced to E. coli O157:H7, a team of Japanese investigators (Hara-Kudo and others 1997) demonstrated the ability of this pathogen to grow in experimentally contaminated radish seeds with populations increasing 3 to 5 logs during germination. Several immunofluorescence and scanning electron microscopy images by Itoh and others (1998) provide further proof that E. coli O157:H7 can migrate through surface stomata and become internalized within the inner tissue of sprouts. Several years later, Gandhi and others (2001) were among the first to study the interaction between Salmonella and alfalfa sprouts during seed germination and sprouting. When alfalfa seeds were immersed in a GFPlabeled suspension of Salmonella Stanley (∼7 log CFU/ml) for 5 min at 22 °C, an initial population of 3.7 log CFU/g was achieved in the seed. These numbers increased thousandfold during initial germination with the target pathogen penetrating to a depth of at least 18 μm in mature sprouts as determined by CLSM. Subsequently, Barak and others (2002) demonstrated that S. enterica and several plant-associated bacteria were

Internalization of Pathogens in Produce

71

more adherent than E. coli O157:H7 when 3-day-old alfalfa sprouts were used in a 4 h adhesion assay, with greater growth of S. enterica also observed on alfalfa seeds 2 days after sprouting (Charkowski and others 2002). Using GFP-labeled strains, increased colonization of alfalfa seeds and roots by S. enterica as opposed to E. coli O157:H7 (Charkowski and others 2002; Dong and others 2003) suggests that alfalfa sprouts are more prone to invasion by Salmonella than E. coli O157:H7. However, the same is not true for bean sprouts, with Warriner and others (2003b) reporting that bioluminescent strains of E. coli and S. Montevideo labeled with the lux gene similarly infiltrated bean sprouts, particularly through the roots, after 24 h of germination. Given that exposing the alfalfa seedling root area to as few as 1 Salmonella CFU led to infiltration and growth to high numbers 5 days later (Dong and others 2003), the chemical sanitizers currently used by sprout manufacturers do not provide the needed assurance in terms of sprout safety (Taormina and Beuchat 1999). Leafy Greens The safety of leafy greens, particularly lettuce and spinach, has become a top food safety priority with 20 of 24 leafy green outbreaks reported since 1996 traced to E. coli O157:H7. Consequently, three common reservoirs for E. coli O157:H7– contaminated irrigation water, manure composts, and soil—have been targeted as probable routes for dissemination of this pathogen in the field. However, contamination can also occur during harvesting, field coring, and further processing (e.g., shredding, conveying, washing, drying) with the ice used in cooling lettuce recently identified as another means for the spread of pathogens during melting (Kim and Harrison 2008). Using a series of lettuce and parsley plots treated once with E. coli O157:H7inoculated manure compost (107 CFU/g) before transplanting or spraying with E. coli O157:H7–inoculated irrigation water (105 CFU/ml) 3 weeks after transplanting, Islam and others (2004) reported that the pathogen persisted 126 to 154 and 154 to 217 days in soil samples in which lettuce and parsley were grown, respectively, with E. coli O157:H7 detected up to 77 days on lettuce and 177 days on parsley. In a similar study by Solomon and others (2003), E. coli O157:H7 persisted 10 to 22 days on 30-day-old lettuce plants after a single spray irrigation treatment with 100 ml of natural irrigation water containing 102 CFU/ml of a GFP-labeled E. coli O157:H7 strain. When plants were subjected to two or three spray irrigation treatments during the first 2 weeks, the pathogen persisted on the leaf surface for the entire 30-day duration of the study, underscoring safety concerns surrounding the potential for internalization of E. coli O157:H7 through surface stomata (Seo and Frank 1999; Takeuchi and Frank 2000). Given a choice, surface irrigation is preferred with Solomon and others (2002a) reporting that 29 of 32 as opposed to 6 of 32 similarly grown lettuce plants yielded the same GFP-labeled strain of E. coli O157:H7 after spray and surface irrigation, respectively, 1, 10, or 20 days before harvest. Working at Rutgers University, Solomon and others (2002b) were first to conclusively demonstrate uptake and subsequent internalization of a GFP-expressing E. coli O157:H7 from inoculated cow manure and irrigation water into leaf lettuce plants via the plant vascular system. Using soil inoculated at 106 CFU/g, E. coli O157:H7 was cultured from the interior of 6- and 9-day-old HgCl2-treated lettuce seedlings with

72

Section I. Microbial Contamination of Fresh Produce

CLSM confirming internalization up to 45 μm below the surface. When mature 50day-old plants were given a one-time exposure to 200 ml of irrigation water or manure slurry inoculated at 107 to 108 CFU/ml, E. coli O157:H7 was detected in irrigated and manure-treated plants at least 5 and 3 days after exposure, respectively. However, using a somewhat similar study design, Johannessen and others (2005) later reported no infiltration of a non-GFP labeled antibiotic resistant strain of E. coli O157:H7 into crisp head lettuce plants that were fertilized once with a manure slurry inoculated at 104 CFU/g. These later findings, which suggest that E. coli populations of at least 106 CFU/g or ml are needed during a contamination event for subsequent internalization through the root system, are also supported by the results of a related study (Wachtel and others 2002b) in which a mixture of nonpathogenic root-, shoot- and seed coat–adhering strains of E. coli naturally present in improperly treated sewage wastewater and a cabbage field colonized the roots but did not infiltrate edible portions of cabbage plants. Several additional publications also attest to the ability of E. coli O157:H7 and Salmonella to become internalized in lettuce plants via the root system. Using strains of E. coli O157:H7 labeled with GFP and nonpathogenic strains of E. coli in a simple seed and seedling adherence model, Wachtel and others (2002a) observed similar adherence of both organisms to the shoots and seed coats of hydroponically grown leaf lettuce seedlings. However, E. coli O157:H7 adherence rates to roots and root hairs were almost tenfold higher compared to the nonpathogen. Although similar findings were obtained when inoculated seeds were later grown in soil, GFP-labeled cells of E. coli O157:H7 were also seen on cotyledons in close proximity to stomatal pores, suggesting another entry route that has been confirmed by others (Seo and Frank 1999; Takeuchi and Frank 2000). Using GFP-labeled strains of E. coli O157:H7, Salmonella and L. monocytogenes, Jablasone and others (2005) subsequently reported similar colonization and growth on dip-inoculated lettuce and other seed types that were gnotobiotically cultivated, with E. coli O157:H7 growth confined to the seedling roots and root junctions. Although E. coli O157:H7 was quantifiable in internal tissue from surface-sterilized lettuce as well as cress, radish, and spinach plants after 9 but not 49 days, S. Typhimurium counts could be obtained only from 9-day-old lettuce and radish seedlings with no evidence of L. monocytogenes internalization in any of the nine seed types examined. Finally, when lettuce seeds were germinated in pathogen-inoculated soil, Franz and others (2007) reported greater internalization of E. coli O157:H7 as opposed to Salmonella into lettuce plants, with average E. coli O157:H7 and S. Typhimurium populations of 3.95 and 2.47 log CFU/g, respectively, recovered from AgNO3 surface-sterilized leaves. When grown hydroponically, only S. Typhimurium was recovered from surface-sterilized leaves, suggesting that root damage in soilgrown lettuce plants may play an important role in E. coli O157:H7 internalization. Using S. enterica, Bernstein and others (2007) subsequently confirmed this theory, reporting that 33-day-old lettuce leaves from lettuce plants transplanted to pots with and without prior root decapitation harbored mean Salmonella populations of 3.7 and 2.7 log CFU/g, respectively, with E. coli O157:H7 internalization into the aerial portions of maize plants also enhanced when the root system is damaged. In addition to the aforementioned preharvest routes of internalization, certain postharvest handling practices (e.g., field coring, shredding, and washing) may also

Internalization of Pathogens in Produce

73

promote the internalization of foodborne pathogens in leafy greens. After immersing iceberg lettuce leaves for 20 min in an E. coli O157:H7 suspension containing 107 CFU/ml, Seo and Frank (1999) reported that the pathogen adhered to the surface, trichomes, and stomata with a tenfold increase in attachment seen to cut edges. When these lettuce leaves were incubated in the suspension for 24 h, E. coli O157:H7 infiltrated the stomata and cut surfaces to depths of 20 to 100 μm with more deeply internalized cells surviving a 5 min exposure to a 20 ppm chlorine solution. In a similar follow-up study, Takeuchi and Frank (2000) found that E. coli O157:H7 penetrated the cut edges of iceberg lettuce to a mean depth of 73 μm when stored at 4 °C compared to 27 to 39 μm at 7 to 37 °C, with internalized cells protected during a 5 min exposure to a 200 ppm chlorine solution. In contrast to chemical sanitizers, gamma (Fan and others 2008; Niemira 2007) and x-ray irradiation (Jeong and others 2008), which have been shown to effectively reduce populations of E. coli O157:H7 on and/ or in leafy greens, may provide an industry-viable nonthermal means to better ensure the safety of fresh produce. Prior to the 2006 spinach outbreak in the United States, this product attracted minimal attention compared to lettuce, which is in large part due to the vast difference in consumption rates between these two products. When Warriner and others (2003a) assessed the interaction of a bioluminescent lux gene-labeled nonverotoxigenic E. coli strain with growing spinach plants, E. coli grew around the roots of developing seedlings and internalized within the root tissue and hypocotyls. However, infiltration occurred only when the plants were grown hydroponically and not in soil. The authors suggested that competitive microflora in the soil may have negated root colonization by E. coli. When 20-day-old spinach seedlings were grown in soil inoculated with E. coli at 2 log CFU/g, intact surface-sterilized 32- and 35-day-old plants contained 2.16 to 2.91 log E. coli CFU/g. Separate surface-sterilized root and leaf samples from mature 42-day-old plants yielded average E. coli counts of 3.78 and <1.20 log CFU/g. Because no infiltration of the leaves was seen when similar spinach seedlings were grown hydroponically, this nonpathogenic E. coli strain could not gain entry into growing spinach plants through the root system. In a follow-up study using a fluorescently labeled strain of E. coli O157:H7, Hora and others (2005) were still unable to detect this pathogen inside spinach leaves taken from growing plants despite prior mechanical disruption of the roots, coinoculation of the spinach plants with P. syringae via vacuum infiltration, or inoculation with the northern root-knot nematode Meloidogyne hapala. Although internalization of E. coli O157:H7 through the root system into the leaves of growing spinach plants currently appears to be unlikely, this does not preclude direct infiltration of the spinach leaves through natural openings or damaged areas. Among the remaining varieties of leafy greens, parsley and cilantro have attracted some attention in response to several earlier outbreaks. In limited work with parsley, Duffy and others (2005) used both surface sanitization (2000 ppm chlorine) and CLSM to demonstrate the uptake of three GFP-labeled Salmonella strains into parsley that had been submerged in the inoculum for 3 or 15 min and then stored for 7 or 30 days at 25 and 4 °C, respectively, with the pathogen growing internally in samples stored at 25 °C. The following year, Lapidot and others (2006) reported that one S. Typhimurium strain formed a biofilm on parsley after dip inoculation. When compared

74

Section I. Microbial Contamination of Fresh Produce

to a biofilm-forming deficient mutant, the biofilm-forming strain was better able to survive a subsequent chlorine treatment with infiltration of the tissue during dipinoculation likely enhancing survival. Following a 1999 California outbreak linked to cilantro tainted with Salmonella Thompson (Campbell and others 2001), Brandl and Mandrell (2002) used CLSM to confirm that this same S. Thompson strain labeled with GFP could infiltrate natural lesions on cilantro leaves after the plants were dipinoculated and incubated for 24 h under high humidity. These findings again stress the importance of pre- and postharvest damage as a plausible route for pathogen uptake into plant tissue.

Strategies for Minimizing Internalization Strategies to minimize produce contamination should be in place long before internalization occurs. Since some plant pathogens enhance the internalization of human pathogens, it is helpful to keep plants free of plant pathogens. Diseased fruits and vegetables should be eliminated during storage and processing. In the field, Good Agricultural Practices should be followed that include proper soil management (Beuchat 2002; Ibekwe and others 2007), use of certified seed (Janse and Wenneker 2002) that has been decontaminated by chemical sanitizers (Annous and others 2001), sublethal heating or irradiation (Erickson and Doyle 2007), careful selection of plant varieties and crop rotation to minimize insect infestations and disease (Gniwotta and others 2005; Kortekamp and Zyprian 1999), and careful harvesting to minimize microbial contamination, mechanical damage and bruising (Altekruse and others 1997). During processing, standard sanitizing procedures should be followed with special attention given to the temperature of both the product and wash water as well as the type of sanitizer, concentration and treatment time during processing, recognizing that these sanitizer washes are unable to completely eliminate pathogenic and spoilage organisms from the surface of fresh produce (Beuchat and Ryu 1997). Several reports have also demonstrated the potential use of natural microbes isolated from soil and plants as biological control agents for human pathogens both during and after harvest (Whipps 2001). Bacillus spp., Pseudomonas aeruginosa, P. fluorescens, and yeasts isolated from the native microflora of green bell peppers, romaine lettuce, and prepeeled baby carrots were found to inhibit the growth of Salmonella Chester and L. monocytogenes on green pepper 1 and 2 logs, respectively, (Liao and Fett 2001). Some strains of Gluconobacter asaii, Candida sp., Discosphaerina fagi, and Metschnikowia pulcherrima were also reportedly able to decrease the growth and survival of L. monocytogenes, E. coli O157:H7 and S. Poona on fresh-cut apples (Janisiewicz and others 1999; Leverentz and others 2006). Although lytic bacteriophages also reduced Salmonella populations 3.5 logs on honeydew melon slices, these same bacteriophages were ineffective against Salmonella on apple slices (Leverentz and others 2001). Bacterial endophytes could play an important role in inhibiting internalized human pathogens, with two reports suggesting that these endophytes may trigger induced systemic mechanisms that foster resistance against human pathogens (Zehnder and others 2001; Sturz and others 2000). Hence, application of indigenous bacteria could provide a potential biologically based strategy for preventing preharvest and postharvest internalization of human pathogens in fresh produce.

Internalization of Pathogens in Produce

75

References Alfano JR, Collmer A. 1996. Bacterial pathogens in plants: Life up against the wall. Plant Cell 8: 1683–1698. Altekruse SF, Cohen ML, Swerdlow DL. 1997. Emerging foodborne diseases. Emerg Infect Dis 3: 285–293. Annous BA, Sapers GM, Mattrazzo AM, Riordan DCR. 2001. Efficacy of washing with a commercial flatbed brush washer, using conventional and experimental washing agents, in reducing populations of Escherichia coli on artificially inoculated apples. J Food Prot 64:159–163. Annous BA, Solomon EB, Cooke PH, Burke A. 2005. Biofilm formation by Salmonella spp. on cantaloupe melons. J Food Microbiol 25:276–287. Araújo WL, Marcon J, Maccheroni Jr W, van Elsas JD, van Vuurde JWL, Azevedo JL. 2002. Diversity of endophytic bacterial populations and their interaction with Xylella fastidiosa in citrus plants. Appl Environ Microbiol 68:4906–4914. Artursson V, Jansson JK. 2003. Use of bromodeoxyuridine immunocapture to identify active bacteria associated with arbuscular mycorrhizal hyphae. Appl Environ Microbiol 69:6208–6215. Aruscavage D, Lee K, Miller S, LeJeune JT. 2006. Interactions affecting the proliferation and control of human pathogens on edible plants. J Food Sci 71:R89–R99. Barak JD, Whitehand LC, Charkowski AO. 2002. Differences in attachment of Salmonella enterica serovars and Escherichia coli O157:H7 to alfalfa sprouts. Appl Environ Microbiol 68:4758–4763. Bartz JA, Showalter RK. 1981. Infiltration of tomatoes by aqueous bacterial suspensions. Phytopathol 71:515–518. Bernstein N, Sela S, Neder-Lavon S. 2007. Assessment of contamination potential of lettuce by Salmonella enterica serovar Newport added to the plant growing medium. J Food Prot 70:1717–1722. Beuchat LR. 2002. Ecological factors influencing survival and growth of human pathogens on raw fruits and vegetables. Microbes Infect 4:413–423. ———. 2004. Difficulties in eliminating human pathogenic microorganisms on raw fruits and vegetables. Acta Hort 542:151–160. ———. 2006. Vectors and conditions for preharvest contamination of fruits and vegetables with pathogens capable of causing enteric diseases. Brit Food J 108:38–53. Beuchat LR, Ryu JH. 1997. Produce handling and processing practices. Emerg Infect Dis 3:459–465. Beuchat LR, Scouten AJ. 2004. Factors affecting survival, growth and retrieval of Salmonella Poona on intact and wounded cantaloupe rind and in stem scar tissue. Food Microbiol 21:683–694. Bidol SA, Daly ER, Rickert RE, Hill TA, Al Khaldi S, Taylor Jr. TH, Lynch MF, Painter JA, Braden CR, Yu PA, Demma L, Barton Behravesh C, Olson CK, Greene SK, Schmitz AM, Blaney DD, Gershman M. 2007. Multistate outbreaks of Salmonella infections associated with raw tomatoes eaten in restaurants—United States, 2005–2006. MMWR 56:909–911. Bowen A, Fry A, Richards G, Beuchat L. 2006. Infections associated with cantaloupe consumption: a public health concern. Epidemiol Infect 134:675–685. Brandl MT. 2006. Fitness of human enteric pathogens on plants and implications for food safety. Ann Rev Phytopathol 44:367–392. Brandl MT, Mandrell RE. 2002. Fitness of Salmonella enterica serovar Thompson in the cilantro phyllosphere. Appl Environ Microbiol 68:3614–3621. Branquinoho Bordini ME, Ristori AC, Jakabi M, Gelli DM. 2007. Incidence, internalization and behavior of Salmonella in mangoes, var Tommy Atkins. Food Contr 18:1002–1007. Bressan W, Borges MT. 2004. Delivery methods for introducing endophytic bacteria into maize. Biocontrol 49:315–322. Buchanan RL, Edelsen SG, Miller RL, Sapers GM. 1999. Contamination of intact apples after immersion in an aqueous environment containing Escherichia coli O157:H7. J Food Prot 62: 444–450. Burnett SL, Chen JR, Beuchat LR. 2000. Attachment of Escherichia coli O157:H7 to the surfaces and internal structures of apples as detected by confocal scanning laser microscopy. Appl Environ Microbiol 66:4679–4687. Caldwell KN, Anderson GL, Williams PL, Beuchat LR. 2003. Attraction of a free-living nematode, Caenorhabditis elegans, to foodborne pathogenic bacteria and its potential as a vector of Salmonella Poona for preharvest contamination of cantaloupe. J Food Prot 66:1964–1971.

76

Section I. Microbial Contamination of Fresh Produce

Campbell J, Mohle-Boetani J, Reporter R, Abbott S, Farrer R, Mandrell R, Werner S. 2001. An outbreak of Salmonella serotype Thompson associated with fresh cilantro. J Infect Dis 183:984–987. Charkowski AO, Barak JD, Sarreal CZ, Mandrell RE. 2002. Differences in growth of Salmonella enterica and Escherichia coli O157:H7 on alfalfa sprouts. Appl Environ Microbiol 68:3114–3120. Conway WS, Leverentz B, Saftner RA, Janisiewicz WJ, Sams CE, Leblanc E. 2000. Survival and growth of Listeria monocytogenes on fresh-cut apple slices and its interaction with Glomerella cingulata and Penicillium expansum. Plant Dis 84:177–181. Cooley MB, Chao D, Mandrell RE. 2006. Escherichia coli O157:H7 survival and growth on lettuce is altered by the presence of epiphytic bacteria. J Food Prot 69:2329–2335. Cooley MB, Miller WG, Mandrell RE. 2003. Colonization of Arabidopsis thaliana with Salmonella enterica and enterohemorrhagic Escherichia coli O157:H7 and competition by Enterobacter asburiac. Appl Environ Microbiol 69:4915–4926. Danhorn T, Fuqua C. 2007. Biofilm formation by plant-associated bacteria. Ann Rev Microbiol 61:401–422. DeWaal CS, Hicks G, Barlow K, Alderton L, Vegosen L. 2006. Foods associated with food-borne illness outbreaks from 1990 to 2003. Food Prot Trends 26:466–473. Dingman DW. 2000. Growth of Escherichia coli O157:H7 in bruised apple (Malus domestica) tissue as influenced by cultivar, date of harvest, and source. Appl Environ Microbiol 66:1077–1083. Dong Y, Iniguez AL, Ahmer BMM, Triplett EW. 2003. Kinetics and strain specificity of rhizosphere and endophytic colonization by enteric bacteria on seedlings on Medicago sativa and Medicago truncatula. Appl Environ Microbiol 69:1783–1790. Duffy EF, Cisneros-Zevallos L, Castillo A, Pillai SD, Ricke SC, Acuff GR. 2005. Survival of Salmonella transformed to express green fluorescent protein on Italian parsley as affected by processing and storage. J Food Prot 68:687–695. Dumont MG, Neufeld JD, Murrell JC. 2006. Isotopes as tools for microbial ecologists. Curr Opinion Biotechnol 17:57–58. Eblen BS, Walderhaug MO, Edelsen-Mammel S, Chirtel SJ, De Jesus A, Merker RI, Buchanan RL, Miller AJ. 2004. Potential for internalization, growth, and survival of Salmonella and Escherichia coli O157:H7 in oranges. J Food Prot 67:1578–1584. Erickson MC, Doyle MP. 2007. Food as a vehicle for transmission of Shiga toxin-producing Escherichia coli. J Food Prot 70:2426–2449. Fan XT, Niemira BA, Prakash A. 2008. Irradiation of fresh fruits and vegetables. Food Tech 62:36–42. Fett WF. 2000. Naturally occurring biofilms on alfalfa and other types of sprouts. J Food Prot 63:625– 632. Franz E, Visser AA, van Diepeningen AD, Klerks MM, Termorshuizen AJ, van Bruggen AHC. 2007. Quantification of contamination of lettuce by GFP-expressing Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium. Food Microbiol 24:106–112. Gandhi M, Golding S, Yaron S, Matthews KR. 2001. Use of green fluorescent protein expressing Salmonella Stanley to investigate survival, spatial location, and control on alfalfa sprouts. J Food Prot 64: 1891–1898. Gniwotta F, Vogg G, Gartmann V, Carver TLW, Riederer M, Jetter R. 2005. What do microbes encounter at the plant surface? Chemical composition of pea leaf cuticular waxes. Plant Physiol 139:519–530. Guan TTY, Blank G, Holley RA. 2005. Survival of pathogenic bacteria in pesticide solutions and on treated tomato plants. J Food Prot 68:296–304. Guo X, Chen J, Brackett RE, Beuchat LR. 2002b. Survival of Salmonella on tomatoes stored at high relative humidity, in soil, and on tomatoes in contact with soil. J Food Prot 65:274–279. Guo X, van Iersel MW, Chen J, Brackett RE, Beuchat LR. 2002a. Evidence of association of salmonellae with tomato plants grown hydroponically in inoculated nutrient solution. Appl Environ Microbiol 68:3639–3643. Hajdock C, Warriner K. 2007. Decontamination of fresh produce by using a combination of ultraviolet light and hydrogen peroxide. J Food Prot (suppl)70:40. Hallmann J, Kloepper JW, Rodriguez-Kabana R. 1997a. Application of the Scholander pressure bomb to study on endophytic bacteria of plants. Can J Microbiol 43:411–416. Hallmann J, Quadt-Hallmann A, Mahaffee WF, Kloepper JW. 1997b. Bacterial endophytes in agricultural crops. Can J Microbiol 43:895–914.

Internalization of Pathogens in Produce

77

Han Y, Sherman DM, Linton RH, Nielsen SS, Nelson PE. 2000. The effects of washing and chlorine dioxide gas on survival and attachment of Escherichia coli O157:H7 to green pepper surfaces. Food Microbiol 17:521–533. Hara-Kudo Y, Konuma H, Iwaki M, Kasuga F, Sugita-Konishi V, Ito Y, Kumagai S. 1997. Potential hazard of radish sprouts as a vehicle of Escherichia coli O157:H7. J Food Prot 60:1125–1127. Hollis JP. 1951. Bacteria in healthy potato tissue. Phytopathol 41:350–367. Hora R, Warriner K, Shelp BJ, Griffiths MW. 2005. Internalization of Escherichia coli O157:H7 following biological and mechanical disruption of growing spinach plants. J Food Prot 68:2506–2509. Ibarra-Sánchez LS, Alvarado-Casillas S, Rodríguez-García MO, Martínez-Gonzáles NE, Castillo A. 2004. Internalization of bacterial pathogens in tomatoes and their control by selected chemicals. J Food Prot 67:1353–1358. Ibekwe AM, Grieve CM, Ching-Hong Y. 2007. Survival of Escherichia coli O157:H7 in soil and on lettuce after soil fumigation. Can J Microbiol 53:623–635. Islam M, Doyle MP, Phatak SC, Millner P, Jiang X. 2004. Persistence of enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. J Food Prot 67:1365–1370. Itoh Y, Sugita-Konishi Y, Kasuga F, Iwaki M, Hara-Kudo Y, Saito N, Noguchi Y, Konuma H, Kumagai S. 1998. Enterohemorrhagic Escherichia coli O157:H7 present in radish sprouts. Appl Environ Microbiol 64:1532–1535. Jablasone J, Brovko LY, Griffiths MW. 2004. A research note: the potential for transfer of Salmonella from irrigation water to tomatoes. J Sci Food Agric 84:287–289. Jablasone J, Warriner K, Griffiths M. 2005. Interactions of Escherichia coli O157:H7, Salmonella Typhimurium and Listeria monocytogenes in plants cultivated in a gnotobiotic system. Int J Food Microbiol 99:7–18. Janisiewicz WJ, Conway WS, Brown MW, Sapers GM, Fratamico P, Buchanan RL. 1999. Fate of Escherichia coli O157:H7 on fresh-cut apple tissue and its potential for transmission by fruit flies. Appl Environ Microbiol 65:1–5. Janisiewicz WJ, Conway WS, Leverentz B. 1999. Biological control of postharvest decays of apples can prevent growth of Escherichia coli O157:H7 in apple wounds. J Food Prot 62:1372–1375. Janse JD, Wenneker M. 2002. Possibilities of avoidance and control of bacterial plant diseases when using pathogen-tested (certified) or -treated planting material. Plant Pathol 51:523–536. Jeong S, Marks BP, Yan ZN, Ryser ET. 2008. Unpublished data. Johannessen GS, Bengsston GB, Heier BT, Bredholt S, Wasteson Y, Rørvik LM. 2005. Potential uptake of Escherichia coli O157:H7 from organic manure into crisphead lettuce. Appl Environ Microbiol 71:2221–2225. Kado CI. 1992. Plant pathogenic bacteria. In: The Prokaryotes. edited by Balows A, Truper HG, Dworkin M, Harder W, and Schleifer KH. pp. 660–662. Springer-Verlag, New York. Kenney SJ, Burnett SL, Beuchat LR. 2001. Location of Escherichia coli O157:H7 on and in apples as affected by bruising, washing, and rubbing. J Food Prot 64:1328–1333. Kim JK, Harrison MA. 2008. Transfer of Escherichia coli O157:H7 to romaine lettuce due to contact water from melting ice. J Food Prot 71:252–256. Kluepfel D. 1993. The behavior and tracking of bacteria in the rhizosphere. Ann Rev Phytopathol 31:441–472. Kortekamp A, Zyprian E. 1999. Leaf hairs as a basic protective barrier against downy mildew of grape. J Phytopathol 147:453–459. Lamb TG, Tonkyn DW, Kluepfel DA. 1996. Movement of Pseudomonas aureofaciens from the rhizosphere to aerial plant tissue. Can J Microbiol 42:1112–1120. Lang MM, Harris LJ, Beuchat LR. 2004. Survival and recovery of Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes on lettuce and parsley as affected by method of inoculation, time between inoculation and analysis, and treatment with chlorinated water. J Food Prot 67:1092–1103. Lapidot A, Romling U, Yaron S. 2006. Biofilm formation and the survival of Salmonella on parsley. Int J Food Microbiol 109:229–233. Leben C. 1972. Micro-organisms associated with plant buds. J Gen Microbiol 71:327–331. Lee SW, Cooksey DA. 2000. Genes expressed in Pseudomonas putida during colonization of a plantpathogenic fungus. Appl Environ Microbiol 66:2764–2772.

78

Section I. Microbial Contamination of Fresh Produce

Leverentz B, Conway WS, Alavidze A, Janisiewicz WJ, Fuchs Y, Camp MJ, Chighladze E, Sulakvelidze A. 2001. Examination of bacteriophage as a biocontrol method for Salmonella on fresh-cut fruit: A model study. J Food Prot 64:1116–1121. Leverentz B, Conway WS, Janisiewiez W, Abadias M, Kurtzman CP, Camp MJ. 2006. Biocontrol of the food-borne pathogens Listeria monocytogenes and Salmonella enterica serovar Poona on fresh-cut apples with naturally occurring bacterial and yeast antagonists. Appl Environ Microbiol 72:1135–1140. Liao CH, Fett WF. 2001. Analysis of native microflora and selection of strains antagonistic to human pathogens on fresh produce. J Food Prot 64:1110–1115. Lodewyckx C, Vangronsveld J, Porteous F, Moore ERB, Taghavi A, Mezgeay M, derLelie D van. 2002. Endophytic bacteria and their potential applications. Crit Rev Plant Sci 21:583–606. Mahaffee WH, Bauske EM, Van Vuurde JWL, Van der Wolf JM, Van den Brink M, Kleopper JW. 1997. Comparative analysis of antibiotic resistance, immunofluorescent colony stain and a transgenic marker (bioluminescence) for monitoring the environment fate of a rhizobacterium. Appl Environ Microbiol 63:1617–1622. Marcais B, Breda N. 2006. Role of an opportunistic pathogen in the decline of stressed oak trees. J Ecol 94:1214–1223. Mendonca AF. 2005. Bacterial infiltration and internalization in fruits and vegetables. In: Produce Degradation: Pathways and Prevention, edited by Lamikanra O, Iman S, Ukuku D. pp. 442–460. CRC Press, Boca Raton, FL. Meneley JC, Stanghellini ME. 1974. Detection of enteric bacteria within locular tissue of healthy cucumbers. J Food Sci 36:1267–1268. Merker R, Edelson-Mamel S, Davis V, Buchanan RL. 1999. Preliminary experiments on the effect of temperature differences on dye uptake by oranges and grapefruit. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. Available from: . Last accessed 3/20/2008. Morris CE, Monier JM. 2003. The ecological significance of biofilm formation by plant-associated bacteria. Ann Rev Phytopathol 41:429–453. Morris CE, Monier JM, Jacques MA. 1997. Methods for observing microbial biofilms directly on leaf surfaces and recovering them for isolation of culturable microorganisms. Appl Environ Microbiol 63:1570–1576. Musson G, McInroy JA, Kloepper JW. 1995. Development of delivery systems for introducing endophytic bacteria into cotton. Biocontrol Sci Technol 5:407–416. Niemira BA. 2007. Relative efficacy of sodium hypochlorite wash versus irradiation to inactivate Escherichia coli O157:H7 internalized in leaves of romaine lettuce and baby spinach. J Food Prot 70:2526–2532. Papendic RI, Cook RJ, Shipton PJ. 1971. Plant water stress and development of Fusarium root rot in wheat. Phytopathol 61:905–907. Payne GA, Thompson DL, Lillehoj EB. 1982. Effect of water-stress on infection and aflatoxin production by Aspergillus flavus in corn. Phytopathol 72:972. Penteado AL, Eblen BS, Miller AJ. 2004. Evidence of Salmonella internalization into fresh mangos during simulated postharvest insect disinfestation procedures. J Food Prot 67:181–184. Plotnikova JM, Rahme LG, Ausubel FM. 2000. Pathogenesis of the human opportunistic pathogen Pseudomonas aeruginosa PA14 in Arabidopsis. Plant Physiol 124:1766–1774. Quadt-Hallmann A, Benhamou N, Kloepper JW. 1997. Bacterial endophytes in cotton: mechanisms of entering the plant. Can J Microbiol 43:577–582. Reinders RD, Beisterveld S, Bijker RGH. 2001. Survival of Escherichia coli O157:H7 ATCC 43895 in a model apple juice medium with different concentrations of praline and caffeic acid. Appl Environ Microbiol 67:2863–2866. Richards GM, Beuchat LR. 2004. Attachment of Salmonella Poona to cantaloupe rind and stem scar tissues as affected by temperature of fruit and inoculum. J Food Prot 67:1359–1364. ———. 2005a. Infection of cantaloupe rind with Cladosporium cladosporioides and Penicillium expansum, and associated migration of Salmonella Poona into edible tissues. Intern J Food Microbiol 103: 1–10. ———. 2005b. Metabiotic associations of molds and Salmonella Poona on intact and wounded cantaloupe rind. Intern J Food Microbiol 97:327–339.

Internalization of Pathogens in Produce

79

Riordan DCR, Sapers GM, Annous BA. 2000. The survival of Escherichia coli O157:H7 in the presence of Penicillium expansum and Glomerella cingulata in wounds on apple surfaces. J Food Prot 63:1637–1642. Riordan DCR, Sapers GM, Hankinson TR, Magee M, Mattrazzo AM, Annous BA. 2001. A study of US orchards to identify potential sources of Escherichia coli O157:H7. J Food Prot 64:1320– 1327. Rosenblueth M, Martínez-Romero E. 2006. Bacterial endophytes and their interactions with hosts. Mol Plant-Microbe Interact 19:827–837. Ryan RP, Germaine K, Franks A, Ryan DJ, Dowling DN. 2008. Bacterial endophytes: recent developments and applications. FEMS Microbiol Lett 278:1–9. Samish Z, Etinger-Tulczynska R. 1963a. Distribution of bacteria within the tissue of healthy tomatoes. Appl Microbiol 11:7–10. Samish Z, Etinger-Tulczynska R, Bick M. 1961. Microflora within healthy tomatoes. Appl Microbiol 9:20–25. ———. 1963b. The microbiota within the tissues of fruits and vegetables. J Food Sci 28:259–266. Sasaki T, Kobayashi M, Agui N. 2000. Epidemiological potential of excretion and regurgitation by Musca domestica (Diptera:Muscidae) in the dissemination of Escherichia coli O157:H7 to food. J Med Entomol 37:945–949. Schulz B, Boyle C. 2006. What are endophytes? In: Microbial Root Endophytes, edited by Schulz BJE, Boyle CJC, Sieber TN, pp. 1–13. Springer-Verlag, Berlin. Seeman BK, Sumner SS, Marini R, Kniel KE. 2002. Internalization of Escherichia coli in apples under natural conditions. Dairy Food Environ Sanitat 22:667–673. Segall RH, Henry FE, Dow AT. 1977. Effect of dump-tank temperature on the incidence of bacterial soft rot of tomatoes. Proc FL State Hort Soc 90:204–205. Sela S, Nestel D, Pinto R, Nemny-Lavy E, Bar-Joseph M. 2005. Mediterranean fruit fly as a potential vector of bacterial pathogens. Appl Environ Microbiol 71:4052–4056. Seo KH, Frank JF. 1999. Attachment of Escherichia coli O157:H7 to lettuce leaf surface and bacterial viability in response to chlorine treatment as demonstrated by using confocal scanning laser microscopy. J Food Prot 62:3–9. Solomon EB, Brandl MT, Mandrell RE. 2006. Biology of foodborne pathogens on produce. In: Microbiology of Fresh Produce, edited by Matthew KR, pp. 55–83. ASM Press, Washington, D.C. Solomon EB, Matthews KR. 2005. Use of fluorescent microspheres as a tool to investigate bacterial interactions with growing plants. J Food Prot 68:870–873. Solomon EB, Pang H-J, Matthews KR. 2003. Persistence of Escherichia coli O157:H7 on lettuce plants following spray irrigation with contaminated water. J Food Prot 66:2198–2202. Solomon EB, Potenski CJ, Matthews KR. 2002a. Effect of irrigation method on transmission to and persistence of Escherichia coli O157:H7 on lettuce. J Food Prot 65:673–676. Solomon EB, Yaron S, Matthews KR. 2002b. Transmission of Escherichia coli O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalization. Appl Environ Microbiol 68:397–400. Stone JK, Bacon CW, White JF. 2000. An overview of endophytic microbes: endophytism defined. In: Microbial Endophytes, edited by Bacon CW, White JF, pp.3–29. CRC Press, Boca Raton, FL. Sturz AV, Christie BR, Matheson BG, Arsenault WJ, Buchanan NA. 1999. Endophytic bacterial communities in the periderm of potato tubers and their potential to improve resistance to soil-borne plant pathogens. Plant Pathol 48:360–369. Sturz AV, Christie BR, Nowak J. 2000. Bacterial endophytes: Potential role in developing sustainable systems of crop production. Crit Rev Plant Sci 19:1–30. Takeuchi K, Frank JF. 2000. Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and the effect of chlorine treatment on cell viability. J Food Prot 63:434–440. Taormina PJ, Beuchat LR. 1999. Behavior of enterohemorrhagic Escherichia coli O157:H7 on alfalfa sprouts during the sprouting process as influenced by treatments with various chemicals. J Food Prot 62:850–856. Wachtel MR, Whitehand LC, Mandrell RE. 2002a. Association of Escherichia coli O157:H7 with preharvest leaf lettuce upon exposure to contaminated irrigation water. J Food Prot 65:18–25.

80

Section I. Microbial Contamination of Fresh Produce

———. 2002b. Prevalence of Escherichia coli associated with a cabbage crop inadvertently irrigated with partially treated sewage wastewater. J Food Prot 65:471–475. Wade WN, Beuchat LR. 2003a. Metabiosis of proteolytic moulds and Salmonella in raw, ripe tomatoes. J Appl Microbiol 95:437–450. ———. 2003b. Proteolytic fungi isolated from decayed and damaged raw tomatoes and implications associated with changes in pericarp pH favorable for survival and growth of foodborne pathogens. J Food Prot 66:911–917. Wade WN, Vasdinnyei R, Deak T, Beuchat LR. 2003. Proteolytic yeasts isolated from raw, ripe tomatoes and metabiotic association with Geotrichum candidum with Salmonella. Intern J Food Microbiol 86:101–111. Warriner K. 2005. Pathogens in vegetables. In: Improving the Safety of Fresh fruits and Vegetables, edited by Jongen WMF, pp.3–43. CRC Press, Boca Raton, FL. Warriner K, Ibrahim F, Dickinson M, Wright C, Waites WM. 2003a. Interaction of Escherichia coli with growing salad spinach plants. J Food Prot 66:1790–1797. Warriner K, Spaniolas S, Dickinson M, Wright C, Waites WM. 2003b. Internalization of bioluminescent Escherichia coli and Salmonella Montevideo in growing bean sprouts. J Appl Microbiol 95:719–727. Waterfield NR, Wren BW, French-Constant RH. 2004. Invertebrates as a source of emerging human pathogens. Nat Rev Microbiol 2:833–841. Wells JM, Butterfield JE. 1997. Salmonella contamination associated with bacterial soft rot of fresh fruits and vegetables in the marketplace. Plant Dis 81:867–872. ———. 1999. Incidence of Salmonella on fresh fruits and vegetables affected by fungal rots or physical injury. Plant Dis. 83:722–726. Whipps JM. 2001. Microbial interactions and biocontrol in the rhizosphere. J Exp Bot 52:487–511. Wilson D. 1995. Endophyte—the evolution of a term, and clarification of its use and definition. Oikos 73:274–276. Wilson M, Hirano SS, Lindow SE. 1999. Location and survival of leaf-associated bacteria in relation to pathogenicity and potential for growth within the leaf. Appl Environ Microbiol 65:1435–1443. Wulff EG, van Vuurde JWL, Hockenhull J. 2003. The ability of the biological control agent Bacillus subtilis, strain BB, to colonise vegetable brassicas endophytically following seed inoculation. Plant Soil 255:463–474. Yan, ZN, Reddy MS, Kloepper JW. 2003. Survival and colonization of rhizobacteria in a tomato transplant system. Can J Microbiol 49:383–389. Zehnder GW, Murphy JF, Sikora EJ, Kloepper JW. 2001. Application of rhizobacteria for induced resistance. Eur J Plant Pathol 107:39–50. Zhuang RY, Beuchat LR, Angulo FJ. 1995. Fate of Salmonella montevideo on and in raw tomatoes as affected by temperature and treatment with chlorine. Appl Environ Microbiol 61:2127–2131. Zottola EA. 1994. Microbial attachment and biofilm formation—a new problem for the food industry. Food Technol 48(7):107–114.

Section II Preharvest Strategies

4 Produce Safety in Organic vs. Conventional Crops Francisco Diez-Gonzalez and Avik Mukherjee

Introduction The demand for fresh fruits and vegetables as popular food choices in our daily diet has been increasing in recent years. As consumers are becoming increasingly aware of the potential health benefits, they are consuming more fresh produce in their diets. The market for fresh produce expanded by almost 20% over the last three decades, but at the same time the number of foodborne outbreaks from contaminated produce have sharply increased. In addition to the fastest growth of the fresh produce market, the organic foods segment has also experienced sustained increases of an average of 20% annually (Dimitri and Oberholtzer 2005). Under these circumstances, the increased popularity of organic foods in general and organic produce in particular, adds a new dimension to the food safety concerns with fresh fruits and vegetables. Organic growers are largely limited to the use of animal waste for fertilization of produce and this constraint may pose a greater risk for contamination with foodborne pathogens. Because of this potentially enhanced risk, it would be reasonable to think that organic produce is less safe than conventional fruits and vegetables. The existing data in the literature does not seem to support those concerns associated with organic produce, but there are some studies that have seriously questioned the validity of current organic manure practices. This chapter summarizes some of the most relevant findings that shed light on assessing the safety of organic as compared to conventional produce.

Organic Foods History Jerome Rodale introduced the principles of “organic” agriculture to the U.S. in the 1940s, but for most of the following 40 years, organic production was adopted by only a very small number of farmers (Conford 2001). During the early 1990s, organic food production started gaining momentum in the United States as the interest in foods produced more “naturally” increased. The Organic Foods Production Act (OFPA) was passed in 1990 as part of the U.S. Farm Bill. This law was intended for streamlining national standards for production, processing, handling, and marketing of organic foods. This regulation was also meant to ensure consistent quality and standards for organic foods produced, processed, handled, and marketed across the nation. Under this Act, the National Organic Standard Board (NOSB) was established in 2000. The NOSB issued the National Standards on Organic Agricultural Production and Handling, also known as the Organic Rule. This set of regulations has been 83

84

Section II. Preharvest Strategies

implemented after some modifications and amendments since October 2002 (NOP/ USDA 2002b). According to the Organic Rule, organic foods are defined as the foods produced without use of genetic engineering, ionizing radiation, and sewage sludge, and only by using tillage, cultivation practices such as crop rotation, cover crop, and fertilization with properly treated crop and animal wastes. The NOSB recognized the impossibility of obtaining all inputs and ingredients produced entirely organic and issued a list of synthetic substances that are allowed for crops, livestock, and processing. Organic livestock producers are allowed to use only 100% organic feed, and the animals must have access to outdoors and pastures. The use of preventive management practices such as vaccination is allowed, but antibiotics, growth hormones, and other synthetic drugs are not allowed (NOP/USDA 2002b). The Organic Rule also mentions that these standards are not intended to ensure safety of organically produced foods. In the U.S., organic is also referred to as a labeling term and depending on the content of organic ingredients, a particular food can be labeled differently (NOP/ USDA 2002b). Organic foods labeled as “100 percent organic” must contain all organically produced ingredients, those labeled as “organic” must contain at least 95% organically produced ingredients, and those labeled as “made with organic …” must contain at least 70% organic ingredients. Only those foods labeled as “100 percent organic” and “organic” are allowed to have the USDA organic seal on their label. Before the Organic Rule was issued, other countries and the U.S. relied on a system of third-party certifying agencies to ensure that products sold as “organic” were indeed using organic practices. The NOSB adopted the same oversight system and the role of the National Organic Program (NOP) is to license those certifying agencies. Certifying agencies typically inspect and audit organic producers and processors once a year. During the inspection, certifiers verify the use of organic practices and ingredients as well as the existence of record keeping. Conventional producers that want to adopt organic practices should go through a “transition” period of 3 years before they can be certified. The Codex Alimentarius Commission also established a set of guidelines in 1999 to address the issue of uniform standards of production, processing, handling, labeling, and global trade of organically produced foods, (Codex Alimentarius Commission 1999). The Codex defines organic agriculture as “a holistic production management system which promotes … biodiversity, biological cycles, and soil biological activities … using cultural, biological and mechanical methods as opposed to synthetic materials … to … maintain long term soil fertility, recycle wastes of plant and animal origin, …. ” The Codex Alimentarius also emphasizes that organic agriculture relies on renewable resources. Market Organic agriculture on a commercial scale began in the European Union countries in the early 1920s, and in the U.S. during the 1940s (Dimitri and Oberholtzer 2005). On both continents, these early stages of organic agricultural production were largely supply driven and motivated by farmers’ preference to use lesser amounts of chemicals in food production. Since the 1980s, however, the fast expansion of organic agriculture was primarily market driven. This fast expansion of market for organic foods started

Produce Safety in Organic vs. Conventional Crops

85

for different reasons, among the American and European consumers. Food scares, such as the bovine spongiform encephalopathy (BSE) epidemic, as well as an inherent cultural fear of consumption of genetically engineered foods, played a major role for market expansion in the EU countries. In the U.S., the consumers’ concerns about environmental health and excessive use of chemicals in food production were the primary driving forces behind the rapid expansion of the organic foods market. In recent surveys, however, consumer motivation for preferring organic foods is shifting more toward personal reasons such as health and food safety, in both the EU and the U.S. (Winter and Davis 2006). Consumers in the U.S. and the EU purchase 95% of the total global sales of organic foods (Dimitri and Oberholtzer 2005). Government policies and growth patterns in organic agriculture differ markedly between these two continents. Although growth rates of organic food sales are at various stages of maturity in the European countries, sales in the United States have still grown at an average rate of 20% annually since 1998. According to most estimates, the growth in the U.S. will continue at an annual rate of 10 to 16% in the following years (Table 4.1). Sales of organic foods were 1.4% of the total U.S. food sales in 2001, and they reached almost 2% in 2005. This increase in the share of organic food sales of the total market in the U.S. is expected to reach 3.5% by 2010 (Dimitri and Oberholtzer 2005). Another indication of growing preference of organically produced foods among Americans is the fact that the distribution of organic food sales has shifted from health food stores to conventional retail stores since 1998. In 2003, 47% of organic food sales occurred in health food stores, while 44% took place in conventional retail stores. In 1998, this distribution was 63% and 31% for health food stores and conventional retail outlets, respectively. Fresh fruits and vegetables have been the top-selling organic food category in the U.S. for quite some time (Oberholtzer and others 2005). In terms of amounts purchased annually, tomatoes, carrots, peaches, squash, leafy vegetables, and apples are some of the leading organic produce types in this country. In 2005, the sales of organic Table 4.1. Organic food sales and growth rates in the U.S., 1997–2004 (Dimitri and Greene 2002; Oberholtzer and others 2005) Year

Sales ($ bill.)

1990

0.6

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2010

3.6 4.3 5.0 6.1 7.4 8.6 10.4 12.4 14.5 17.7 23.8*

* = estimated.

Growth Rate (%)

19.7 18.2 21.0 20.7 17.3 20.2 19.2 17.0 22.0

86

Section II. Preharvest Strategies

produce accounted for $5.4 billion (40%) out of $13.6 billion total organic food sales in the U.S. (Winter and Davis 2006). As much as 93% of organic fruits and vegetables are sold in the form of fresh produce in this country (Dimitri and Greene 2002). It is estimated that by 2010, organic fruit and vegetable sales in the U.S. will reach $8.5 billion, about four times the organic produce sold in 2000. In the U.S., California is the leading producing state and Minnesota is the third largest producer in overall organic food production. California is also the leading state in the U.S. in production of organic fruits and vegetables.

Safety Issues Associated with Organic Fresh Produce According to a number of consumer surveys, “safer foods” is one of the qualifiers often given to organic foods, but this perception is largely driven by the association of “unsafe” and synthetic ingredients (Magkos and others 2006). In the case of fresh produce consumers frequently consider that organic fruits and vegetables are safer because they have less pesticide residues. Although the long-term health effects of ingesting pesticide residues are yet to be corroborated, to date there is convincing evidence that organic fruits and vegetables have significantly lower prevalence and concentration of pesticide residues than their conventional counterparts (Baker and others 2002; Pussemier and others 2006). This consumer perception of safer organic foods often ignores the microbiological safety issues resulting from organic production practices. Organic practices limit the number of synthetic inputs used in food production, and in the case of fertilizers all synthetic compounds have been banned (NOP/USDA 2002a). According to the Organic Rule, farmers can use a variety of natural wastes as organic fertilizers, but the most widely available and least expensive fertilizing material is livestock manure. The organic regulations do not restrict the type of animal manure used on crops, but in the U.S. the NOP established specific guidelines for its utilization on crop fields (NOP/USDA 2002b). The Organic Rule recommends the application of manure after composting, but it also allows the use of raw manure if it is applied from 90 to 120 days before harvest. The term compost is defined as “the product of a managed process through which microorganisms break down plant and animal materials into more available forms suitable for application to the soil … must be produced through a process that combines plant and animal material with an initial C:N ratio of between 25:1 to 40:1” (NOP/USDA 2002a). For composting, organic farmers using an in-vessel or aerated pile system are required to maintain the compost pile at 131 to 170 °F for 3 days, and users of the windrow system must maintain the compost material in the same temperature range for 15 days, during which the materials must be turned a minimum of five times (NOP/USDA 2002b). Raw animal manure may be used only when it is applied to soil at least 120 days prior to harvest of products whose edible portion comes in direct contact with the soil surface or soil particles, and at least 90 days prior to harvest of products whose edible portion does not come in contact with the soil surface or soil particles. The main food safety concern of organic fresh produce as a vehicle of foodborne infections stems from the utilization of manure as crop fertilizer. Pathogenic bacteria

Produce Safety in Organic vs. Conventional Crops

87

such as Salmonella, Campylobacter, Listeria monocytogenes, and Escherichia coli O157:H7 are natural inhabitants of the gastrointestinal tract of livestock and may be excreted in feces and remain viable in manure. The almost obligatory use of manure in organic agriculture has led people to believe that organic produce would be more susceptible to contamination with those pathogenic organisms than their conventional counterparts. However, manure may also be used by conventional farmers who are not restricted on how and when it can be spread onto fields. The limitations in manure usage by the Organic Rule offer some level of protection but it has become apparent that they would not be able to control the transmission of foodborne pathogens from animal manure to fresh produce. If composting is done following the organic guidelines, almost none of those pathogenic bacteria would be able to survive, but it has been reported that composting in windrows is often difficult to control (Forshell and Ekesbo 1993; Lung and others 2001). Some of the problems associated with improper composting are due to the lack of uniform heating, the cross-contamination of treated and raw manure, and the diversity of composting systems (Brinton and Storms 2004). Although proper composting would control pathogenic organisms, the guidelines for raw manure use may not offer sufficient protection. When the Organic Rule was issued in the late 1990s there was very limited scientific information about the survival of foodborne pathogens in manure and soil, but the NOSB members had to formulate time recommendations (Sideman 2006). With the few publications available they determined that 90 and 120 days would be sufficient to minimize risks, but recent studies have suggested that pathogenic bacteria are capable of surviving in the environment more than 120 days (Gagliardi and Karns 2002; Islam and others 2004). Given these potential problems of current organic practices it seems apparent that the safety of fresh produce can be compromised. As a result of these concerns, a significant number of studies have been devoted to investigate whether organic fruits and vegetables pose a greater risk for transmission of foodborne pathogens. Researchers have used a variety of approaches to assess the risk of organic fresh produce and to address the question of whether they are more susceptible to pathogen contamination. The three major strategies that scientists have taken include 1) pathogen survival studies as affected by organic practices, 2) product testing surveys in which the prevalence of pathogenic and indicator bacteria of organic produce is compared to their conventional counterparts, and 3) identification of organic practices with increased risk of contamination. The following three sections discuss these approaches.

Pathogen Survival in Manure, Compost, and Soil Survival of Salmonella and E. coli O157:H7 in Manure and Compost Application of untreated or improperly treated animal waste as fertilizers in agricultural soil used for food production is a major concern for the production of any type of fresh fruits and vegetables that will be eaten raw. A number of researchers have studied survival of pathogens like E. coli O157:H7 and Salmonella in animal manure, and various survival rates have been reported (Table 4.2). The range of survival

88

Section II. Preharvest Strategies

Table 4.2. Summary of survival studies of foodborne pathogens in livestock manure Manure and Conditions Cattle manure

S. Dublin S. Senftenberg S. Typhimurium E. coli O157:H7

Maximum Survival (Days) 183 204 204 70 at 5 °C

Piles and slurry ovine manure Field manure

E. coli O157:H7

>365

E. coli O157:H7

99

Cattle manure and slurries at 4 °, 22 °, and 37 °C Chicken manure at 20 °C

E. coli O157 S. Typhimurium

100 at 4 °C

S. Typhimurium

100 and aw = 0.07

Chicken manure at 4 °, 22 °, and 37 °C Cattle slurries at RT

E. coli O157 S. Typhimurium

261

Campylobacter Salmonella E. coli O157 S. Newport

90

Manure in lab

Dairy manure

Bacteria

184

Other Findings

Reference

Inactivated in Forshell and composted manure Ekesbo 1993 after 14 days Survival increases at Wang and others low temp. and 1996 high moisture Little survival in Kudva and aereated piles others 1998 Fastest inactivation Bolton and on soil others 1999 6-log reduction after Himathongkham 38 and 48 days and Riemann 1999 Himathongkham 6-log decline at and others aw = 1 after 22 days 1999 Maximum Himathongkham DRT = 150 days and others 2000 Inactivated after 30 Nicholson and days at 55 °C others 2005 D-values from 14 to 32 days

You and others 2006

in manure, depending on the conditions, has been reported from 70 to 260 days. In general, survival of bacterial pathogens in animal wastes appears to be affected by temperature, moisture content, pH, and holding time (Warriner 2005). Wang and colleagues (1996) reported longer survival of E. coli O157:H7 in bovine manure at 5 °C compared to 30 °C. In another study, survival of E. coli O157:H7 and Salmonella Typhimurium were recorded in cattle manure and cattle manure slurry at different temperatures (Himathongkham and others 1999). In that study, the decimal reduction times for both the pathogenic bacteria at 4 °C were seven to eight times longer than that at 37 °C. Other reports have observed more than a year of survival of E. coli O157:H7 in animal feces at temperatures below 5 °C, compared to temperatures above 20 °C (Kudva and others 1998). Composting is considered an effective way of minimizing pathogenic counts in animal wastes. Lung and others (2001) used laboratory cultures and bench-scale composting to study effectiveness of the process in reducing E. coli O157:H7 and Salmonella Enteritidis in cattle manure. In that study, composting at 45 °C for 3 days decreased the pathogen counts to nondetectable levels (Lung and others 2001). Using a similar approach in laboratory reactors, another study reported a 5-log reduction in 7 days in the viable count of E. coli O157 during manure composting as long as the temperature was greater than 52 °C (Jiang and others 2003). In one of the few publications that have investigated the survival of E. coli O157 on compost heaps, Shepherd

Produce Safety in Organic vs. Conventional Crops

89

and others (2007) used a nontoxigenic strain to determine the effect of heap turning. They observed that the pathogen reached almost undetectable levels in less than 2 weeks when the compost piles were turned, but the organism was still detected after 4 months if the surface did not reach higher than 50 °C and there was significant temperature stratification. Survival of Pathogenic Organisms in Soil A thorough assessment of the ability of pathogenic organisms to survive in soil after the soil has been amended with manure is critical to determine the likelihood of contamination in organic produce. Recent outbreaks caused by contaminated fresh vegetables have stressed the potential role of soil as a vehicle of transmission. Determining the survival rates of pathogenic organisms in soil has been a very active area of research in preharvest food safety; some of the most relevant studies are summarized in Table 4.3. From the results of these reports, the type of soil and crop type are factors that have an important effect on survival of pathogenic organisms.

Table 4.3. Summary of survival studies of foodborne pathogens in soil Soil, Manure Type, and Conditions Soil cores in lab

E. coli O157

Maximum Survival (Days) 130

Clay loam and sandy loam, cattle slurry Pig manure

E. coli O157

28

S. Typhimurium

<21

Campsite with sheep manure Clay soil/ loamy sands, cow manure Fallow soil and silt loam soil, cattle manure Clay loam and channery silt loam Composted dairy and poultry manure Sandy loam soil, dairy cow manure Cattle slurries at RT

E. coli O157

105

Reduction similar to indicator E. coli Noted that it could survive up to 299 days Outbreak-related

S. Typhimurium

63

3- to 4-log reduction

E. coli O157

41 (fallow) 92 (silt loam) 500 (frozen) 120 (90%) 164 (99%) 154 (lettuce) 214 (parsley)

Clay increased persistence

Gagliardi and Karns 2002

High variability in survival 4-log reduction after 42 days

Kato and others 2004 Islam and others 2004

196 (onion) >84 (carrot) 90

3-log reduction w/ onions at 64 days Inactivated after 30 days at 55 °C

Islam and others 2005 Nicholson and others 2005

96

Natural contamination linked to a human case No difference between soil types

Mukherjee and others 2006a Semenov and others 2008

Garden soil

Organic and conventional soils

Organism

Cryptosporidium parvum E. coli O157

E. coli O157 Campylobacter Salmonella E. coli O157 E. coli O157

E. coli O157

>60

Other Findings

Reference

Only 2-log reduction

Maule 2000 Ogden and others 2001 Baloda and others 2001 Ogden and others 2002 Natvig and others 2002

90

Section II. Preharvest Strategies

A range of maximum survival times from 28 days at room temperature to 500 days in frozen soil have been reported for E. coli O157:H7. In should be noted, however, that more than four reports were able to recover pathogenic organisms long after 120 days. In general, it appears that the survival of Salmonella and E. coli O157:H7 was stimulated by increasing moisture in soil and by lower temperature. In one study, E. coli and Salmonella Typhimurium inoculated in moist grey loam topsoil showed greater persistence than that in drier soil (Chandler and Craven 1980). On average Salmonella seems to adapt better than E. coli O157:H7 serotype to the harsh, competitive soil environment, and it survives longer (Warriner 2005). Foodborne pathogens also persist longer in the rhizosphere, compared to bulk soil alone. Some researchers have suggested that plant roots provide better protection for the pathogens against protozoa (Maule 2000). Several studies have focused on survival of foodborne pathogens and fecal indicator bacteria in soil fertilized with animal wastes inoculated with laboratory grown cultures. In one study, fecal indicator E. coli and other enteric bacteria belonging to the enterococci group were inoculated in bovine manure (Lau and Ingham 2001). E. coli showed greater survival than enterococci, in two different types of soils in Wisconsin amended with bovine manure. In another study, in Denmark, S. enterica serovar Typhimurium DT12 was inoculated in animal waste slurry from a piggery, and the contaminated wastes were disposed of by spreading them on agricultural soil (Baloda and others 2001). The pathogen remained detectable for 2 weeks in the soil. That study emphasized the significance of improper agricultural practices that may lead to transmission of human pathogens into agricultural soil, and then to human foods like produce cultivated in such soils. In another field study, a laboratory grown E. coli O157:H7 culture was inoculated in commercially manufactured composts, made from poultry and cattle manure, and these inoculated composts were mixed into agricultural soil plots (Islam and others 2005). The pathogen was detected in the manure-amended soil for as long as 196 days, and in onions grown in those plots for almost 170 days. Hutchison and colleagues (2004) reported survival of E. coli O157:H7 and Salmonella when untreated and inoculated livestock wastes were applied on the top of agricultural soil. In this latter study, when untreated pig and cattle manure was left on the top of agricultural soil for at least a week before incorporation into the soil, the D-values for Salmonella decreased by 1 to 3 days. For E. coli O157:H7, the D-values decreased by 5 days when incorporation of untreated cattle manure and pig slurry into agricultural soil was delayed by at least a week. The majority of survival studies have been conducted using inoculation with strains grown in the laboratory, and the different inoculum preparations may also explain the wide variability of results found in the literature. Among the few studies that have monitored the survival of naturally occurring pathogenic bacteria, a documented case of environmental survival of an E. coli O157:H7 strain in soil was related to an outbreak of gastroenteritis among participants in a boy scout camp (Howie and others 2003; Strachan and others 2001). In that outbreak, 20 boys developed gastroenteritis after camping on a pasture field previously used for grazing sheep. Investigations of the field reported that a low level of 1.8 log CFU/g of O157:H7

Produce Safety in Organic vs. Conventional Crops

91

serotype was initially detected in the soil, but after 3 weeks from the onset of symptoms, the pathogen could no longer be detected (Strachan and others 2001). This decline was similar with the findings of a study conducted in our laboratory, where the survival of a naturally occurring strain of E. coli O157 that had been responsible for the infection of a baby was tracked for several weeks in a residential garden (Mukherjee and others 2006a). The E. coli O157:H7 count was reduced by 1.5 log CFU/g after 23 days from an initial count of 3.4 log CFU/g. The pathogenic strains remained viable in the manure-amended garden for at least 69 days.

Bacterial Prevalence in Organic and Conventional Produce The number of foodborne outbreaks and cases of illnesses from consumption of contaminated produce has increased markedly in recent years. However, the isolation of pathogenic bacteria from fresh fruits and vegetables has been reported less frequently (Rangel and others 2005; Sivapalasingam and others 2004). Surveys focusing on isolation of contaminating bacteria from fresh produce have been conducted almost exclusively on postharvest, retail samples (Johnston and others 2005; Loncarevic and others 2005; McMahon and Wilson 2001; Phillips and Harrison 2005; Sagoo and others 2001). Other surveys focused on fresh vegetables that were cultivated in experimental plots treated with manure and/or irrigation water inoculated with laboratory-grown pathogenic bacteria (Johannessen and others 2004, 2005; Lang and others 2004). Clearly, there is only limited data in the literature, reporting the isolation of contaminating bacteria from fresh fruits and vegetables at farms before they are harvested. Isolation of Foodborne Pathogens from Fresh Fruits and Vegetables Surveys conducted mostly on retail and postharvest produce samples have found a very low number of fresh fruits and vegetables containing foodborne pathogenic bacteria (Table 4.4). In one such survey, 6 out of 3,826 samples of bagged ready-toeat salad vegetables tested positive for Salmonella (Sagoo and others 2003). These contaminated samples were traced back to a national outbreak of salmonellosis infection in the U.K. Likewise, pathogenic bacteria has been isolated from produce samples that were suspected as vehicles of foodborne outbreaks (MDH 2005). More recently, Johnston and others (2005) reported that among approximately 400 postharvest samples of leafy greens, herbs, and cantaloupe, only 3 cantaloupe samples were contaminated with Salmonella. Studies focusing on the isolation of foodborne pathogens from organic produce have also been published. Several of these studies were carried out in the U.K. and Norway. In one such report, fewer than 100 samples of retail organic vegetables were tested and neither E. coli O157:H7 nor Salmonella was detected in any of the samples (McMahon and Wilson 2001). A larger study in the U.K. analyzed more than 3,000 retail samples of ready-to-eat organic vegetables, and all the samples tested negative for foodborne pathogens such as Salmonella and E. coli O157:H7 (Sagoo and others 2001). Loncarevic and others (2005) reported similar results in the study in Norway, in which 179 organic lettuce samples were analyzed for the presence of Salmonella and E. coli O157:H7. None of these studies, however, assessed the comparative

92

Section II. Preharvest Strategies

Table 4.4. List of testing surveys that analyzed the presence of foodborne pathogens in fresh produce Produce Type Imported

Total Samples 1,003

Leafy Greens # 377

Domestic

1,028

357

Organic

3,200

623

Bagged salad mixes

3,826

3,826

Organic and conventional

605

143

Organic lettuce

179

179

Organic and conventional

2,029

442

Organic and conventional spring mix Leafy greens, cantaloupe

213

213

398

211

Sprouts

200

0

Mexican and domestic

466

175

Bacteria Tested Shigella Salmonella E. coli O157 Shigella Salmonella E. coli O157 Salmonella E. coli O157 Salmonella L. mono cytogenes E. coli O157 Campylobacter Salmonella E. coli O157

Positive Samples # 9 35 0 5 6 0 0 0

Place

Reference

U.S.

FDA 2002

U.S.

FDA 2003

U.K.

Sagoo and others 2001 Sagoo and others 2003

6 1 0 0

U.K.

2 0

MN

Salmonella L. mono cytogenes E. coli O157 Salmonella E. coli O157

0 2 0

Norway

0 0

MN, WI

Salmonella L. mono cytogenes Salmonella L. mono cytogenes E. coli O157 Salmonella L. mono cytogenes E. coli O157 Shigella Salmonella L. mono cytogenes E. coli O157

0 0

CA

3 0 0

Southern U.S.

14 0 3 0 0 3 0

WA

Southern U.S.

Mukherjee and others 2004 Loncarevic and others 2005 Mukherjee and others 2006b Phillips and Harrison 2005 Johnston and others 2005 Samadpour and others 2006 Johnston and others 2005

microbiological quality and safety between organically and conventionally grown fresh produce, because they are cultivated using different farm management practices. The first reports on the comparative evaluation of presence of foodborne pathogens in preharvest fresh produce grown by organic and conventional farmers were

Produce Safety in Organic vs. Conventional Crops

93

published by our research group in 2004 and 2006 (Mukherjee and others 2004, 2006b). From more than 2,600 samples of fresh fruits and vegetables collected directly at the field, we detected Salmonella contamination in only 2 samples grown by farmers who claimed to be organic, but were not certified. None of the organic and conventional produce tested positive for E. coli O157:H7. The U.S. Food and Drug Administration conducted two surveys: one on domestic and the other on imported fruits and vegetables (FDA 2002, 2003). Approximately 1,000 samples of broccoli, cantaloupe, different types of leafy vegetables, strawberries, and tomatoes were collected from 21 different countries, and analyzed for the presence of Shigella, Salmonella, and E. coli O157:H7. As many as 44 samples (4% of the total sample lot) tested positive for Shigella and Salmonella. Almost all of these contaminated samples were either cantaloupe or leafy vegetables and 7.3% of the cantaloupes and 4.9% of leafy vegetables tested positive for either one or both of these two pathogenic bacteria. Only one sample of strawberries and none of the broccoli and tomato samples tested positive for pathogens. None of the 1,003 imported produce samples had E. coli O157:H7 contamination. The survey on domestic fresh produce included samples of the same produce types from 18 states. Eleven samples (1% of the total sample lot) tested positive for Shigella and Salmonella. Out of these 11 contaminated samples, 5 were cantaloupe (3% of the cantaloupes) and 6 were leafy vegetables (1% of the leafy vegetables). Like the survey on imported produce, none of the domestically produced tomatoes were contaminated with pathogens, and none of the 1,028 domestic produce samples tested positive for E. coli O157:H7. Overall, the occurrence of foodborne pathogens in produce testing studies is relatively rare. From the 11 studies listed in Table 4.4, Salmonella was detected in 6 of them, but only 2 studies reported prevalence values greater than 3%. The detection of E. coli O157:H7 in fresh vegetables is an extremely rare event. The first documented isolation of this pathogenic organism from a vegetable sample was accomplished by the Minnesota Department of Agriculture from a sample of bagged lettuce implicated in an outbreak that affected 21 individuals in Minnesota and Oregon (MDH 2005). The only produce survey that has reported the detection of this serotype was a study conducted in Washington (state) when three samples of sprouts tested positive (Samadpour and others 2006). In addition, as many as 13 samples of spinach linked to human cases yielded positive isolations during a large national outbreak in 2006 (CDC 2006). Reliability of Escherichia coli as an Indicator for Fecal Contamination in Produce Ideally, a fecal indicator bacterium should demonstrate occurrence only in intestinal environments and should also possess other criteria for bacterial indicators, such as easy and reliable detection even when present in small numbers (Jay and others 2005). Fecal coliforms and Escherichia coli have been used as a fecal indicator in water by the EPA since the early 1900s (Doyle and Erickson 2006; EPA 1904). During much of the 20th century, use of these bacteria as a fecal indicator has been extended to foods. Over the years, as detection of E. coli became rapid and more reliable, this bacterium has been more frequently used as direct indication of fecal contamination in foods. However, in recent years, some reports have questioned the reliability of E. coli as a fecal indicator in foods, because this bacterium has been documented to

94

Section II. Preharvest Strategies

naturalize environmental niches other than intestinal tract (Ishii and others 2006). Other reports have demonstrated that the reliability of fecal coliforms as a fecal indicator in various types of foods, including fruits and vegetables, was not better than that of E. coli (Dogan-Halkman and others 2003). Over the years, numerous studies have reported that among the fecal coliforms, certain genera such as Klebsiella, Citrobacter, and some Enterobacter are natural inhabitants of fresh fruits and vegetables (Duncan and Razzell 1972; Mukherjee and others 2004, 2006b; Zhao and others 1997). This evidence suggests that, although E. coli can naturalize environments other than intestinal tract, at present there is no other bacterium that could be a more reliable substitute for E. coli as the fecal indicator in fresh fruits and vegetables. Escherichia coli Contamination in Fresh Fruits and Vegetables A number of surveys reported the presence and counts of E. coli in retail samples of fresh produce. However, comparative evaluations of E. coli contamination between organic and conventional produce have rarely been reported. One recent report compared overall bacterial counts and counts of E. coli, coliform, lactic acid bacteria, yeasts, and molds between organic and conventional spring mix collected from a California processor, before washing (Phillips and Harrison 2005). They found that counts of the different bacterial groups, including E. coli, did not differ significantly between the organic and conventional spring mixes. Johnston and others (2005) reported very low average E. coli counts in postharvest produce samples. The large survey involving more than 3,000 retail organic produce samples reported that only 1.5% of the produce had E. coli contamination (Sagoo and others 2001). A vast majority of these contaminated samples had less than 100 CFU/g E. coli counts. A smaller study on retail organic produce in the U.K. didn’t detect any E. coli from any of the 86 samples (McMahon and Wilson 2001). Loncarevic and others (2005) reported considerably greater E. coli prevalence (8.9%) in the study on postharvest, unwashed organic lettuce. In our previously published studies on preharvest organic and conventional produce samples in the upper Midwest, we were able to observe some differences in the prevalence of generic E. coli between organic and conventionally produced vegetables, but these differences were markedly influenced by a larger contribution of lettuce and leafy green samples from organic farms (Mukherjee and others 2004, 2006b). Among the different fresh vegetable types collected at farms, lettuce and leafy greens were significantly more susceptible to E. coli contamination. Because of this factor, we could not conclude that any type of farm management system was more susceptible to E. coli contamination.

Evaluation of Organic Practices That Could Be Linked to Produce Contamination Relatively limited research has been conducted to assess the risk of pathogen contamination of organic management and to reevaluate the current guidelines of existing organic practices. Our research group conducted a statistical analysis on the E. coli prevalence results of the longitudinal preharvest testing survey conducted in Minnesota and Wisconsin, and a few management practices used by organic farmers were associated with increased odds of contamination (Mukherjee and others 2007).

Produce Safety in Organic vs. Conventional Crops

95

The use of animal manure (either composted or raw) for animal production was clearly linked to contamination with generic E. coli (Odds ratio = 13). Among the different types of manure, cattle manure was approximately seven times more likely to be associated with E. coli contamination in produce. The season of manure application (fall or spring) did not appear to increase E. coli prevalence. When manure was aged for more than 6 months before application, it was four times less likely to contaminate produce than in farms that used manure aged for less than 6 months. Researchers at the University of Wisconsin conducted a couple of studies specifically designed to reevaluate the requirement for 120 days of manure application before harvest based on generic E. coli survival (Ingham and others 2004, 2005). In the first study, the count of naturally occurring E. coli declined 3 log CFU/g within 90 days in manure-amended soil from an initial count of 4.4 log CFU/g, but it was still detected after 168 days. Based on the rapid die-off, the researchers indicated that the Organic Rule could be shortened to less than 120 days. In the second study, there was also an initial decline of 2.5 log CFU/g in the E. coli count during the first 7 weeks, but the indicator was still detected in radish, lettuce, and carrot samples. The investigators concluded that this data was not sufficient evidence to reduce the 120-day requirement. A couple of recent publications compared the effect of conventional and organic soil type on the survival of E. coli O157:H7 (Franz and others 2008; Semenov and others 2008). In the first study, there was no significant difference between organic and conventional soils and the range of maximum survival times ranged from 54 to 105 days. In the second report, soil types were paired by organic and conventional categories and the decline in pathogenic E. coli appears to be very irregular in conventional soils as compared to organic, but there were again no significant differences in the overall rate of inactivation.

Epidemiology of Foodborne Disease Linked to Organic Produce In the last 10 years there have been from 1,000 to 1,300 recorded outbreaks of foodborne disease per year in the U.S., but there is no recorded outbreak in the Centers for Disease Control Outbreak Surveillance Data that was confirmed to an organic food (CDC 2007). However, a few reports have linked some infections to consumption of organic fresh produce. In Germany, a rare outbreak of gastroenteritis and hemolytic uremic syndrome were caused by organically grown parsley contaminated with verotoxigenic Citrobacter freundii (Tschape and others 1995). The parsley had been grown in an organic garden fertilized with pig manure, but it is not clear whether this garden had been certified by an organic agency. In a couple of outbreaks of E. coli O157:H7 due to contaminated lettuce, the traceback investigations reported that manure had been used in the fields where the lettuce had been grown, but none of them indicated that the farms were using certified practices (Ackers and others 1998; Hilborn and others 1999). In one of the most serious outbreaks due to fresh produce, bagged spinach produced in California was responsible for more than 200 infections with E. coli

96

Section II. Preharvest Strategies

O157:H7 (CDC 2006). After a thorough investigation by public health authorities, the source of contamination was traced back to a single farm in central California (Jay and others 2007). Further investigation detected the same outbreak strain in a neighboring cattle farm and in feral pigs on the spinach farm. Since the spinach farm was in its second year out of the 3 years required for transition to organic practices the spinach was sold as conventional produce. Based on this outbreak, it is questionable to argue that organic practices were responsible because the farm was yet to be certified.

Summary The continued occurrence of foodborne outbreaks due to contaminated fresh produce is a very serious public health concern. Regardless of whether the implicated produce is conventional or organic, it is critical for any management practices to eliminate pathogen transmission via the fecal-oral route. The widespread adoption of good agricultural practices (GAPs) will be essential to minimize the number of outbreaks. However, the lack of incentives and/or regulations may limit the acceptance of management practices intended to minimize foodborne disease. The reevaluation of the rationale for the current organic regulations for manure use as fertilizer needs to be conducted. This urgency not only stems from the growing importance of organic produce, but from the fact that the current GAPs have adopted almost the same recommendations. From the existing literature discussed in this chapter, a reevaluation is warranted and overdue. Some experts are now predicting a slowdown in the demand for organic foods as consumers are now more interested in local foods and other sustainable production systems. This consumer shift could lead to a wider diversity of management systems, which in turn could lead to less adoption of food safety measures. We have to note here that, to this date, organic produce farmers are the only agriculture system that has a system of oversight as compared to most conventional farmers who have none. The long-term prevention of foodborne disease from any management system will depend on the adoption of comprehensive science-based production and processing practices.

References Ackers M-L, Mahon BE, Damrow T, Hayes PS, Bibb WF, Rice DH, Barrett TJ, Hutwagner L, Griffin PM, Slutsker L. 1998. An outbreak of Escherichia coli O157:H7 infections associated with leaf lettuce consumption. J. Infect. Dis. 177:1588–1593. Baker BP, Benbrook CM, Groth E, Benbrook KL. 2002. Pesticide residues in conventional, integrated pest management (IPM)–grown and organic foods: insights from three U.S. datasets. Food Addit. Contam. 19:427–446. Baloda SB, Christensen L, Trajcevska S. 2001. Persistence of Salmonella enterica serovar Typhimurium DT12 clone in a piggery and in agricultural soil amended with Salmonella-contaminated slurry. Appl. Environ. Microbiol. 67:2859–2862. Bolton DJ, Byrne CM, Sheridan JJ, McDowell DA, Blair IS. 1999. The survival characteristics of a nontoxigenic strain of Escherichia coli O157:H7. J. Appl. Microbiol. 86:407–411. Brinton W, Storms P. 2004. Microbiological test qualities and risk factors for composts prepared from farm-based manures and municipal collected green wastes. In: Todd ECD, editor. First World Congress on Organic Food 2004, Lansing, Michigan. National Food Safety and Toxicology Center, Michigan State University.

Produce Safety in Organic vs. Conventional Crops

97

CDC. 2006. Ongoing Multistate Outbreak of Escherichia coli serotype O157:H7 Infections Associated with Consumption of Fresh Spinach—United States, September 2006. MMWR 55:1–2. ———. 2007. U.S. Foodborne Disease Outbreaks. Foodborne Outbreak Response and Surveillance Unit. Chandler DS, Craven JA. 1980. Relationship of soil moisture to survival of Escherichia coli and Salmonella typhimurium in soils. Aust. J. Agric. Res. 31:547–555. Codex Alimentarius Commission. 1999. Guidelines for the production, processing, labelling, and marketing of organically produced foods. Rome: FAO/WHO. Conford P. 2001. The Origins of the Organic Movement. Glasgow, Great Britain: Floris Books. Dimitri C, Greene C. 2002. Recent growth patterns in the U.S. organic food market. ERS/USDA. pp. 1–15. Dimitri C, Oberholtzer L. 2005. Market-led versus Government-facilitated growth: development of the U.S. and EU organic agricultural sector. ERS/USDA. pp. 1–26. Dogan-Halkman HB, Cakir I, Keven F, Worobo RW, Halkman AK. 2003. Relationship among fecal coliform and Escherichia coli in various foods. Eur. Food Res. Technol. 216:331–334. Doyle MP, Erickson MC. 2006. Closing the door on the fecal coliform assay. Microbe 1(4):162–163. Duncan DW, Razzell WE. 1972. Klebsiella biotypes among coliforms isolated from forest environments and farm produce. Appl. Microbiol. 24:933–938. EPA. 1904. Safe drinking water act. U.S. EPA. FDA. 2002. FDA survey of imported fresh produce; FY 1999 field assignment. Center for Food Safety and Applied Nutrition, FDA. ———. 2003. FDA survey of domestic fresh produce; FY 2000/2001 field assignment. Center for Food Safety and Applied Nutrition, FDA. Forshell LP, Ekesbo I. 1993. Survival of salmonellas in composted and not composted solid animal manures. J. Vet. Med. 40:653–658. Franz E, Semenov AV, Termorshuizen AJ, de Vos OJ, Bokhorst JG, van Bruggen AH. 2008. Manureamended soil characteristics affecting the survival of E. coli O157:H7 in 36 Dutch soils. Environ. Microbiol. 10:313–27. Gagliardi JV, Karns JS. 2002. Persistence of Escherichia coli O157:H7 in soil and on plant roots. Environ. Microbiol. 4:89–96. Hilborn ED, Mermin JH, Mshar PA, Hadler JL, Voetsch A, Wojtkunski C, Swartz M, Mshar R, LambertFair M-A, Farrar JA, and others. 1999. A multistate outbreak of Escherichia coli O157:H7 infections associated with consumption of mesclun lettuce. Arch. Intern. Med. 159:1758–1764. Himathongkham S, Bahari S, Riemann H, Cliver D. 1999. Survival of Escherichia coli O157:H7 and Salmonella typhimurium in cow manure and cow manure slurry. FEMS Microbiol. Lett. 178:251–257. Himathongkham S, Riemann H. 1999. Destruction of Salmonella typhimurium, Escherichia coli O157:H7 and Listeria monocytogenes in chicken manure by drying and/or gassing with ammonia. FEMS Microbiol. Lett. 171:179–182. Himathongkham S, Riemann H, Bahari S, Nuanualsuwan S, Kass P, Cliver DO. 2000. Survival of Salmonella typhimurium and Escherichia coli O157:H7 in poultry manure and manure slurry at sublethal temperatures. Avian Dis. 44:853–860. Howie H, Mukerjee A, Cowden J, Leith J, Reid T. 2003. Investigation of an outbreak of Escherichia coli O157 infection caused by environmental exposure at a scout camp. Epidemiol. Infect. 131:1063–1069. Hutchison ML, Walters LD, Moore A, Crooks KM, Avery SM. 2004. Effect of length of time before incorporation on survival of pathogenic bacteria in livestock wastes applied to agricultural soil. Appl. Environ. Microbiol. 70:5111–5118. Ingham SC, Fanslau MA, Engel RA, Breuer JR, Breuer JE, Wright TH, Reith-Rozelle JK, Zhu J. 2005. Evaluation of fertilization-to-plant and fertilization-to-harvest intervals for safe use of noncomposted bovine manure in Wisconsin vegetable production. J. Food Prot. 68:1134–1142. Ingham SC, Losinski JA, Andrews MP, Breuer JE, Breuer JR, Wood TM, Wright TH. 2004. Escherichia coli contamination of vegetables grown in soils fertilized with noncomposted bovine manure: gardenscale studies. Appl. Environ. Microbiol. 70:6420–6427. Ishii S, Ksoll WB, Hicks RE, Sadowsky MJ. 2006. Presence and growth of naturalized Escherichia coli in temperate soils from Lake Superior watersheds. Appl. Environ. Microbiol. 72:612–621. Islam M, Doyle MP, Phatak SC, Millner P, Jiang X. 2004. Persistence of enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. J. Food Prot. 67:1365–1370.

98

Section II. Preharvest Strategies

———. 2005. Survival of Escherichia coli O157:H7 in soil and on carrots and onions grown in fields treated with contaminated manure compost or irrigation water. Food Microbiol. 22:63–70. Jay JM, Loessner MJ, Golden DA. 2005. Modern food microbiology. New York: Springer Science and Business Media, Inc. Jay MT, Cooley M, Carychao D, Wiscomb GWS, Sweitzer RA, Crawford-Miksza L, Farrar J, Lau DK, O’Connell J, Millington A, Asmundson RV, and others. 2007. Escherichia coli O157:H7 in feral swine near spinach fields and cattle, central California coast. Emerg. Infect. Dis. 13:1908–1911. Jiang X, Morgan J, Doyle MP. 2003. Fate of Escherichia coli O157:H7 during composting of bovine manure in a laboratory-scale bioreactor. J. Food Prot. 66:25–30. Johannessen GS, Bengtsson GB, Heier BT, Bredholt S, Wasteson Y, Rorvik LM. 2005. Potential uptake of Escherichia coli O157:H7 from organic manure into crisphead lettuce. Appl. Environ. Microbiol. 71: 2221–2225. Johannessen GS, Froseth RB, Solemdal L, Jarp J, Wasteson Y, Rorvik LM. 2004. Influence of bovine manure as fertilizer on the bacteriological quality of organic iceberg lettuce. J. Appl. Microbiol. 96:787–794. Johnston LM, Jaykus LA, Moll D, Martinez MC, Anciso J, Mora B, Moe CL. 2005. A field study of the microbiological quality of fresh produce. J Food Prot. 68:1840–1847. Kato S, Jenkins M, Fogarty E, Bowman D. 2004. Cryptosporidium parvum oocyst inactivation in field soil and its relation to soil characteristics: analyses using the geographic information systems. Sci. Total Environ. 321:47–58. Kudva IT, Blanch K, Hovde CJ. 1998. Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Appl. Environ. Microbiol. 64:3166–3174. Lang MM, Harris LJ, Beuchat LR. 2004. Survival and recovery of Escherichia coli O157:H7, Salmonella and Listeria monocytogenes on lettuce and parsley as affected by method of inoculation, time between inoculation and analysis, and treatment with chlorinated water. J. Food Prot. 67:1092–1103. Lau MM, Ingham SC. 2001. Survival of fecal indicator bacteria in bovine manure incorporated into soil. Lett. Appl. Microbiol. 33:131–136. Loncarevic S, Johannessen GS, Rorvik LM. 2005. Bacteriological quality of organically grown leaf lettuce in Norway. Lett. Appl. Microbiol. 41:186–189. Lung AJ, Lin C-M, Kim JM, Marshall MR, Nordstedt R, Thomson NP, Wei CI. 2001. Destruction of Escherichia coli O157:H7 and Salmonella entritidis in cow manure composting. J. Food Prot. 64:1309–1314. Magkos F, Arvaniti F, Zampelas A. 2006. Organic food: buying more safety or just peace of mind? A critical review of the literature. Crit. Rev. Food Sci. Nutr. 46:23–56. Maule A. 2000. Survival of verotoxigenic Escherichia coli O157 in soil, water and on surfaces. J. Appl. Microbiol. Symp. Suppl. 88:71S–78S. McMahon MAS, Wilson IG. 2001. The occurrence of enteric pathogens and Aeromonas species in organic vegetables. Int. J. Food Microbiol. 70:155–162. MDH. 2005. Health officials investigate E. coli O157:H7 cases related to Dole prepackaged lettuce mixes sold at Rainbow Foods. Minneapolis: Minnesota Department of Health. Mukherjee A, Cho S, Scheftel J, Jawahir S, Smith K, Diez-Gonzalez F. 2006a. Soil survival of Escherichia coli O157:H7 acquired by a child from garden soil recently fertilized with cattle manure. J. Appl. Microbiol 101:429–436. Mukherjee A, Speh D, Diez-Gonzalez F. 2007. Association of farm management practices with risk of Escherichia coli contamination in pre-harvest produce grown in Minnesota and Wisconsin. Int. J. Food Microbiol. 120:296–302. Mukherjee A, Speh D, Dyck EA, Diez-Gonzalez F. 2004. Pre-Harvest Evaluation of coliforms, Escherichia coli, Salmonella and E. coli O157:H7 in organic and conventional produce grown by Minnesota farmers. J. Food Prot. 67:894–900. Mukherjee A, Speh D, Jones AT, Buesing KM, Diez-Gonzalez F. 2006b. Longitudinal microbiological survey of fresh produce grown by farmers in the Upper Midwest. J. Food Prot. 69:1928–1936. Natvig EE, Ingham SC, Ingham BH, Cooperband LR, Roper TR. 2002. Salmonella enterica serovar Typhimurium and Escherichia coli contamination of root and leaf vegetables grown in soils with incorporated bovine manure. Appl. Environ. Microbiol. 68:2737–2744. Nicholson FA, Groves SJ, Chambers BJ. 2005. Pathogen survival during livestock manure storage and following land application. Bioresour. Technol. 96:135–143.

Produce Safety in Organic vs. Conventional Crops

99

NOP/USDA. 2002a. National list of allowed and prohibited substances. USDA. ———. 2002b. Organic production and handling standards. AMS/USDA. Oberholtzer L, Dimitri C, Greene C. 2005. Price premiums hold on as U. S. organic produce market expands. ERS/USDA. pp. 1–22. Ogden ID, Hepburn NF, MacRae M, Strachan NJ, Fenlon DR, Rusbridge SM, TH. P. 2002. Long-term survival of Escherichia coli O157 on pasture following an outbreak associated with sheep at a scout camp. Lett. Appl. Microbiol. 34:100–104. Ogden LD, Fenlon DR, Vinten AJ, Lewis D. 2001. The fate of Escherichia coli O157 in soil and its potential to contaminate drinking water. Int. J. Food Microbiol. 66:111–117. Phillips CA, Harrison MA. 2005. Comparison of the organically and conventionally grown spring mix from a California grower. J. Food Prot. 68:1143–1146. Pussemier L, Larondelle Y, Van Peteghem C, Huyghebaert A. 2006. Chemical safety of conventionally and organically produced foodstuffs: a tentative comparison under Belgian conditions. Food Control 17:14–21. Rangel JM, Sparling PH, Crowe C, Griffin PM, Swerdlow DL. 2005. Epidemiology of Escherichia coli O157:H7 outbreaks, United States, 1982–2002. Emerg. Infect. Dis. 11:603–609. Sagoo SK, Little CL, Mitchell RT. 2001. The microbiological examination of ready-to-eat organic vegetables from retail establishments in the United Kingdom. Lett. Appl. Microbiol. 33:434–439. Sagoo SK, Little CL, Ward L, Gillespie IA, Mitchell RT. 2003. Microbiological study of ready-to-eat salad vegetables from retail establishments uncovers a national outbreak of Salmonellosis. J. Food Prot. 66:403–409. Samadpour M, Barbour MW, Nguyen T, Cao TM, Buck F, Depavia GA, Mazengia E, Yang P, Alfi D, Lopes M, and others. 2006. Incidence of enterohemorrhagic Escherichia coli, Escherichia coli O157, Salmonella, and Listeria monocytogenes in retail fresh ground beef, sprouts, and mushrooms. J. Food Prot. 69:441–443. Semenov AV, Franz E, van Overbeek L, Termorshuizen AJ, van Bruggen AH. 2008. Estimating the stability of Escherichia coli O157:H7 survival in manure-amended soils with different management histories. Environ. Microbiol. 10:1450–1459. Shepherd MWJ, Liang P, Jiang X, Doyle MP, Erickson MC. 2007. Fate of Escherichia coli O157:H7 during on-farm dairy manure-based composting. J. Food Prot. 70:2708–2716. Sideman, E. 2006. Personal communication. Sivapalasingam S, Friedman CR, Cohen L, Tauxe RV. 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J. Food Prot. 67:2342–2353. Strachan NJ, Fenlon DR, Ogden ID. 2001. Modeling the vector pathway and infection of humans in an environmental outbreak of Escherichia coli O157. FEMS Microbiol. Lett. 203:69–73. Tschape H, Prager R, Streckel W, Fruth A, Tietze E. 1995. Verotoxinogenic Citrobacter freundii associated with severe gastroenteritis and cases of haemolytic uraemic syndrome in a nursery school: green butter as the infection source. Epidemiol. Infect. 114:441–450. Wang G, Zhao T, Doyle MP. 1996. Fate of enterohemorrhagic Escherichia coli O157:H7 in bovine feces. Appl. Environ. Microbiol. 62:2567–2570. Warriner K. 2005. Pathogens in vegetables. In: Jogen W, editor. Improving the Safety of Fresh Fruit and Vegetables. Boca Raton, FL: CRC Press. pp. 3–43. Winter CK, Davis SF. 2006. Organic foods. Food Technol. 71:R117–R124. You Y, Rankin SC, Aceto HW, Benson CE, Toth JD, Dou Z. 2006. Survival of Salmonella enterica serovar Newport in manure and manure-amended soils. Appl. Environ. Microbiol. 72:5777–5783. Zhao T, Clavero MRS, Doyle MP, Beuchat LR. 1997. Health relevance of the presence of fecal coliforms in iced tea and leaf tea. J. Food Prot. 60:215–218.

5

The Role of Good Agricultural Practices in Produce Safety Robert B. Gravani

Introduction Fresh fruits and vegetables are an important component in the diets of people throughout the world. Nutritionists and health professionals have shown that diets low in fat and high in fiber, with at least five servings of fruits and vegetables are protective against many types of cancers, diabetes, and possibly other chronic diseases. Fruits and vegetables are excellent sources of many vitamins (A, C, E, and K), micronutrients (folate, potassium, and magnesium), and phytochemicals that play key roles as antioxidants (Brown 2008). MyPyramid, developed by the U.S. Department of Agriculture (USDA), (USDA a. 2005) recommends 1.5–2 cups of fruit and 2.5–3 cups of vegetables daily for adults each day (based on age, gender, and physical activity), while a recent WHO/FAO report recommends a minimum of 400 grams of fruits and vegetables daily (WHO 2005). As countries develop strategies to reduce the number of overweight and obese individuals in their populations, increased fruit and vegetable consumption can help displace foods high in saturated fats, sugar, and salt. Advances in agronomic practices, preservation technologies, packaging, temperature control and cold chain management, and shipping, combined with improvements in marketing and merchandising strategies, have resulted in increased global production and distribution of fresh fruits and vegetables (National Research Council 2003). Today, many fresh produce items are available year-round. While shopping in a typical retail food store, consumers have a wide variety of products to choose from and can select from an average of 345 different produce items that come from over 130 countries around the world (Rangurajan and others 1999). U.S. consumers have listened to the positive public messages about produce being a key component of a healthy diet and have increased their consumption of these foods. Since 1970, the per capita consumption of fresh vegetables increased from 150 to nearly 200 pounds in 2006, while fresh fruit consumption during the same time period increased from 101 to 129 pounds. Overall, fresh fruit and vegetable consumption in the U.S. in 2006 was more than 325 pounds per person (USDA b. 2008)! As produce consumption was increasing, epidemiologists at the Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia, noticed another important trend that was developing. From the early 1970s to the present, there was a significant increase in the number of foodborne outbreaks associated with fresh produce. The national data that CDC was studying indicated that: 101

102

Section II. Preharvest Strategies

• The number of fresh produce-associated outbreaks steadily increased and more than doubled. • The number of people affected in these outbreaks more than doubled. • Produce outbreaks, as a proportion of outbreaks associated with known vehicles, also increased. • A wide variety of fresh produce items were involved in the outbreaks. • Bacteria, viruses, and parasites were identified as causative agents. • Both domestic and imported fresh produce items were implicated in outbreaks. • Due to complicated packing and repacking schemes, as well as the perishability of the products, tracing outbreaks back to their source was very difficult. Figure 5.1 illustrates the increase in produce-associated outbreaks by category of causative agent and by decade from the 1970s through the 1990s (Tauxe 2008). Figure 5.2 depicts the number of produce-associated outbreaks (72) and illness (8,791) in the U.S. by year, from 1996 to 2008, compiled by the Center for Food Safety and Applied Nutrition (CFSAN) of the U.S. Food and Drug Administration (FDA) in 2008 (Vierk 2009). The data in Figure 5.2 and all subsequent data compiled for produce-associated foodborne illnesses by FDA 1) represents only those outbreaks and illnesses associated with FDA-regulated foods, 2) do not contain information on outbreaks/illnesses where the point of contamination is a retail food setting or home, 3) do not include illnesses transmitted from person-to-person, and 4) do not include illness that may have occurred but were not recorded. From the FDA data provided in Figure 5.2, it can be seen that produce-associated outbreaks have fluctuated from between 1 to 10 incidents per year from 1996 to 2008 and have involved from 57 to 1,702 people (Vierk 2009). According to the FDA,

Figure 5.1. 2008).

Produce-associated outbreaks by causative agent and decade (Tauxe

The Role of Good Agricultural Practices

Outbreaks (N = 82)

Illnesses (N = 10,505) 1800 1702

1533

10

10 1389 # of outbreaks

10

9 8

8

1161

1600 1400 1200

7 6 4 3

6

5

6

1000

6

7 946 951 725

554

800 694

# of illnesses

12

4

103

600 400

2

368 173

0

244

65

1

200 0

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Figure 5.2.

Produce outbreaks and illnesses by year.

in 1996 produce was associated with about 4.3% of all outbreaks (produce and nonproduce) in the U.S. and 45.5% of illnesses; in 2007, produce was associated with 4.2% of outbreaks and 12.5% of illnesses (Vierk 2008). The Center for Science in the Public Interest (CSPI) recently published a report on produce outbreaks from 1990–2005 (Smith DeWaal and Bhuiya 2008). They indicated that between 1990 and 2005, there were 713 outbreaks and 34,049 cases of illness linked to produce in their database and with an average of 48 individuals afflicted per outbreak. According to CSPI, produce outbreaks accounted for 13% of foodborne illness outbreaks and 21% of illnesses in their database (Smith DeWaal and Bhuiya 2008). Clearly, outbreaks and illnesses linked to produce continue to be a major health and safety concern in the U.S. Produce-related outbreaks are much different than “traditional or classical” foodborne outbreaks. In a typical outbreak scenario, there is an error involving a food preparation practice in a commercial kitchen that results in the bacterial contamination of food that is served to a group of people within a community, who all come down with similar symptoms of the illness in a relatively short period of time. This type of outbreak is relatively easy to investigate, since it involves a relatively small number of people from a specific geographic area. In contrast, produce outbreaks present significantly greater challenges to scientists because they are much more complex and difficult to investigate (Guzewich 2008). Produce items are grown in different areas of the country or are imported from other countries. A very complex national distribution system results in fresh produce being shipped all over the country—and in some cases, internationally. Consumers who become ill are usually scattered across the country and it often takes some time to determine that the illnesses are due to the consumption of a common produce item. In addition, the complexity of the distribution system, where commodities from different growers are often packed and repacked

104

Section II. Preharvest Strategies

at packinghouses and regional markets, results in commingling of produce items from different farms, and makes traceback and recall of contaminated produce items very difficult. With fresh-cut produce items (like leafy greens), this issue is a real concern, since a batch of contaminated product from one farm can contaminate very large amounts of product from a variety of other farms during processing. In addition, when regulatory agency officials are investigating the cause of outbreaks, they often cannot locate the product due to its perishability, since it has either been consumed or spoiled and discarded (Guzewich 2008). In either case, the suspect produce item is often no longer available to sample and test for pathogens. When officials go to farms to investigate, they find that in many cases, the suspect crops have already been harvested and there is no remaining produce from the suspected fields. So, produce outbreaks are very difficult to trace back to their sources and in many outbreak investigations, the route of produce contamination is often suspected, but never scientifically proven. Fresh fruits and vegetables can become contaminated by biological hazards, such as pathogenic organisms including bacteria, viruses and parasites, chemical hazards, and physical hazards. These hazards have all caused illness or injury in fresh produce. Biological hazards, particularly pathogenic bacteria, are the greatest concern because the risk that they pose may be amplified by growth prior to consumption (National Research Council 2003) and have been responsible for most of the produce associated outbreaks in the U.S. This chapter focuses primarily on the reduction of these hazards through the implementation of good agricultural practices (GAPs).

The Nature of Pathogen Contamination Produce can become contaminated from a variety of production and harvesting techniques on the farm, through washing and sorting in the packinghouse, through distribution in retail food stores and food-service facilities, and in the home. Contamination can result from contact with the soil; manure; improperly composted manure; irrigation water; fecal material from wild and domestic animals; farm, packinghouse, and terminal market workers; contaminated equipment in fields, packinghouse, and distribution system; wash, rinse, and flume water; ice; cooling equipment and transportation vehicles; cross-contamination from other foods; and improper storage, packaging, display, and preparation (Bihn and Gravani 2006; FDA, USDA, and CDC 1998). Produce-associated outbreaks have been caused by a variety of bacterial pathogens including Salmonella species, E. coli O157:H7, Shigella, Campylobacter, Bacillus cereus, the parasites Crytosporidium and Cyclospora, and viruses hepatitis A and norovirus. Table 5.1 shows a list of pathogens that have been implicated in produceassociated outbreaks from 1994–2004 (Tauxe 2008). Table 5.2 contains data available from the FDA and provides a list of fresh produce items that have been contaminated and associated with outbreaks in the U.S. from 1996 to 2008, such as leafy greens, including lettuce, romaine, spinach, mixed lettuce, tomatoes, cantaloupes and other melons, raspberries and frozen strawberries, green onions (scallions), and basil and basil-containing products (Guzewich 2008; Vierk 2009). Figure 5.3 illustrates the types of produce, by percentage, associated with the 82 outbreaks in the U.S. between 1996 and 2008 (Vierk 2008, 2009). It is noteworthy

The Role of Good Agricultural Practices

105

Table 5.1. Foodborne Outbreaks Associated with Fresh Produce, by Causative Agent, 1998–2004* 䉬 Bacterial

97

ⴰ Salmonella

53

ⴰ E. coli 0157 ⴰ Shigella ⴰ Campylobacter ⴰ Other 䉬 Viral ⴰ Calicivirus/ Norovirus ⴰ Hepatitis A

19 6 6 13 81 73

䉬 Parasite ⴰ Cyclospora ⴰ Other 䉬 Chemical

6 5 1 6

8

Tauxe (2008). * electronic Foodborne Outbreak Reporting System (eFORS), preliminary analysis and subject to change. The causative agent was identified in 190 (48%) of the 384 produce-associated outbreaks during this time period.

Table 5.2. Produce Outbreaks by Commodity 1996–2008 Tomatoes Lettuce Romaine lettuce Mixed lettuce Cabbage Spinach Cantaloupe Melons Honeydew melon Raspberries/berries Green onions Total = 82 outbreaks

14 15 6 1 1 3 9 2 2 7 3

Jalapeno/Serrano Mango Almonds Parsley Basil Cilantro Green grapes Snow Peas Basil or Mesclun Squash Unknown

1 2 2 2 4 1 1 1 2 1 2 As of May, 2008

Guzewich 2008; Vierk 2009.

that leafy greens, tomatoes, cantaloupe, herbs (such as basil and parsley), and berries were responsible for approximately 78% of the produce-associated outbreaks during this time period. To reduce the likelihood of illnesses, it is important to understand the sources (or reservoirs for the pathogenic microorganisms implicated in produce-associated outbreaks as compiled in Table 5.3 (Guzewich 2008; Vierk 2009). From 1996–2008, zoonotic (animal) reservoirs were responsible for outbreaks from E. coli O157:H7 and Salmonella species; human reservoirs were responsible for Cyclospora, hepatitis A, and Shigella outbreaks. The source of the zoonotic reservoirs involved in outbreaks included 35 that were domestic, 9 foreign, and 15 unknown for a total of 59; the source of human reservoirs were 1 that was domestic, 11 foreign, and 10 unknown for a total of 22 (Guzewich 2008; Vierk 2009). Produce has always been considered a very safe food, and it still is, but many people wonder why there has been an increase in the number of produce-associated

106

Section II. Preharvest Strategies

9.8%

1.2%

2.4%

8.5% 3.7% Berries 8.5%

Green onions Herbs Leafy greens Melons

17.1%

32.9%

15.9% Figure 5.3.

Tomatoes Others Multiple Unknown

Types of produce associated with outbreaks, 1996–2008 (N = 82).

Table 5.3. Produce outbreak reservoirs, 1996–2008 (N = 82*) Zoonotic

Human

25

E coli 0157:H7

17

Cyclospora

34

Salmonella sp

3 2

Hepatitis A Shigella

35 9 15 59

Domestic Foreign Unknown TOTAL

Source 1 11 10 22

Domestic Foreign Unknown TOTAL As of May, 2008

Guzewich 2008; Vierk 2009. *One outbreak due to cucrubitacic toxin.

foodborne illnesses. There are many contributory factors involved, and they include those discussed in the following sections (Rangurajan and others 1999). Changing Demographics People are living longer than ever before and as a result, many people in the U.S. are elderly, have weakened immune systems, or are suffering from chronic diseases. It is estimated that approximately 25–30% of the U.S. population has a weakened or compromised immune system (Morris and Potter 1997; Smith 1997). People ingesting immunosuppressive agents due to chemotherapy treatments or to prevent the rejection of organ transplants, people who suffer from late-stage HIV infections or AIDS, pregnant women, and young children are more vulnerable to foodborne illnesses than are healthy, nonimmunocompromised people, and they are at an increased risk for serious foodborne illnesses (Rangurajan and others 1999).

The Role of Good Agricultural Practices

107

Changing Food Systems A wide variety of fresh fruits and vegetables are grown, transported, and distributed from numerous farms and orchards in the U.S. and from countries throughout the world. Although this complex food system provides a greater diversity of produce items available to large numbers of people, it also potentially increases the exposure of more consumers to different types of microorganisms on produce. In our complex food supply chain, with the multiple handling of produce during harvesting, sorting, washing, transportation, and storage, there is a greater chance of product contamination and temperature abuse to occur. Fresh fruits and vegetables are often eaten raw, without cooking, so they are not subject to a lethal heat treatment to kill pathogenic microorganisms that may be present. When an outbreak occurs, it is often very difficult to trace the source of the problem. Changing Consumer Preferences The growing popularity of salad bars and the increase in the number of meals eaten outside the home can increase the risk of contamination through poor food handling and preparation practices. Minimally processed produce, such as fresh-cut fruits and vegetables, although convenient, has not been heat treated to kill pathogens. If this processing is followed by long storage periods, especially at warm temperatures, pathogens that may be present can survive and grow, increasing the risk of foodborne illness (Rangurajan and others 1999). Poor home food preparation practices can also contribute to an increased risk of produce contamination. When consumers prepare raw foods of animal origin on a cutting board, and then reuse the cutting board to prepare fresh produce without having first thoroughly cleaned and sanitized it, there is an increased risk of foodborne illness in those people who consume the raw produce item. Following all of the principles of safe home food preparation, including the tips for the safe handling of fresh produce developed by the FightBac campaign (Partnership for Food Safety Education 2004) will reduce the risk of foodborne illness. Attention to detail; proper cleaning and sanitizing of all equipment and utensils used to prepare foods; and good personal hygiene and frequent hand washing before, during, and after food preparation are important considerations for all who prepare food for others. More detailed information about consumer and food-service handling of fresh produce is provided in chapter 15. Changing Microorganisms Over the last 30–40 years, microbiologists have observed many genetic changes in microorganisms. These changes include adaptation to stresses in the environment, allowing microorganisms to survive and grow where they once could not survive. Bacteria such as Yersinia enterocolitica and Listeria monocytogenes are capable of growing slowly at refrigerator temperatures and some bacteria, such as Salmonella enteritidis and E. coli O157:H7, can cause serious human illness when only a small number of cells are ingested. Microbiologists are actively studying these adaptive stress responses to learn more about how these specialized mechanisms work, so that better control strategies can be developed. Pathogens that attach to fruits and vegetables are very difficult to remove, so it is important that prevention strategies are designed to keep these microorganisms from contaminating produce.

108

Section II. Preharvest Strategies

The Economic Impact of Produce Outbreaks When there is a produce-associated outbreak, it affects all produce consumption. Even when outbreaks occur in areas far from where they live, consumers will avoid the consumption of the produce item involved in the outbreak. In fact, they often reduce their consumption of all fresh produce. In 1996, a large national outbreak involving Cyclospora cayetanensis–contaminated imported raspberries resulted in the strawberry industry losing an estimated $50 million after mistakenly being identified as the source of the pathogen early in the investigation (Bihn and Gravani 2006). In 2006, due to the large national outbreak of E. coli O157:H7–contaminated packaged spinach, the spinach industry lost an estimated $75 million. U.S. sales of packaged spinach were off 53% for the 16-week period after that outbreak, which began in August–September 2006. In March 2007, sales were still down 14% versus before the outbreak. Sales of all packaged salads (not containing spinach) were down 14% in the 16-week period after the outbreak and were still down about 1% in March 2007 (Pool 2007). The recent multistate Salmonella St. Paul outbreak that sickened over 1,284 people in 43 states in 2007, and was initially thought to be associated with Roma and Red Round tomatoes, but was later traced to jalapeno/ serrano peppers, cost the tomato industry an estimated $100 million. These are some of the many examples indicating how the produce industry experiences a considerable economic impact and a loss of reputation and consumer trust when produce-associated outbreaks occur. Additional detailed information about the economic impact of produce outbreaks is contained in chapter 22.

The Development and Implementation of Good Agricultural Practices (GAPs) Produce poses special challenges because of the overall complexity of the industry, the variety of fruits and vegetables grown, the multitude of production conditions and strategies in different parts of the country, including soil types, water sources, irrigation methods, feral animals, proximity to domestic animal operations, harvest techniques, equipment, transportation, water use, packing, distribution, and market outlets. Some commodities have natural characteristics that make them more susceptible to microbial contamination and have been implicated in more outbreaks than some others (Bihn and Gravani 2006). Since each farming operation is unique, growers need to carefully evaluate every phase of their operation and develop a specific food safety plan (program) that addresses the hazards and risks that are present (Bihn and Gravani 2006). In 1998, the FDA released the Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables (FDA, USDA, and CDC 1998). The Guide outlines a set of production practices called good agricultural practices (GAPs) and intervention strategies that can be implemented on farms in the production of fresh produce. GAPs are analogous to the good manufacturing practices (GMPs) used in the food-processing industry, but address agricultural activities designed to reduce microbial hazards and risks. Simply stated, GAPs are the basic environmental and operational conditions necessary for the production of safe and wholesome fresh fruits and vegetables. GAPs can also be thought of as any operational or management practice that reduces microbial hazards in the growing, harvesting, sorting, packing, and

The Role of Good Agricultural Practices

109

storing of fresh fruits and vegetables. The FDA Guide provides a broad range of guidelines that address many of the risk areas that growers need to evaluate before implementing a farm food safety plan and encourages growers to use proper management strategies to minimize potential contamination. The Guide was written by the FDA in collaboration with USDA and CDC to provide guidance and recommendations about GAPs to fruit and vegetable growers; it is not a codified regulation. The Guide provides the following eight principles of microbial food safety that can be applied to the growing harvesting, packing, and transportation of fresh fruits and vegetables (FDA, USDA and CDC 1998; Beru and Salsbury 2002; Bihn and Gravani 2006). Principle 1: The prevention of microbial contamination of fresh produce is favored over reliance on corrective actions once contamination has occurred. Principle 2: To minimize microbial food safety hazards for fresh produce, growers or packers should use GAPs in those areas over which they have a degree of control while not increasing other risks to the food supply or the environment. Principle 3: Anything that comes in contact with fresh produce has the potential of contaminating it. For most foodborne pathogens associated with produce, the major source of contamination is human or animal feces. Principle 4: Whenever water comes in contact with fresh produce, its source and quality dictate the potential for contamination. Principle 5: Agricultural practices using manure or municipal biosolid wastes should be closely managed to minimize the potential for microbial contamination of fresh produce. Principle 6: Worker hygiene and sanitation practices during production, harvesting, sorting, packing, and transportation play a critical role in minimizing the potential for microbial contamination of fresh produce. Principle 7: Follow all applicable local, state, and federal laws and regulations, or corresponding or similar laws, regulations, or standards for agricultural practices for operations outside the United States. Principle 8: Accountability at all levels of the agricultural environment (farms, packing facilities, distribution centers, and transportation operations) is important to a successful food safety program. There must be qualified personnel and effective supervision to ensure that all elements of the program function correctly and to help track produce back through the distribution channels to the producer. Because a wide variety of fruits and vegetables are grown using diverse production methods, the principles in the Guide are quite general. The Guide focuses only on microbial hazards for fresh produce and does not address pesticide residues or chemical contaminants. Prevention is the key to reducing the microbial contamination of fresh fruits and vegetables, and the Guide emphasizes that growers should focus on risk reduction strategies, not risk elimination, because present technologies cannot eliminate all potential food safety hazards associated with fresh produce that will be eaten raw (FDA, USDA and CDC 1998; Beru and Salsbury 2002; Bihn and Gravani 2006). In 2000, several large food retailers around the country began writing letters to their produce suppliers requiring that the fresh fruits and vegetables that were being supplied to them be grown in accordance with GAPs (Bihn and Gravani 2006). In addition to requiring GAPs, many buyers introduced purchasing specifications, letters

110

Section II. Preharvest Strategies

of guarantee, and supplier certification programs and also required independent thirdparty audits of farms and packinghouses to assure that GAPs were being implemented (Institute of Food Technologists 2001; International Fresh-Cut Produce Association 2001; National Research Council 2003; Bihn and Gravani 2006). This concept of produce supplier quality and safety assurance was widely adopted (with variations) by food retailers throughout the U.S. These produce-buying arrangements are still in effect today and many retailers are not only requiring large produce suppliers to comply with them, but are also requiring implementation of GAPs and third-party audits of small, local produce growers. The role of third-party farm and packinghouse audits in produce safety is discussed in chapter 17.

Components of GAPs The concept of GAPs and their importance in produce safety can be visualized in Figure 5.4, through the Produce Safety Assurance Pyramid (Gravani 2006). The pyramid consists of four main components (Bihn and Gravani 2006). At the peak of the pyramid, produce safety depends on the total commitment and leadership of top management in the farm operation. Management must believe in and promote the importance of food safety in their business and provide the necessary resources (finances, personnel, training, equipment, etc.) to achieve this goal. Without a total commitment from upper management, food safety will never become a top priority for employees in the company. Many companies in the food system have built a culture of food safety and foster it among all employees throughout their organizations.

Produce Safety Assurance Pyramid

Total Management Commitment Produce Safety Assurance Biological Hazards

Education and Training

Good Agricultural Practices Microbial Manure Use Worker Health Water Quality & Composting & Hygiene

Cleaning & Sanitation

Animal & Pest Management

Farm Biosecurity

Recall & Traceback

Crisis Management

© Robert B. Gravanl

Figure 5.4.

Produce safety assurance pyramid.

The Role of Good Agricultural Practices

111

The next component essential for produce safety is a well-designed and effective education and training program that is provided to employees throughout the company. Farm, packinghouse, and transportation workers who possess the knowledge, skills, and motivation to perform their job tasks properly, correctly, and consistently will reduce microbial risks. The expectation of job responsibilities and the importance of food safety requirements must be clearly communicated to all workers, and appropriate resources necessary to allow them to do their jobs properly must be provided. The education and training of workers is often overlooked or neglected and not given the importance that it deserves, and this can result in significant lapses in the safety of fresh produce. The third component in the top portion of the pyramid is a thorough understanding of biological hazards (including bacteria, viruses, and parasites) that have been involved in produce-associated outbreaks; this is a key factor in controlling them. Hazards that are not properly identified or understood cannot be easily addressed (Bihn and Gravani 2006). The safety of fresh fruits and vegetables is dependent on these three components of the Produce Safety Assurance Pyramid which are built on the fourth component, a strong and solid foundation of GAPs, including many practices that control hazards on the farm. The eight foundation blocks that make up GAPs are microbial water quality, manure use and composting, worker health and hygiene, cleaning and sanitation, animal and pest management, recall and traceback strategies, crisis management, and farm biosecurity (Bihn and Gravani 2006). Although the latter three foundation blocks are not agricultural practices, they focus on areas of concern that should be addressed in a farm food safety plan. Therefore in developing an effective food safety plan, all of these areas must be addressed and a well-organized record-keeping system must be established, implemented, and maintained to document relevant activities. If a strong foundation of GAPs is not well established and actively maintained, the safety of fresh fruits and vegetables can be compromised. Each of these important foundation blocks of GAPs is briefly addressed in the following sections. Microbial Water Quality The microbiological quality of water used in the irrigation, washing, sorting, and packing of fresh fruits and vegetables as well as for crop protection is of critical importance in assuring the safety of these foods. At the present time, there are no national standards in the U.S. for microbiological quality of irrigation water, but there are several grower groups and produce trade associations that are addressing this important issue. There are some states, grower groups, and marketing orders that are making recommendations on the level of microorganisms that should be considered for irrigation water. These are derived from a number of sources, but clearly more research is needed to determine the appropriate level of microorganisms present in irrigation water. The role of water and water testing in produce safety is addressed in chapter 7. Manure Use and Composting When manure is used as a soil amendment, it provides a variety of benefits to the soil by increasing the organic carbon and overall organic matter, and this results in

112

Section II. Preharvest Strategies

improved granulation, water infiltration, water holding capacity, nutrient content, soil biota activity, soil fertility, soil tilth, and overall productivity (National Research Council 1989). Using manure also serves the animal production industry as a way to manage animal waste (Bihn and Gravani 2006). Raw animal manure from cattle, sheep, swine, and chickens may contain a variety of pathogens, and, if not properly applied to fields, can contaminate fruits and vegetables that are grown there. Abandoning manure use to reduce risk is one option, but the best option is to understand the risks that exist and minimize these risks through GAPs implementation (Bihn and Gravani 2006). The USDA National Organic Program guidelines recommend a 90-day interval for crops whose edible portion does not come in contact with the soil and a 120-day interval between the application of raw manure and harvest of crops whose edible portion has direct contact with the soil (National Organic Standards Board 2002). Another option is to properly compost raw manure to reduce the pathogen load prior to soil application. The role of manure and compost in produce safety is discussed in detail in chapter 8. Worker Health and Hygiene The health and hygiene of all workers who handle fresh produce, whether it’s on the farm, or in the packinghouse, terminal market, grocery store, or foodservice operation is of paramount importance in preventing produce-associated outbreaks. Organisms such as Shigella spp., E.coli O157:H7, Hepatitis A virus, and norovirus can be easily spread to produce via the fecal-oral route of transmission from infected workers who work when they are ill. Strategies for preventing contamination by workers involve well-designed and -delivered education and training programs that include information on the importance of good health and hygiene to produce safety, proper use of field and packinghouse toilets, effective hand-washing practices, and the appropriate use of gloves. Even though discussing urination and defecation are difficult topics to address, it is vital that workers understand their role in preventing the contamination of the produce they handle. It is also important for workers to be reminded that they are handling ready-to-eat products and are considered food handlers! Growers and packinghouse managers are responsible for providing clean, sanitary, and well-stocked toilets, with adequate hand-washing facilities containing water, soap, and single-use paper towels that are in close proximity to where people are working. It is not only a legal requirement, but it is a matter of common decency and privacy to provide clean and sanitary toilets and hand-washing facilities to all workers (Bihn and Gravani 2006). There are many educational resources available in several languages that can be used in worker education and training programs (National Good Agricultural Practices Program 2008). Additional information about worker health and hygiene can be found in chapter 16. Cleaning and Sanitation There are a number of key areas throughout the farm-to-table produce supply chain where effective cleaning and sanitation must be practiced, but this discussion will be limited to cleaning and sanitation in the packinghouse. Inside the packinghouse, there are several areas to consider when addressing product safety (Bihn and Gravani 2006). First, the flow of product and personnel is

The Role of Good Agricultural Practices

113

important, as is the equipment used to wash, sort, grade, and pack the produce being harvested. Equipment should be made of smooth materials that are impervious to dirt and moisture and easily cleanable. The facility should be designed and constructed to minimize product contamination and facilitate the cleaning and sanitizing process. All equipment should be installed with enough space around it (from walls and other equipment) to facilitate easy disassembly, cleaning, and sanitizing. The surfaces of all equipment that come in contact with produce can accumulate plant residues, exudates and juices, soil, debris, and other materials and can become a source of contamination. Biofilms, which are an accumulation of bacterial cells and food debris, can form on food contact surfaces if proper and thorough cleaning and sanitizing procedures are not performed on a regular basis (Bihn and Gravani 2006). Once biofilms form on surfaces, they are difficult to remove, so a master schedule of cleaning and sanitizing procedures must be developed, implemented, maintained, and properly recorded for equipment and environmental surfaces throughout the facility. Because there are a wide variety of cleaners and sanitizers on the market today, it is recommended that growers and packinghouse operators contact chemical suppliers for advice on their specific cleaning and sanitizing needs (Bihn and Gravani 2006). Employee facilities such as locker rooms, toilets, and lunchrooms should be properly lit, clean, orderly, in good repair, and well maintained. These facilities should be located in an area of the packinghouse where the contamination of produce and equipment cannot occur (Bihn and Gravani 2006). Since toilets and hand-washing facilities are vital to the hygiene of employees and to produce safety, they should also be cleaned on a regular basis. The reader is referred to chapter 16. Animal and Pest Management Because most fruits and vegetables are grown in soil and outdoors, contamination can occur from many sources. The data in Table 5.3 indicate the need for domestic and feral animal control in fields and pest control in the packinghouse. There are several examples of produce-associated outbreaks where traceback investigations have showed that animal intrusions were linked to the contaminated products. Manure from domestic animals living in close proximity to irrigation water sources and near production fields can find its way into an irrigation water source and then onto crops. In the spinach outbreak that occurred in 2006, the same serotype of E.coli O157:H7 found in victims was also found in dairy cattle from farms that were in close proximity to the spinach operation and in feral pigs that were found in the spinach fields. It is important to site produce fields away from cattle operations and to not allow any animals, including poultry or pets, to roam in crop areas. Building fences or other barriers to keep feral animals out of production areas is vital to the safety of the products grown there. Also, steps should be taken to minimize wild animal and bird traffic in ponds and through fields where possible. In the packinghouse, sanitary design and sanitation are important components in pest management. The packinghouse should be properly designed and constructed, protected from the environment, and maintained in a clean and sanitary manner (Bihn and Gravani 2006). Some packing facilities are open structures where insects, rodents (mice and rats), birds, reptiles, amphibians, and other pests can easily enter and contaminate equipment, food contact surfaces, and product. The most important pest

114

Section II. Preharvest Strategies

management strategy is prevention and that is achieved through sanitary building design and construction, an effective sanitation program and vigilant attention to detail (Bihn and Gravani 2006). Recall and Traceback Strategies A key foundation block of a Farm Food Safety Plan is the development and testing of an effective recall and traceback plan. Should there be a need to retrieve a product from the marketplace, a plan that traces the product at least one step forward and one step back is imperative. One of the best ways to approach this task is to develop a product identification system that will enable the grower of that crop to identify, code, or label each item, container, or lot from the field of origin to the distributor. Some growers who grow commodities that are field packed (such as iceberg lettuce), print bar-coded labels in the field as produce is being harvested and use this system to identify the boxes of the produce, date of harvest, field, row, harvest crew, time of day, etc. The codes that a company uses must be simple, understandable, and well known to employees in the operation, so that if a problem occurs, the product and its origins and routes of handling can be quickly identified and traced. A recall plan consists of a team of key employees with designated specific tasks, a list of events that would trigger a recall, the recall procedures, and how the recall procedures will be implemented. It should also contain an up-to-date list of important contacts (with land-line telephone numbers, cell phone numbers, emergency numbers, pager numbers, email addresses, etc.), including distributors, customers, transportation companies, warehouses, terminal markets, scientific advisors, produce trade associations, legal counsel, regulatory agency officials, the media, and other important stakeholders. Examples of correspondence to all stakeholders, including media releases, etc., should be part of a company’s recall manual. The recall plan should also address strategies for handling and disposition of the recalled products, as well as methods for verifying the effectiveness of the recall procedures (Rangurajan and others 1999). The best way to test a recall plan is to conduct mock recalls and identify any weaknesses that need to be corrected. It is imperative that recall plans be regularly updated as contact information of personnel and organizations change. Produce trade associations have education materials that provide detailed information and examples of documents needed to develop effective fresh produce recall plans. Crisis Management A crisis is a specific, unexpected, nonroutine event (or series of events) that creates uncertainty and often threatens an organization’s goals (Seeger and others 2003). The outcome of a crisis determines whether possible negative consequences will follow. Events such as foodborne outbreaks; product recalls; natural disasters including floods, tornadoes, and hurricanes; power outages; and injury or death of a critical farm staff member are some examples of situations where a crisis management plan would be crucial to an operation (Bihn and Gravani 2006). Every farm operation, no matter what its size, needs to have procedures in place to address and manage a crisis effectively. Crises are unique and unpredictable, involve multiple audiences and stakeholders, are often very expensive, have the potential to damage a company’s reputation and brand image, and usually involve the media (Andrew 2007). Having a basic

The Role of Good Agricultural Practices

115

knowledge of risk and crisis communication practices, and specific plans and procedures in place for managing a crisis is very useful should a crisis occur. No company or organization ever expects a crisis, but through proper planning, company management will be able to address many issues and concerns in a clear and organized manner. The risk communication team of the National Center for Food Protection and Defense, a Department of Homeland Security–funded center at the University of Minnesota, developed 10 best practices of risk communication, which provide excellent advice for those interested in communicating with all stakeholders during a crisis (NCFPD 2006). In a crisis that involves public health, messages that provide consumers with information on what is known about the issues involved in the crisis, what is not known, when additional information will be available, and what they can do to protect their families should be shared openly, honestly, and regularly, with compassion, concern, and empathy. See chapter 6 for more information. Farm Biosecurity (Food Protection and Defense) With the tragic events of September 11, 2001, our country has become aware of the need to develop strategies to protect against the intentional contamination of our food supply, and in this case, fresh produce. A biosecurity (food protection and defense) plan addresses the intentional contamination of produce with biological or chemical agents (Bihn and Gravani 2006). Although the risk of the intentional contamination of produce is very difficult to predict, there are some simple measures that can be implemented to reduce this risk and make farm operations and packinghouse facilities more secure and deter would-be perpetrators. Farm buildings where harvested commodities are stored and chemical storage areas should be secured by locking doors and windows and limiting the access to these facilities to only trusted personnel. Farm security can be tightened by limiting access and controlling the flow of people on and off the farm, whether they are employees or visitors (Bihn and Gravani 2006). Visitors, contractors, and vendors should report to the office where they sign in, provide information about their visit, and receive a badge and escort during their stay. Vigilance on the part of all employees is important in deterring intentional contamination. Everyone should be asked to report any unusual sightings or events; and if any strangers are spotted on the premises, their whereabouts should be reported to a supervisor immediately. For larger operations, food protection can include security fences, closed circuit television cameras, gates, and guards, but for the majority of small farm operations, this level of security and the expenses incurred may far outweigh the value (Bihn and Gravani 2006). Every farm and packinghouse operation should review the security of their operations and seriously consider changes that can be implemented to strengthen their defenses.

Advancing GAPs Implementation Although most growers and packers are now aware of the importance of food safety and the need to reduce the microbial hazards and risks associated with the growing, harvesting, and packing of fresh fruits and vegetables, implementation remains inconsistent. As advances in research provide greater knowledge about the ecology of pathogens, their transmission, and their attachment to produce, and effective

116

Section II. Preharvest Strategies

mitigation strategies are identified for use throughout the production and distribution system, commodity-specific GAPs will be developed to further reduce the microbial hazards and risks associated with fresh produce.

References Andrew, P. 2007. Reputation Management. Crisis Communication Planning, Response and Recovery. A presentation for the National Restaurant Association/Ecolab Conference on Preparing for a Crisis— Business Sustainability and Continuity. Boston, MA. November 7, 2007. Beru, N. and P.A. Salsbury. 2002. FDA’s produce safety activities. Food Safety Mag. February–March: 14–19. Bihn, E.A. and R.B. Gravani. 2006. Role of Good Agricultural Practices in Fruit and Vegetable Safety. Microbiology of Fresh Produce. Karl R. Matthews, Ed. ASM Press, Washington D.C. pp. 21–53. Brown, Judith, E. 2008. Nutrition Now. Fifth Edition. Thompson Wadsworth Publishing Co., Belmont, CA. 33 Units. FDA (Food and Drug Administration), U.S. Department of Agriculture and Centers for Disease Control and Prevention. October 26, 1998 posting date. Guidance for Industry: Guide to Minimize Microbial Food Safety Hazard for Fresh Fruits and Vegetables.(online) Food & Drug Administration, Washington, D.C. www.foodsafety.gov/∼dms/prodguid.html. Gravani, R.B. 2006. Produce Safety Assurance Pyramid. Role of Good Agricultural Practices in Fruit and Vegetable Safety. Microbiology of Fresh Produce. Karl R. Matthews, Ed. ASM Press, Washington, D.C. pp. 21–53. Guzewich, J. 2008. Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration. Personal communication. Institute of Food Technologists. 2001. Analysis and Evaluation of Preventive Control Measures for the Control and Reduction/Elimination of Microbial Hazards on Fresh and Fresh-Cut Produce. Chicago, IL. International Fresh-Cut Produce Association (now known as United Fresh Produce Association). 2001. Food Safety Guidelines for the Fresh-Cut Produce Industry, 4th Ed. Alexandria, VA. Morris, J.G. and M. Potter. 1997. Emergence of new pathogens as a function of changes in host susceptibility. Emerg. Infect. Dis. 3:433–441. NCFPD (National Center for Food Protection and Defense), Risk Communication Team. 2006. Best Practices for Effective Risk Communication. http://www.ncfpd.umn.edu/docs/NCFPDRiskCommBestPractices.pdf. Accessed 8/08. National Good Agricultural Practices Program, Cornell University. 2008. Education Materials for Farm Workers and Growers. www.gaps.cornell.edu. Accessed 7/08. National Organic Standards Board. 2002. Compost Task Force Recommendations, as amended by the NOP. http://ams.usda.gov/nosb/NOSBrecommendations/compost.pdf. Accessed 8/08. National Research Council, Board on Agriculture, Committee on the Role of Alternative Farming Methods in Modern Production Agriculture. 1989. Alternative Agriculture. National Academy Press, Washington, D.C. National Research Council. 2003. Scientific Criteria to Ensure Safe Food. National Academy Press, Washington, D.C. 401 pp. Partnership for Food Safety Education. 2004. Safe Handling of Fresh Produce. http://www.FightBac.org/ content/view/203. Accessed 7/08. Pool, W. 2007. Wegmans Food Markets. Personal communication. Rangurajan, A., E.A. Bihn, R.B. Gravani, D.L. Scott and M.P. Pritts. 1999. Food Safety Begins on the Farm. A Growers Guide. National GAPs Program, Department of Food Science, Cornell University, Ithaca, NY. 28 pp. Seeger, M.W., T.L. Sellnow and R.R. Ulmer. 2003. Communication and Organizational Crisis. Praegar Publishers, Westport, CT. 297 pp. Smith DeWaal, C.S. and F. Bhuiya. 2008. Outbreaks by the Numbers: Fruits and Vegetables, 1990–2005. Center for Science in the Public Interest, Washington, D.C. (online) http://cspinet.org/foodsafety/ IAFPposter.pdf. Accessed 10/08. Smith, J.L. 1997. Long-term consequences of foodborne toxoplasmosis: Effects on the unborn, the immunocompromised, the elderly and the immunocompetent. J. Food Prot. 60:673–676.

The Role of Good Agricultural Practices

117

Tauxe, Robert. 2008. Centers for Disease Control and Prevention. Personal communication. USDA (U.S. Department of Agriculture). a. 2005. My Pyramid. www.mypyramid.gov. Accessed 8/08. USDA (U.S. Department of Agriculture). b. 2008 U.S. per capita food availability: Fruits & Vegetables. www.ERS.gov/foodconsumption/FoodAvailWQuereable.aspx. Accessed 9/08. Vierk, K., E. Elliot, T. Gondwe, J. Guzewich, T. Hill, K. Klontz, P. McCarthy, S. McGarry, M. Ross, J. Sanders, D. Street and B. Timbo. 2008. Outbreaks Associated with FDA/CFSAN-regulated Foods: 1996–2007. Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, College Park, MD. PDF file, 27 Slides. Vierk, K. 2009. Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration. Personal Communication. WHO (World Health Organization). 2005. Global Strategy on Diet, Physical Activity and Health. Promoting Fruit and Vegetable Consumption around the World. Information Sheet. www.who.int/ dietphysicalactivity/fruit/en/index2. 2 pp. Accessed 11/07.

6 Effectively Managing through a Crisis Will Daniels and Michael P. Doyle

Introduction Earthbound Farm is the world’s largest grower and marketer of organic produce, and it is recognized for its high-quality products. The company is also known as the specialty salad leader and processes conventional specialty salad greens for other brands under its company’s legal name, Natural Selection Foods (NSF). On September 14, 2006, the company received a call from the California Department of Health Services (CDHS) informing it that there was a multistate E. coli O157:H7 outbreak associated with packaged spinach packed by NSF. By the next day, NSF voluntarily recalled all spinach brands it packaged. The details remain vivid, and the employees’ perspectives were forever changed. A robust Incident Management Plan was already in place, as well as an Incident Management Team with representatives from all areas of the company. Even if you cannot imagine a reason your company might need to be so prepared, that unexpected incident with the potential to disrupt your business is out there. Developing an Incident Management Plan is simply sound business practice, particularly today when international communication, facilitated by the web, is instantaneous. An incident doesn’t have to be as huge as a nationwide outbreak and product recall; it can be anything that will have a significant impact on the business: Natural disaster, the sudden loss of a key executive, a production accident, or a work stoppage are only a few examples. Companies that have developed and practiced Incident Management Plans are better positioned to work with emergency responders, government agencies, and the media to quickly address the needs of those affected, communicate the company’s message, provide accurate and timely information, and ultimately reduce the adverse impact of the incident on the company’s day-to-day activities. Good plans help reduce the pressures on those involved, demonstrate the company’s commitment and professionalism, control the flow and accuracy of information, and manage resources most effectively. The degree to which key leaders and staff in the organization embrace the plan and understand their roles in its implementation will have a significant impact on the company’s ability to manage and withstand a crisis effectively. Any successful business strategy must come from the top down and have the support of management. Having this commitment, the Incident Management Plan must then be developed, implemented, practiced, and perfected as a team. In today’s age of instantaneous communication and citizen journalists, it is next to impossible to avoid the seemingly microscopic scrutiny as an incident unfolds, even though most would prefer to work through a problem inconspicuously. 119

120

Section II. Preharvest Strategies

The Incident Management Plan should provide sound and reliable guidance and roadmaps for most of the decisions that will have to be made very quickly. And even with all the tools in place, it is possible to skip a step or deviate from the plan under pressure, so regular review and revision is essential, even during the moment of crisis. Incident Management Plans come in many variations, but in developing such plans you should begin by taking a comprehensive look at your business and consider the kinds of incidents that could potentially disrupt its activities. The Incident Management Plan described in this chapter focuses on the critical communications during an incident: internal and external communication about the issue, what the public and employees should do, what the company should do, and how the issue has been resolved. Other aspects important to the Incident Management Plan include delineating the processes for the product recall as well as changing a process or the management. This includes standard operating procedures or programs such as a Recall Program, a Change Management Procedure, or a Succession Plan. These documents should be in place prior to development of the Incident Management Plan. Developing these procedures while working your way through a crisis will likely lead to mismanagement of the event. Managing a crisis can take minutes to years, depending upon the nature and magnitude of the issue. An incident can be as routine as the loss of a key employee or as rare as a product recall. When considering whether to activate your Incident Management Plan, you should evaluate your situation and determine the degree to which your company could be affected by an incident. No matter how well prepared you are, a significant crisis will affect your day-to-day business. Crises come in two main categories: crises with people and crises with products. People crises usually impact the company’s productivity and generally relate to employee matters and include, but are not limited to, personal injury, threats, disgruntled employees, strike, work stoppage, facility issues, and changes in management. In most cases, communication will be internal with employees and external with customers concerning how their orders will be affected. Product crises usually impact the company’s brand and relate to the products’ quality or safety, which can include product tampering, contamination with an adulterant, and packaging and product defects. In most cases, communication will be with employees, customers, and consumers (the public). Many of these incidents will involve some level of communication with local, state, or federal authorities or regulators. This chapter provides the tools and a process for developing an Incident Management Plan. The success of the plan depends completely on a company’s willingness to embrace and support the plan from the company’s top leadership on down. If key decision makers and role players do not understand the relevance and value of investing in and practicing such a plan, it can undermine the entire effort, thereby leaving you as unprepared as if you had no plan at all.

Defining the Incident Management Plan Some critical prerequisite programs must be in place before developing an Incident Management Plan to help guide how to address an incident that causes the crisis, such

Effectively Managing through a Crisis

121

as a recall. If your company does not know how to carry out the logistics of a product recall (or other action that must be taken), the response will be extemporaneous and uncertain, rather than a preplanned, deliberate response that results in a well-controlled situation. Examples of prerequisite programs include Change Management Plans/ Procedures, Succession Plans/Procedures, and Evacuation Plans/Procedures. Developing the Incident Management Plan document is the first step in the process. This can be done by one or more people who are key members of the Incident Management Team. It should include an outside consultant if an experienced crisis manager is not available as a member of the current staff. Format and content of the plan can vary from company to company, but some fundamental elements should be in every plan: • • • • • • • • •

Purpose Definition of a Crisis Guiding Principles Definition of Threat or Action Levels Team Members’ Roles and Responsibilities Media Communication Policy Incident Response Procedure Communication Procedures Appendixes to include a contact list of your team members (including home contact information) and documents critical to managing the crisis

The Purpose and Definition sections are designed to communicate the intent of the plan and should provide a focus to the team, so that no one is guessing about whether an incident has reached a crisis level. The Guiding Principles assist the Incident Management Team in developing the plan as well as providing guidance during a crisis. Examples of statements include respond quickly, comply with all laws and regulations, and measure your response and act appropriately. Definition of Threat or Action Levels should provide guidance for Incident Management Team members to determine the seriousness of the incident and identify the appropriate response. Team members’ roles should be clearly articulated and well suited to the expertise of team members. For example, sales representatives should be the customer contacts and liaisons, and the quality assurance representatives should perform the traceback in the event of a recall. The Media Communication Policy is one of the most important statements in the document. Controlling the message can be a significant challenge, so it is important the company have one voice and one clear message communicated to all constituencies. The Incident Response Procedure should include guidance on topics such as determining when to convene the team, collecting information and measuring media attention, taking action steps, critiquing responses, and identifying additional action steps that may be needed. Communication Procedures should ensure that all constituencies receive the attention and information they need. Although a crisis brings the need for external

122

Section II. Preharvest Strategies

communication with the media, customers, consumers, and the government to the forefront, it is important not to forget the importance of having a strong communication plan for employees.

The Incident Management Team The next step is to assemble the Incident Management Team. This team should be a cross-functional group that includes key decision makers from all areas of the business, including operations, sales, human resources, communications, and the executive committee. Additional team members could include, but are not limited to, legal counsel, external communications counsel (e.g., an experienced public relations/crisis management expert), a situational expert (depending on the on-staff expertise, a food safety expert, microbiologist, or security expert may be needed) and a historian. The historian can play a crucial role in any crisis management scenario, being the eyes and ears of the process as it is happening and unfolding. The historian documents meetings, discussions, inspections, Q&A between inspectors and the company, etc. This chronological account of events will be helpful as a reference for what was said or submitted and will be crucial in the event that the crisis leads to regulatory enforcement and/or litigation. Appointees to the Incident Management Team should be decision makers that possess sufficient knowledge and expertise to understand the impact of their decisions. In addition, each team member should have an alternate who can assume the responsibilities in the event that the primary member is unavailable. The alternate should have the same level of knowledge and familiarity with the plan as the primary member. Members of the team must understand the importance of their input in the development of the plan and their role and responsibilities as team members. Team members must understand that, depending on the business, they may be required to be accessible any hour of the day, any day of the week. The Incident Management Team members are usually selected by the CEO, president, or owner of the company. Regardless of how the team is chosen, each member must understand the responsibilities associated with a position on this team. These responsibilities should be emphasized at the time the request to join the team is extended and they should be reemphasized at every meeting. Any team members who are reluctant, not motivated, or who may need some additional encouragement to make relevant contributions should be identified and replaced.

Developing the Incident Management Plan At the first meeting of the Incident Management Team, roles should be clarified and details of the plan should be reviewed so that everyone understands them fully. A brainstorming session to identify incidents that could rise to the level of a crisis and affect the business is a useful team development exercise. Several techniques to brainstorm the situations can be used as long as everyone participates, and they should include as many potential situations as possible. This list of scenarios can be used as incident drills to practice team and company responses, providing the team opportunities to discuss them and better understand its role in each situation. This conversation should be thorough, including as much detail

Effectively Managing through a Crisis

123

as possible. For example, the team may have identified an earthquake as a threat that could affect the business. In this case, start with what you would do to prepare your employees and facility, such as evacuation plans and earthquake retrofitting. Next address the event itself. Imagine the worst case scenario, such as the facility sustaining enough damage to shut it down for repair and some employees being killed in the building collapse. Using this scenario, go to the moment the quake ends: • • • •

Does the evacuation plan cover facility shutdown procedures? Does it address injuries or deaths (calling emergency responders)? Who will check employee rosters to account for everyone? What will be done with product in the facility (finished goods and raw materials)? • Who will let customers know that orders will be delayed? Next, work on the communication component of the crisis. What should be said to the workforce, the customers, and the public? Assume a scenario in which the facility can recover in time to keep all employees, but some distressed employees will need immediate attention, and some employees will need hospital care. Everyone should discuss his/her role(s) in handling this type of situation effectively. Assuming a worstcase scenario in which some employees were killed, the remaining workforce will need to be notified about them as well as how the incident affects their own job security. Part of this scenario’s plan should also address production. If the plant is down for repair, can the product be made elsewhere? Will customers not receive their entire orders? If so, for how long? As part of this exercise, messaging should be developed to address all the areas outlined previously. Many ideas brought up in this brainstorming session may be unique to the facility and product. Don’t discount any idea. If it can be thought of, it may happen. Include all team members in the discussion as you work through each potential crisis threat. These work sessions should be taken seriously so that the organization is properly prepared. If the organization is not prepared to handle a crisis, it will have little chance of controlling the message during an actual crisis. This can be a complex and even emotional process. Hence, it is beneficial to work with an experienced crisis manager to facilitate the practice runs. The facilitator will be less involved in the details of the company’s operation and more focused on the principles of good incident management. An experienced crisis manager will be able to solicit details that may have been missed, suggest successful strategies, and identify potential flaws in the plan. Depending on the business, this crisis brainstorming could take hours. It may be done in several meetings, rather than one, but it should not be taken lightly and no scenario should be left out. This is the foundation of the Incident Management Plan, so it should be comprehensive. Once the brainstorming is complete, these scenarios and documents specific to each scenario should be added to the plan. However, the Incident Management Plan should be a living document, revised at least annually, or as often as the business, products, facilities, or personnel change.

124

Section II. Preharvest Strategies

The Incident Management Team should convene at least annually to review the plan and make recommendations for modifications, develop new crisis scenarios, and work through some “mock” scenarios to practice the plan. Practice is extremely important. From year to year, staff will change and new members may not have had the benefit of being part of the original development of the plan. Hence, it may be important to work directly with new employees who have a key role on the team. New team members must understand the plan, their role, and how they will work with others through the crisis, understanding that a crisis may occur tomorrow. Another way to perform mock incident management drills is to link the practice with mock prerequisite drills, such as a mock recall. In most cases, both plans will be used simultaneously and it is important to understand the interactions among all role players: team members, personnel resources, communication, products, etc. When running a mock drill, utilize the documentation you have in place so that the drill simulates a real-world situation: draft press releases and recall announcements, use the team directory (check numbers to be sure they are accurate), and review corrective action tracking documents. Although it is often said that these events should not disrupt your day-to-day activity, the reality is that, in most cases, a crisis will at a minimum affect day-to-day activities and is likely to be disruptive. Not only will it impact production and customer service, but it will also impact personnel. In some cases, it will not only affect their ability to get their daily tasks done but it may affect them personally. They may suffer from stress, which should be anticipated and managed as another part of the plan. If the interruption of business can be anticipated, the company will be even more prepared in the face of an actual crisis.

Implementing the Plan during a Crisis The moment there is a crisis is not the time to begin planning how to respond. The plan should be already developed and ready for implementation. Review the plan as soon as the crisis occurs and as often as necessary to ensure there are no deviations. In some cases, however, it may be necessary to modify the plan due to an unanticipated event or missed scenario during the brainstorming. It is important to apply as much of the preplanned information to the new scenario as possible to aid in controlling the message. For this discussion, a product recall due to a foodborne illness outbreak will be used as the crisis example. The first step in the process is for the team leader to identify the need to convene the team. Based on the guidelines in the plan, the leader may determine that there is no need to convene the team (the crisis threat is low and the problem may have already been resolved). If a need is identified, the team should meet immediately. Once the team is convened, it is extremely important to brief the team on the issue so it can determine the crisis threat level. The information provided should be based on facts and not include speculation or rumors. It is important to know the source of the information. There should also be a review of media activity surrounding the issue. If the media attention is high, it should be considered and factored into the response. If it is low, it may indicate less of a crisis threat than anticipated. Once the threat level is determined, the trained team should immediately move into action mode. A review of

Effectively Managing through a Crisis

125

responsibilities for each team member should be discussed. This is especially important if some team members are missing and responsibilities must be reassigned. The next step is the notification of key stakeholders. Depending on the issue, these key stakeholders may include customers, employees, suppliers, consumers, government, and the media and should have been clearly defined in the plan during its development. Use the Guiding Principles to help develop communication messages with each stakeholder and to keep the messages consistent. Why do the stakeholders need notification? It is important to notify customers because they will need to know whether orders are going to be interrupted, for how long, and what should be done with the product in their possession. Employees will need to know how their work schedule will be affected and how the company is responding to the recall. Employees will likely be the last group that management feels needs attention but do not forget them and be sure they are informed as often as necessary. They are a great source of information outside of work and will be approached by friends, competitors, or the media who want more information. In addition to the information about the incident, it is also important to give employees tools for how to handle possible confrontations. The suppliers will need to know how deliveries will be affected and for how long. The government should be involved and should be alerted to the situation immediately. The responsible government regulatory agency will expect a recall announcement soon after you decide to initiate the recall, and the announcement must be written in a prescribed format. Having a statement prepared in advance is advisable. This is usually the peak of the stress level for all involved; and it can be difficult to think clearly, provide all the necessary information, and control the message when you have only minutes to respond. Typically, crises don’t happen on a Monday morning when all resources are available, but often occur on a Friday afternoon when the work week is ending. A government agency involved is no different and may unintentionally cause additional stress. One cannot overemphasize the importance of being well-prepared. The media response is where the plan often shows its greatest value. This stakeholder is the most unpredictable group and is usually looking for a good “story.” It is extremely important to make the message clear, concise and to the point. Even with a well-scripted message, the information may be twisted to make the story more sensational. When this happens, do not be discouraged. Do not be goaded into a reactive response; stick with the message and keep delivering it, and eventually it will be accurate in the media. Another tactic reporters may pursue once you are in the limelight is to find past violations or issues with the company. For example, if a reporter finds that the company failed to file for a building permit for a canopy in its receiving yard, this may be used to illustrate that the company has a history of negligence. If the story is big enough, media may camp outside the facility. This will likely unnerve employees, and the Incident Management Team member in charge of communications should be prepared to reassure employees that this is normal, that the reporters can’t hurt them, and that they should just be polite but not talk to anyone. When responding to stakeholders and delivering the message, it is important that both communications and actions are sincere. Do not force any response or activity— your actions will not be taken seriously if people do not see your sincerity and interest

126

Section II. Preharvest Strategies

in their well-being. Don’t promise anything that you are not willing or able to accomplish. For example, in the case of a recall due to an outbreak, don’t send the message that people who became ill from eating the contaminated product will be reimbursed out-of-pocket medical expenses if you don’t intend to follow through. A lack of commitment will quickly be reported by the media, by your peers in the trade, and even by your employees; and your reputation, credibility, and further actions will be undermined.

The Seven R’s of a Response There are at least “The Five R’s” to rely on as a guide during a response in a crisis: Regret, Responsibility, Restitution, Resolution, and Reform. A sixth and seventh “R” should be added to this list: Review and Repeat. Regret It is extremely important to communicate regret to affected stakeholders, such as expressing your sincere concern for the victims. Be sure that this message of regret is conveyed in all communications and actions moving forward. Responsibility Communicate immediately to customers the details of the product being recalled. Work with the media to deliver to the public the details of the recall, including how to identify product and what should be done with it. Be sure to provide contact information if the media have questions. It may be beneficial to use an outside source for handling these calls. Depending on the size of the recall and media attention, thousands of calls a day could be received, which would overwhelm most in-house consumer hot lines. It may also be helpful to have an easily accessible website to provide all stakeholders up-to-the-minute information regarding the recall, the company, and the message. This website can be prepared in advance of any crisis and set up as a “dark site.” This means the page is fully developed and contains as much relevant information as possible prior to a crisis but not turned on or accessible to the public. This dark site, when activated, can then be linked to your home page to avoid confusion on how to access it. Easy access to critical information is essential for all of the stakeholders, and it will be less stressful for the company managing the crisis. Showing responsibility is also very important in the management of the response. Cooperate fully with investigators in their effort to prevent additional illnesses and protect public health. Public safety should be at the forefront and the top priority of food processors on a daily basis, so it should not be any different during an investigation of an outbreak of foodborne illness. Assist in identifying the cause of the outbreak, and if it was a failure in the system, assume responsibility and correct the problem. Investigations are rigorous and may require repeating information and often defending the company’s food safety programs. Remembering that the inspectors’ questions and audits are part of the regulatory agency’s responsibilities makes it easier to work through the investigation. Taking a defensive stance and not sharing information or appearing uncooperative will likely result in more harsh treatment by inspectors

Effectively Managing through a Crisis

127

because the investigators will assume something is being hidden. If the crisis has progressed to this point, being defensive is unproductive. Accept the situation and cooperate fully with the investigators rather than appearing to be acting against them. Restitution It is likely that your first opportunity for restitution will be providing your customers with assurances that they will be reimbursed for all product recalled, including any additional charges to return or destroy product. Depending on the circumstances, it is also important to offer to cover appropriate out-of-pocket medical expenses for those directly affected by this situation. Consider how these actions will be perceived by lawyers who will be handling these cases later. Resolution Not only is it important to convey in the message what is being done to address the immediate situation, but it is also essential to show that efforts are being made to resolve the problem and proactive steps are being taken to prevent it from happening again. Based on information provided during internal and external investigations, including advice from internal and external experts and the government, develop a process improvement plan or corrective action plan to address those issues. Do not commit to anything that will not be accomplished or corrected in the long term. Be certain the response addresses the concerns of key stakeholders and that the corrective action plan is effectively communicated to them. This is imperative in regaining the stakeholders’ confidence in the company’s product. If the company is still viable, this strategy will prove invaluable in regaining the business that was lost as the result of the crisis. Reform Reform must be addressed if the investigation reveals that even the best practices are not going to prevent such an event from recurring. This requires airing to all stakeholders your learning experiences and the corrective actions needed to prevent this from happening again. This reinforces the company’s commitment to doing the right thing and taking responsibility for the problem. Review Continuously reviewing the communications, corrective actions and stakeholder response in an effort to gauge the success of your Incident Management Plan and response is vitally important. If, for example, there are several of the same questions about product codes, it is necessary to review and revise the communications to make this point clearer. If the corrective actions are not adequate, revise them and make them effective. If stakeholders are responding negatively to the company’s communications and actions, find out why and adjust. This type of analysis is widely used in process improvement and should be included here. It will be extremely valuable as the crisis evolves. Not adjusting during the crisis to the company’s response to the stakeholders’ needs and expectations will be viewed as not taking responsibility for the problem and will undermine the efforts that have been made in addressing the crisis.

128

Section II. Preharvest Strategies

Repeat After reviewing the company’s communications and actions responding to the crisis and making some changes to the plan, it is necessary to go back as far to the beginning of the Incident Management Plan and repeat the steps. Changes to the Incident Management Plan will likely be made several times during the crisis response. The ability to react to information in real time will strengthen the company’s message and resolve. Continue this cycle of change and review until stakeholders are satisfied with the response, business is back to normal, and the crisis has been determined to be over by the Incident Management Team.

Conclusions Depending on the crisis, execution of the Incident Management Plan could require anywhere from minutes to years. The information provided in this chapter is not intended to be comprehensive, but to provide some experience-based guidance and tools to assist in planning for unforeseen crises. Employing external counsel in the development and execution of an Incident Management Plan is invaluable; counsel objectivity will be crucial, because it is often very difficult for company employees to remain objective during a crisis. A company and its employees are not likely to be fully prepared for a crisis and its intricacies unless they have some experience. Even then, it is unlikely enough experience will be gained as that of a professional crisis manager. Hence, have a professional crisis manager preselected to assist, engage this person at the time of the crisis, and then determine whether the crisis manager is needed for the issue at hand. Most importantly, be sincere and focus on doing the right thing. The consequences for not developing an effective Incident Management Plan and executing it properly could well be loss of trust and ultimately loss of the entire business.

7 The Role of Water and Water Testing in Produce Safety Charles P. Gerba

Introduction Water plays a usually overlooked role in terms of importance in produce safety. It is not only essential in the growth of crops, but also in their processing. Water is used in the cooling of produce, removal of debris, decontamination, washing of equipment, and personal hygiene. The microbial quality of this water is important in ensuring that contamination of the produce does not occur during these events. This is especially true of this commodity because the produce is often eaten with no or little processing. The importance of high-quality source water and its separation from human wastes was recognized by the Romans more than 2,000 years ago. The development of the germ theory of disease a little more than 100 years ago provided the basis for modern water quality treatment and personnel hygiene. These two practices together have played the biggest role in disease reduction in human history, more so even than the development of antibiotics and vaccinations. The pathogens most commonly transmitted by produce are enteric pathogens transmitted by the fecal-oral route and are excreted in large numbers in both the feces of man and animals. For example, human rotavirus may be excreted in numbers as great as 1011 per gram (Gerba 2000) and Salmonella 1010 (Feachem and others 1983). When one considers that only 1–10 viruses are needed to cause an infection, a small amount of feces in the water can have a significant impact on disease transmission to susceptible populations (Gerba and others 1996). The largest use of freshwater in the world is in agriculture, with more than 70% being used for irrigation. Approximately 240 million ha, 17% of the world’s cropland, are irrigated, producing one-third of the world’s food supply (Shannan 1998). Nearly 70% of this area is in developing countries. Irrigation of food crops with untreated domestic sewage has long been associated with the transmission of infectious diseases. As a result, the use of wastewater for irrigation is forbidden or the wastewater must be highly treated and rigorously monitored in developed counties. Irrigation with sewage or sewage-contaminated surface waters in developing countries is fairly common and usually not regulated. Although guidelines for wastewater reuse have been developed by the World Health Organization (WHO 2006), their application in developing countries is difficult, due to inadequate institutional capability and general lack of financial resources. Ideally water used for application of herbicides and pesticides and for washing would receive complete treatment similar to that required for drinking water because direct contact of the water with the produce occurs. In the United States surface water 129

130

Section II. Preharvest Strategies

used for drinking is required to be filtered and disinfected. Filtration ensures the removal of protozoan parasites (Giardia and Cryptosporidium), which are very resistant to chlorine. Unless special conditions are met (e.g., absence of coliform bacteria in the water and no potential sources nearby), all groundwater used as drinking water must be disinfected. The minimum treatment in both cases must be capable of reducing the protozoa by at least 99.9% and the enteric viruses by 99.99% (Regli and others 1991). Because enteric bacterial pathogens are more sensitive to disinfectants it is felt that these treatment requirements would be sufficient to eliminate waterborne bacteria below levels of concern. In practice, water used for pesticide and herbicide application is collected from irrigation channels, ponds, or other sources, which are not treated. Because application often takes place by direct spraying, the contamination of produce occurs. Water for washing may also come from untreated sources, but chlorine is often added to the water to reduce possible cross-contamination and numbers of organisms on the surface of the produce. Because pathogens attached to the surface of produce are significantly more resistant to inactivation, much greater concentrations of chlorine are used than to treat drinking water (200 mg/l vs. 1–3 mg/l).

Occurrence of Foodborne Pathogens in Water Groundwater Under normal conditions, coliform bacteria are absent from drinking-water supplies. In most countries, monitoring of treated water supplies for coliform bacteria ensures the absence of E. coli, which is a member of the coliform group. However, small systems and private wells are monitored less frequently or not at all, depending on local regulations. In general, enteric pathogens are far less common in groundwater because of the natural filtering mechanism of soil. However, every groundwater source is potentially susceptible to contamination. The construction of wells (or location of a spring), nature of the substrata, depth to groundwater, and rainfall can affect the microbiological quality of the well water. Pathogens enter groundwater from latrines, septic tanks leach fields, land application of wastewater for irrigation, oxidation ponds, leaking sewer lines, and unlined landfills. Unprotected wells may allow for surface water to run into the well during storm events. In well-structured soils, protozoa and bacteria are easily filtered out during transport through the soil. However, in fractured limestone and clay soils and gravel/sandy soils, long-distance transport is possible. Viruses are more likely to contaminate groundwater and travel long distances (hundreds of meters) because of their small size (Keswick and Gerba 1980). A survey of 144 private water supplies in the Netherlands found that 11% were positive for fecal indicators and 2.7% for E. coli O157:H7. All were located in agricultural areas with grazing animals (Schets and others 2005). In India 79% of the rural water (both surface and groundwater) drinking water supplies contained pathogenic serotypes of E. coli including E. coli O157:H7 (Ramteke and Tewari 2007). Although waterborne outbreaks of E. coli O157:H7 are primarily associated with surface waters, large outbreaks have been associated with contaminated groundwater (Hunter 2003). Rainfall and irrigation provide a mechanism for leaching into groundwater. E. coli O157:H7 has been shown to travel the subsurface for 2 months after initial application

The Role of Water and Water Testing

131

(Gagliardi and Karns 2000). In a survey of 448 drinking water supply wells of 35 states, human enteric viruses were found in 31% before disinfection (Abbaszadegan and others 2003). Another study, also in the United States, reported the detection of Cryptosporidium, Giardia, or both in 11% of 166 sites using groundwater (MoultonHancock and others 2000). Many of the positive sites were from infiltration galleries located under or near streams to supply drinking water. Finally, the majority of drinking water disease outbreaks documented every year in the United States are caused by fecally contaminated wells (Reynolds others 2008). Thus, it should never be assumed that groundwater is free of pathogens. Surface Water Surface water is always more susceptible to contamination than groundwater because of the direct discharge of sewage and impact of runoff from rainfall events. Any freshwater surface source is likely to contain enteric pathogens at one time or another. The greatest levels of pathogens occur in water sources receiving discharges of untreated sewage and in watersheds with intensive levels of animal production. In the United States, sewage discharges are usually disinfected and storm water runoff from agricultural land and septic tanks (referred to as nonpoint sources) are the largest contributors of pathogens to surface water. However, in Europe and many other countries of the world sewage discharges are not disinfected. For these reasons finding enteric pathogens in surface waters in these countries is not uncommon (Bosch and others 2006). The greatest loading of pathogens occurs in surface waters after rainfall events, although concentrations may be elevated during low flow when the proportion of wastewater to natural runoff is highest. During runoff events accumulated feces are washed into nearby streams or collection systems forcing release of partially or untreated sewage from wastewater treatment plants. Studies of waterborne disease outbreaks associated with drinking water in the United States and Canada have been correlated with above average rainfall events with drinking water–associated disease outbreaks (Curriero and others 2001; Thomas and others 2006). Recent data on the occurrence of enteric pathogens in the United States and Europe are fairly limited, probably because of the difficulty in isolating pathogens directly from water. At this time, the prevalence of E. coli O157:H7 in surface waters in the United States is unknown, but sporadic illnesses and outbreak data suggests that surface waters play a role in E. coli O157:H7 disease transmission via recreational water exposures and drinking water sources. Waterborne disease outbreaks caused by E. coli O157:H7 are primarily associated with recreational lake waters (Bruce and others 2003). This is due to the ingestion of untreated water during swimming. Domestic and wild animals are believed to be the source of contamination in most recreation waters. Kurokawa and others (1999) detected 102 to 105 per ml E. coli 0157:H7 in river water heavily polluted by industrial and agricultural wastes using polymerase chain reaction (PCR). However, this method may detect both viable and dead bacteria. In surface water supplies in southern Alberta, E. coli O157:H7 was isolated in 1.7% of 1,608 samples over a 2-year period (Gannon and others 2004). Most isolations of the organism occurred during the summer.

132

Section II. Preharvest Strategies

Vereen and others (2007) studied the occurrence of camplyobacters in six streams in Georgia in watersheds representing different agricultural uses. The isolation of campylobacters ranged from 17 to 100% depending upon the watershed. Mean concentrations ranged from 2 to 158 per 100 ml (highest value 595/100 ml). The mean campylobacter numbers and overall prevalence were greatest downstream from a wastewater treatment plant that handled both human and poultry slaughterhouse waste. A multivariate model showed that the levels were significantly influenced by increasing precipitation, which peaked during the summer months. In the United Kingdom campylobacters were detected in 50% of the lake water samples and 20% of brook water samples (Sails and others 2002). Because of their greater resistance to chlorine disinfection, enteric virus and protozoa can be detected in any surface waters receiving sewage discharges. Chlorine disinfection will reduce the level of enteric viruses, but will have little effect on protozoa (Gennaccaro and others 2003). Because Giardia and Cryptosporidium also have animal reservoirs they can be expected to be present at any time in surface waters. In a nationwide study in the United States Cryptosporidium was detected in 55% of the sites with an average concentration of 43 oocysts per 1,00 l, and Giardia cysts were detected in 16% of the samples with an average concentration of 3 per 1,00 l (Rose and others 1991). In another study in the United States and Canada either Giardia and Cryptosporidium or both were detected in 97% of the surface waters at drinking water treatment plant intakes sampled over a period of several months (LeChevallier and others 1991). Cryptosporidium in surface waters usually originates from cattle and is usually found in greater concentrations in surface water than Giardia. However, Giardia is a common infection in man, and sewage discharges usually have greater concentrations than Cryptosporidium (Smith and Grimason 2003). Thus, surface waters more impacted by sewage discharges or combined sewer overflows may have more Giardia than Cryptosporidium. Humans are believed to be the only significant source of enteric viruses that cause illness in man, although hepatitis E virus is excreted in the feces of pigs and perhaps other animals (Mushahwar 2008). Although the occurrence of enteroviruses has been the most studied, adenoviruses may be the most common in sewage discharges and surface waters (Mena and Gerba 2008). Studies conducted in Europe have reported enteric virus concentrations in surface waters of between 1 and 56 per liter, as determined by conventional cell culture methods (Bosch and others 2006). Jiang and Chu (2004) reported that half of the river water they sampled in southern California heavily impacted by treated sewage discharges contained adenoviruses. In half of the samples collected at surface water inlets to drinking water treatment plants Chapron and others (2000) detected enteroviruses and adenoviruses in cell culture using the polymerase chain reaction. Enteric pathogens are more likely to be detected in sediments than surface waters. In the case of bacteria and viruses this is probably because of their attachment to suspended particles and their settling. The protozoan parasites are large enough that they will eventually settle out on their own. Once in the sediment they appear to be capable of longer-term survival or growth (Hendricks 1971a; Smith and others 1978). Hendricks (1971b) isolated salmonellae from 0.6% (1 of 195) of water samples and from 4.6% (9 of 195) of bottom sediments in a length of river below a sewage outfall. Van Donsel and Geldreich (1971) collected simultaneous sediment and water samples

The Role of Water and Water Testing

133

from lakes and rivers. They isolated salmonellae from only 8% of the water samples and 46% of the sediment samples. Greater numbers of fecal coliforms, Camplyobacter, and enteric viruses have also been detected in sediments than in the overlaying water (Falabi and others 2002; Obiri-Danso and Jones 1999; Goyal and others 1977). The accumulation of enteric pathogens can result in the rapid degradation of the quality of the overlaying water by both man-made and storm events because of resuspension of the sediment (Donovan and others 2008; Grimes 1980).

Survival in Water Survival of waterborne pathogens in water is dependent on a number of factors: • Temperature—longer survival at lower temperatures. Viruses survive longest at freezing temperatures. • Sunlight—longer survival in the dark. Ultraviolet light in sunlight plays the major role in surface waters. • Particle matter—protects against effect of sunlight and antimicrobial factors in the water. • Soluble organic matter—longer survival in the presence of sewage. Temperature is usually the dominate factor in survival and can be used to approximate survival of an enteric pathogen in water. In surface waters the incidence of solar radiation can also be used to predict the survival of indicator bacteria (Kadlec and Knight 1996). Generally, enteric viruses survive the longest followed by protozoan parasites and then bacteria. Table 7.1 lists reported survival time of E. coli O157:H7

Table 7.1. Survival of E. coli O157:H7 in water Study Czajkowska and others 2005

Artz and Killham 2002 Avery and others 2004 Ritchie and others 2003 McGee and others 2002 Geldreich and others (1992) Wang and Doyle (1998)

°C 6 24 6 24 10

T99 4–11 2–8 10–20 5–8 2

T99.99 8–22 5–10 25–39 10–30 15

10 10 15 15 15 5

1–6 1–29 20 35 7.5 14–>35

ND ND >90 58 14.5 ND

8 23 23

>13 9 2–5

ND ND ND

T99 = time in days for a 99% decrease in titer. T99.99 = time in days for a 99.99% decrease in titer.

Comments Surface waters from lakes and rivers Sediments from the same lakes and rivers Groundwater used for drinking. Conclusion: copper in water inactivated E. coli. Raw sewage—10 different samples Treated sewage Soil Groundwater River water with feces (outdoors and lab) Filtered sterilized water samples from Cobool, MO, after outbreak of E. coli O157:H7 Filtered tap water, unfiltered lake water Filtered tap water Unfiltered lake water

134

Section II. Preharvest Strategies

in water. Below 8 °C enteric viruses are capable of surviving for months in aquatic environments (Kutz and Gerba 1988). Survival of E. coli O157:H7 in surface waters was found to be greater at lower temperatures (Czajkowska and others 2005) and was two to three times greater in river and lake sediments at the same temperatures. In water contaminated with manure, E. coli O157:H7 was observed to survive at outside ambient temperatures for 92 days (McGee and others 2002). Enteric bacteria and viruses will also survive for longer periods of time in aquatic sediments than in the overlaying water probably because of the increased presence of organics and the thermostabilization of viruses (Karim and others 2004; Liew and Gerba 1980; Hendricks 1971a). Karim and others (2004) found that the die-off of Giardia cysts was greater in sediments from a pond use to treated secondary sewage effluents.

Water Testing The routine examination of water for the presence of enteric pathogens is often a tedious, difficult, costly, and time-consuming task. Thus, indicator organisms have been used to assess the presence of fecal contamination and the effectiveness of sewage treatment processes. Developed at the turn of the last century for assessing fecal contamination, the indicator concept depends on the fact that certain nonpathogenic bacteria occur in the feces of all warm-blooded animals. These bacteria can easily be isolated and quantified by simple methods. Detection of these bacteria in water means that fecal contamination has occurred and suggests that enteric pathogens may also be present. For example, coliform bacteria, which normally occur in the intestines of all warm-blooded animals, are excreted in great numbers in feces. In polluted water, coliform bacteria are found in densities roughly proportional to the degree of fecal pollution. Because coliform bacteria are generally hardier than diseasecausing bacteria, their absence from water is an indication that the water is bacteriologically safe for human consumption. Indicators have traditionally been used to suggest the presence of enteric pathogens; however, today we recognize that there is rarely a direct correlation between bacterial indicators and human pathogens (Ashbolt and others 2001). As such, the use of indicators is better defined by their intended purpose (Table 7.2). Thus, process indicators

Table 7.2. Definitions and examples of indicator microorganisms Group Process indicator Fecal indicator Index and model organisms

Definition and Examples A group of organisms that demonstrate the efficacy of a process, such as total heterotrophic bacteria or total coliforms for chlorine disinfection A group of organisms that indicate the presence of fecal contamination, such as the fecal coliforms or Escherichia coli A group or species indicative of pathogen presence and behavior, respectively, such as E. coli as an index for Salmonella and male-specific coliphages as models for human enteric viruses

Modified from Ashbolt and others (2001).

The Role of Water and Water Testing

135

are used to assess the efficacy of a treatment process (e.g., drinking water treatment), while fecal indicators indicate the presence of fecal contamination. An index (or model) organism represents the presence and behavior of a pathogen in a given environment. These indicators have been used to judge the safety of drinking and recreational waters, as well as indicating the success of treatment processes. However, how useful these indicators are for assessment of irrigation is currently not known. The coliform group, which includes the genus Escherichia, Citrobacter, Enterobacter, and Klebsiella, is relatively easy to detect. Specifically, this group includes all aerobic and facultatively anaerobic, Gram-negative, non–spore-forming, rod-shaped bacteria that produce gas upon lactose fermentation in prescribed culture media within 48 hours at 35 °C. The coliform group has been used as the standard for assessing fecal contamination of drinking waters for most of the last century. However, they can originate from certain plants and are capable of growth in the environment under certain conditions. They appear to be very common in large numbers in irrigation waters in Arizona not directly impacted by fecal contamination (Kayed 2004). The die-off rate of coliform bacteria depends on the amount and type of organic matter in the water and its temperature. If the water contains significant concentrations of organic matter and is at an elevated temperature, the bacteria may increase in numbers. This phenomenon has been observed in eutrophic tropical waters, waters receiving pulp and paper mill effluents, wastewater, aquatic sediments, and organically enriched soil (i.e., sewage sludge–amended) after periods of heavy rainfall (Zaleski and others 2005; Carrillo and others 1985). Fecal coliforms, which include the genera Escherichia and Klebsiella, are differentiated in the laboratory by their ability to ferment lactose with the production of acid and gas at 44.5 °C within 24 hours. The frequent occurrence of coliform and fecal coliform bacteria in unpolluted tropical waters, and their ability to survive for considerable periods of time outside the intestine in these waters, have suggested that these organisms occur naturally in tropical waters (Solo-Gabriele and others 2000) and that new indicators for these waters need to be developed. E. coli is more commonly being used as an indicator because it can easily be distinguished from other members of the fecal coliform group (e.g., absence of urease and presence of β-glucuronidase) and more likely to indicate fecal pollution. Fecal coliforms also have some of the same limitations in use as the coliform bacteria, i.e., regrowth and less resistance to water treatment than viruses and protozoa. The enterococci have been suggested as useful indicators of risk of gastroenteritis for recreational bathers, and standards have been recommended (Cabbelli 1981). However, they may not be useful in tropical waters (Byappanahalli and Fujioka 2004; Fujioka and others 1999) and are also common in warm irrigation waters used in Arizona (Kayed 2004).

Standards and Criteria for Indicators The use of microbial standards also requires the development of standard methods and quality assurance or quality control plans for the laboratories that will do the monitoring. Knowledge of how to sample and how often to sample is also important. All this information is usually defined in the regulations when a standard is set. For

136

Section II. Preharvest Strategies

example, frequency of sampling may be determined by the size (number of customers) of the utility providing the water. Sampling must proceed in some random fashion so that the entire system is characterized. Because of the wide variability in numbers of indicators in water, some positive samples or tolerance levels or averages may be allowed. Usually, geometric averages are used in a standard setting because of the often skewed distribution of bacterial numbers. This prevents one or two high values from giving overestimates of high levels of contamination, which would appear to be the case with arithmetic averages.

Irrigation Waters Irrigation using surface water and groundwater mirrors similar water quality and pathogen concentrations as discussed in the previous two sections (see above). There are a number of sources that can contribute to fecal contamination of irrigation systems even if there are no sources of sewage discharge: • Recreational use of source waters. Bathers can contribute significant levels of pathogens to reservoirs or that water may be used to store or collect water for use in irrigation systems (Gerba 2000). • Return flows. During flood irrigation non-infiltrated water may be discharged back into the irrigation canal system or be used to irrigate other fields. If these fields have manure applied they may contribute pathogens to the irrigation water. • Wildlife. Migrating birds such as ducks or geese may rest on the canal waters. Deer, rabbits, and other animals may also have access to the channeled water. • Storm water. In some arid areas irrigation channels may be used as storm drains. Rainfall may also allow collected animal feces along the banks of the canal to wash into the channel. In some areas of the world animals are allowed to graze along the canal banks. In some urban areas they are used as linear parks where pet feces may collect to be later washed into the canal during a storm event. Studies are limited on the occurrence of pathogens in irrigation water impacted by nondirect or purposeful reuse of sewage in both developed and developing countries. In a study of irrigation waters in several Central American countries and the United States, the protozoan parasites Giardia, Cryptosporidium, and microsporidia were detected (Thurston-Enriquez and others 2002). Giardia concentrations were similar in almost all countries (60% of the samples), whereas Cryptosporidium concentrations were much greater in Central America. In a study in western Mexico, 48% of the surface irrigation waters were positive for Cryptosporidium oocysts and 50% for Giardia cysts (Chaidez and others 2005). Both of these parasites have also been detected in irrigation water used for bean sprout irrigation in Norway (Robertson and Gjerde 2001). Salmonella has been reported in 2–14% of the irrigation waters used for produce production in Nigeria (Okafo and others 2003) and 23.5% of irrigation water used for cantaloupe production in Brazil (Espinoza-Medina and others 2006). Both E. coli O157:H7 and Salmonella have been reported in irrigation waters in western Canada (Gannon and others 2004). Izumi and colleagues (2008) reported the detection of both Salmonella and E. coli O157:H7 in irrigation water used to irrigate

The Role of Water and Water Testing

137

persimmons. In a study of irrigation water along the Rio Grande, Materon and others (2007) reported both E. coli O157:H7 and Salmonella in water used to irrigate cantaloupe. Kayed (2004) conducted a 2-year study of the occurrence of indicators and pathogens in two irrigation districts in Arizona. Although geometric means of E. coli varied from 4 to 81 per 100 ml, depending on the sample location, levels above two million per 100 ml were detected after periods of rainfall. Samples collected from a water irrigation district located in central Arizona yielded a variety of foodborne pathogens. In 2.3% of the samples Cryptosporidium was detected, 55% yielded Campylobacter, and 18% were positive for norovirus. The source waters were obtained from recreation lakes, but there was no intentional discharge of sewage into the source waters. In the same canals Carpenter (2007) found greater concentrations of E. coli and longer survival of both E. coli O157:H7 and Salmonella in the canal sediment. The likelihood of the edible parts of the plants becoming contaminated during irrigation depends upon a number of factors, including growing location of the edible portion of the produce (e.g., distance from the soil or water surface), frequency of irrigation, surface of the edible portion (i.e., smooth, rough, or webbed), and type of irrigation method (i.e., furrow or flood irrigation, sprinkler, or drip). If the edible part of the crop grows in or near the soil surface, it is more likely to become contaminated than fruit growing in the aerial parts of the plant. Some produce surfaces are furrowed or have structures that retain water (e.g., a pepper vs. cantaloupe). There are three types of irrigation systems: sprinkler, gravity-flow (furrow), and microirrigation systems. Microirrigation systems include surface drip and subsurface drip irrigation. The type of irrigation system greatly influences the degree of crop contamination that occurs during irrigation. Stine and others (2005a,b) and Gerba and Choi (2006) compared coliphage contamination of cantaloupe, iceberg lettuce, and bell peppers by various methods of irrigation. Virus transfer to the lettuce was 4.2%, 0.02%, and 0.00039% for spray, furrow, and drip irrigation, respectively. Drip and flood irrigation did result in some contamination of the cantaloupe, but not of the bell peppers. Alum (2001) observed virus contamination of lettuce, tomato, and cucumber when high levels of coliphage, poliovirus type 1, hepatitis A virus, and adenovirus type 40 were added to irrigation water used to flood irrigate the crops. Flood irrigation of produce with wastewater results in crop contamination with enteric bacteria (Heaton and Jones 2008; Ensink and others 2007; Okafo and others 2003). Extensive contamination of lettuce was reported with spray irrigation of water seeded with E. coli O157:H7 (Solomon and others 2003). Use of contaminated water for pesticide application may also serve as another mechanism of produce contamination. Guan and colleagues (2005) found that Salmonella and E. coli O157:H7 could grow in various pesticide solutions and contaminate tomato plants when they were applied as a spray. Quantitative microbial risk assessment models for the use of reclaimed water show that the risk varies with the crop, with lettuce posing a higher risk than cucumber, but comparable to that of broccoli and cabbage (Hamilton and others 2006). The interval between irrigation and harvest will affect the likelihood of pathogens surviving to reach the consumer. However, some pathogens like hepatitis A can survive for weeks on produce before harvest (Stine and others 2005b) and some salads are harvested

138

Section II. Preharvest Strategies

within 24 hours of the last irrigation event allowing little time for pathogen die-off (Tyrell and others 2006).

Standards for Irrigation Water Quality Most standards for the microbial quality of irrigation water have been developed for the use of treated wastewater. Few standards have been suggested for water derived from other sources. Based on results of a study of irrigation waters in the western United States in the late 1960s Geldreich and Bordner (1971) suggested an irrigation standard of 1,000 fecal coliforms/100 ml based on the absence of Salmonella in irrigation waters, which had values below this level. Several states in the United States have standards for irrigation of food crops if reclaimed wastewater is used; however, irrigation with reclaimed water is seldom practiced. The state of California has a treatment requirement with a coliform standard of <2 coliforms/100 ml for the treated water. The WHO (2006) recently revised its recommendations for the safe reuse of wastewater using a risk-based approach for development of treatment and microbial standards. The WHO guidelines for the safe use of wastewater in agriculture are based on a risk analysis approach and recommends treatment requirements for pathogen reduction, monitoring (verification of treatment performance), and management strategies. The level of protection goal is defined in terms of disability adjusted life years (DALY), a measure that combines years of life lost by premature mortality with years lived with a disability, standardized or weighted by severity of illness (Pruss and Havelaar 2001). The acceptable level of risk from consumption of pathogens on food is defined as 10−6 DALY (WHO 2006). Based on this analysis the minimum requirements for irrigation water for use on root crops are equal or less than 1,000 E. coli per 100 ml and zero helminth eggs per liter. This guideline is based upon a wastewater treatment process that provides a 4-log (99.99%) reduction in pathogens (approximately equivalent to an E. coli of 1,000/100 ml in unchlorinated effluents), a 2-log pathogen reduction due to die-off between the last irrigation and consumption, and a 1-log reduction by washing of the salad crops or vegetables with water prior to consumption. The WHO believes this option provides the needed 7-log pathogen reduction for crops eaten uncooked. For options totally dependent on the treatment to remove pathogens (again a 7-log reduction) to the required level of acceptable risk, the E. coli level in irrigation water for crops eaten uncooked should be equal to or less than one E. coli per 100 ml.

Process Waters Feces-contaminated water postharvest can also result in the contamination of produce. In Ghana it was found that water was an important source of bacterial contamination (Keratia and Drechsel 2004). Contamination of wash water is most likely to occur if tanks of water are used or if the water is recirculated during the process. Chaidez and others (2005) in Mexico found that 16% of the wash water tanks used in the fresh produce industry contained Cryptosporidium oocysts and 83% Giardia cysts. Levels of chlorine normally used in the wash water tanks (200 mg/l) would expect to have little impact against the Cryptosporidium oocysts. Castillo and others (2004) documented the occurrence of Salmonella and E. coli in wash water used to process can-

The Role of Water and Water Testing

139

taloupe along the Rio Grande River in the United States, but observed that the low level observed did not have an impact on the melons being processed. The use of contaminated ice may also be source enteric pathogens (Cannon and others 1991).

Conclusions The amount of information on the microbiological quality of water used in the production of produce is limited, especially for enteric pathogens. Such information is needed to better assess the role and risks contaminated water plays in disease transmission by fresh produce. The role of irrigation waters in produce contamination is clear when feces-contaminated waters are used. However, application of quantitative microbial risk assessment is needed to better understand the impact on low-level transmission. The levels of enteric pathogens acceptable for irrigation and process water need to be defined. Standards need to be developed for irrigation waters that are meaningful and reduce the risk of produce contamination taking into consideration the means of irrigation and the type of produce. Irrigation systems are very complex, and sources and ecology of fecal contamination from nonsewage sources need to be better studied. The role water plays in produce contamination has been overlooked and clearly needs more study to ensure it is not a source of pathogens in produce consumed raw.

References Abbaszedegan M, LeChevallier M, Gerba C. 2003. Occurrence of viruses in US groundwaters. J Am Water Works Assoc 95:107–120. Alum A. 2001. Control of viral contamination of reclaimed water irrigated vegetables by drip irrigation. Ph.D. diss., University of Arizona, Tucson. Artz RR, Killham K. 2002. Survival of Escherichia coli O157:H7 in private drinking water wells: influence of protozoan grazing and elevated copper concentrations. FEMS Environ Microbiol 7:241–248. Ashbolt NJ, Grabow WOK, Snozzi M. 2001. Indicators of microbial water quality. In Water Quality: Guidelines and Health, edited by Fewtell L, Bartram J. pp. 289–315. London: IWA Publishing. Avery LM, Hill P, Killham K, Jones D. 2004. Escherichia coli O157:H7 survival following the surface and sub-surface application of human pathogen contaminated organic waste to soil. Soil Biol Biochem 36:2101–2103. Bosch A, Pinto RM, Abad FX. 2006. Survival and transport of enteric viruses in the environment. In Viruses Foods, edited by Goyal SM. pp. 151–187. New York: Springer. Bruce MG, Curtis MB, Payne MM, Gautom RK, Thompson EC, Bennett AL, Kobayashi JM. 2003. Lakeassociated outbreak of Escherichia coli O157:H7 in Clark County, Washington, August 1999. Arch Pediatr Adolesc Med 157:1016–1021. Byappanahalli M, Fujioka R. 2004. Indigenous soil bacteria and low moisture may limit but allow fecal bacteria to multiply and become a minor population in tropical soils. Water Sci Technol 50:27–32. Cabbelli VJ. 1981. Swimming-associated illness and recreational water quality criteria. Water Sci Technol 21:13–21. Cannon R, Hirschhorn JRB, Rodeheaver DC, Silverman EA, Brown GH, Talbort SE, Stine SE, Monroe SS, Dennis DT. 1991. A multistate outbreak of Norwalk virus gastroenteritis associated with consumption of commercial ice. J Infect Dis 164:860–863. Carpenter K. 2007. The ecology of bacterial pathogens and Salmonella in irrigation canals. M.S. thesis, University of Arizona, Tucson. Carrillo M, Estrada E, Hazen TC. 1985. Survival and enumeration of the fecal indicators Bifidobacterium adolescentes and Escherichia coli in a tropical rain forest watershed. Appl Environ Microbiol 50:468–476.

140

Section II. Preharvest Strategies

Castillo A, Mercado I, Lucia LM, Martinez-Ruiz Y, Ponce de Leon J, Murano EA, Acuff GR. 2004. Salmonella contamination during production of cantaloupe: a binational study. J Food Protect 67:713–720. Chaidez C, Soto M, Gortares P, Mena K. 2005. Occurrence of Cryptosporidium and Giardia in irrigation water and its impact on fresh produce industry. Intl J Environ Hlth Res 15:339–345. Chapron CD, Ballester NA, Fontaine JH, Frades CN, Margolin AB. 2000. Detection of astroviruses, enteroviruses, and adenoviruses types 40 and 42 in surface waters collected and evaluated by the information collection rule and an integrated cell culture-nested PCR procedure. Appl Environ Microbiol 66:2520–2525. Curriero FC, Patz JA, Rose JB, Lele S. 2001. The association between extreme precipitation and waterborne disease outbreaks in the United States, 1948–1994. Am J Public Hlth 91:1194–1199. Czajkowska D, Witkowska-Gwiazdowska A, Sikorska I, Boszczyk-Maleszak H, Horoch M. 2005. Survival of Escherichia coli serotype O157:H7 in water and in bottom-shore sediments. Pol J Environ Stud 14:423–430. Donovan EP, Staskal DF, Unice KM, Roberts JD, Haws LC, Finley BL, Harris MA. 2008. Risk of gastrointestinal disease associated with exposure to pathogens in the sediments of the lower Passaic River. Appl Environ Microbiol 74:1004–1018. Ensink JHJ, Mahmood T, Dalsgaard A. 2007. Wastewater-irrigated vegetables: market handling versus irrigation water quality. Trop Med International Hlth 12(suppl 2):2–7. Espinoza-Medina IE, Rodríguez-Leyva FJ, Vargas-Arispuro I, Islas-Osuna MA, Acedo-Felix E, MartinezTellez MA. 2006. PCR identification of Salmonella: potential contamination sources from production and postharvest handling of cantaloupes. J Food Protect 69:1422–1425. Falabi JA, Gerba CP, Karpiscak MM. 2002. Giardia and Cryptosporidium removal from waste-water by a duckweed (Lemmna gibba L.) covered pond. Lett Appl Microbiol 34:384–387. Feachem R G, Bradley DJ, Garelick H, Mara DD. 1983. Sanitation and Disease. Health Aspects of Excreta and Wastewater Management. New York: Wiley. Fujioka R, Sian-Denton C, Boja M, Morphew K. 1999. Soil: the environmental source of Escherichia coli and enterococci in Guam’s streams. J Appl Microbiol 85:83S–89S. Gagliardi JV, Karns JS. 2000. Leaching of Escherichia coli O157:H7 in diverse soils under various agricultural management practices. Appl Environ Microbiol 66:877–883. Gannon VP, Graham TA, Read S, Ziebell K, Muckie A, Mori J, Thomas J, Selinger B, Townshed I, Bryne J. 2004. Bacterial pathogens in rural water supplies in southern Alberta, Canada. J Toxicol Environ Hlth A 67:1643–1653. Geldreich EE, Bordner RH. 1971. Fecal contamination of fruits and vegetables during cultivation and processing for markets. J Milk Food Technol 34:184–198. Geldreich EE, Fox KR, Goodrich JA, Rice EW, Clark RM, Swerdlow DL. 1992. Searching for a water supply connection in the Cabool, Missouri disease outbreak of Escherichia coli O157:H7. Water Res 26:1127–1137. Gennaccaro AL, McLaughlin MR, Quintero-Bentancourt W, Huffman DE, Rose JB. 2003. Infectious Cryptosporidium parvum oocysts in final reclaimed effluent. Appl Environ Microbiol 69:4983–4984. Gerba CP. 2000. Assessment of enteric pathogens shedding by bathers during recreational activity and its impact on water quality. Quant Microbiol 2:55–68. Gerba CP, Choi CY. 2006. Role of irrigation water in crop contamination by viruses. In Viruses in Foods, edited by Goyal SM. pp. 257–263. New York: Springer. Gerba CP, Rose JB, Haas CN, Crabtree KD. 1996. Waterborne rotavirus: a risk assessment. Water Res 30:2929–2940. Goyal SM, Gerba CP, Melnick JL. 1977. Occurrence and distribution of bacterial indicators and pathogens in canal communities along the Texas coast. Appl Environ Microbiol 34:139–149. Grimes DJ. 1980. Bacteriological water quality effects of hydraulically dredging contaminated upper Mississippi River bottom sediment. Appl Environ Microbiol 39:782–789. Guan TT, Blank G, Holley RA. 2005. Survival of pathogenic bacteria in pesticide solutions and on treated tomato plants. J Food Protect 68:296–304. Hamilton AJ, Stagnitti F, Premier R, Boland A-M, Hale G. 2006. Quantitative microbial risk assessment models for consumption of raw vegetables irrigated with reclaimed water. Appl Environ Microbiol 72:3284–3290.

The Role of Water and Water Testing

141

Heaton JC, Jones K. 2008. Microbial contamination of fruit and vegetables and the behavior of enteropathogens in the phyllosphere: a review. J Appl Microbiol 104:613–626. Hendricks CW. 1971a. Enteric bacterial metabolism of stream sediment eluates. Canad J Microbiol 17:551–556. ———. 1971b. Increased recovery rate of salmonellae from stream bottom sediments versus surface waters. Appl Microbiol 21:379–380. Hunter P. 2003. Drinking water and diarrheal disease due to Escherichia coli. J Water Hlth 1:65–72. Izumi H, Tsukada Y, Poubol J, Hisa K. 2008. On-farm microbial contamination of persimmon fruit in Japan. J Food Protect 71:52–59. Jiang SC, Chu W. 2004. PCR detection of pathogenic viruses in southern California urban rivers. J Appl Microbiol 97:17–28. Kadlec RH, Knight RL. 1996. Treatment Wetlands. Boca Raton: CRC Press. Karim MR, Mashadi FD, Karpiscak MM, Gerba CP. 2004. The persistence and removal of enteric pathogens in constructed wetlands. Water Res 38:1831–1837. Kayed D. 2004. Microbial quality of irrigation water used in the production of fresh produce in Arizona. Ph.D. diss., University of Arizona, Tucson. Keratia BN, Drechsel P. 2004. Agricultural use of untreated urban wastewater in Ghana. In Wastewater Use in Irrigated Agriculture, edited by Scott CA, Faruqui NI, Raschid-Sally L. pp.101–112. Wallingford, UK: Commonwealth Agricultural Bureau International Publishing. Keswick BH, Gerba CP. 1980. Viruses in groundwater. Environ Sci Technol 14:1290–1297. Kurokawa K, Tani K, Ogawa M, Nasu M. 1999. Abundance of bacteria carrying sltil gene in natural river water. Lett Appl Microbiol 28:405–410. Kutz SM, Gerba CP. 1988. Comparison of virus survival in fresh-water sources. Water Sci Technol 20:467–471. LeChevallier MW, Norton WD, Lee RG. 1991. Occurrence of Giardia and Cryptosporidium spp. in surface water supplies. Appl Environ Microbiol 57:2610–2616. Liew PF, Gerba CP. 1980. Thermostabilization of enteroviruses by estuarine sediment. Appl Environ Microbiol 40:305–308. Materon LA, Martinez-Garcia M, McDonald V. 2007. Identification of sources of microbial pathogens from pre-harvest to post-harvest operations. Wld J Microbiol Biotechnol 23:1281–1287. McGee P, Bolton DJ, Sheridan JJ, Earley B, Kelly G, Leonard N. 2002. Survival of Escherichia coli O157:H7 in farm water: its role as a vector in the transmission of the organism within herds. J Appl Microbiol 93:706–713. Mena K, Gerba CP. 2008. Waterborne adenovirus. Rev Environ Contam Toxicol, in press. Moulton-Hancock C, Rose JB, Vasconcelos GJ, Harris SI, Klonicki PT, Sturbaum GD. 2000. Giardia and Cryptosporidium occurrence in groundwater. J Am Water Works Assoc 92(9)117–123. Mushahwar IK. 2008. Hepatitis E virus: molecular virology, clinical features, diagnosis, transmission, epidemiology, and prevention. J Med Virol 80:646–658. Obiri-Danso K, Jones K. 1999. Distribution and seasonality of microbial indicators and thermophilic camplyobacters in two freshwater bathing sites on the River Lune in northwest England. J Appl Microbiol 87:822–832. Okafo CN, Umoh VJ, Galadima M. 2003. Occurrence of pathogens on vegetables harvested from soils irrigated with contaminated streams. Sci Total Environ 311:49–56. Pruss A, Havelaar A. 2001. The global burden of disease study and applications in water, sanitation and hygiene. In Water Quality: Guidelines, Standards and Health, edited by Fewtrell L, Bartram J. pp.43–88. London: IWA Publishing. Ramteke PW, Tewari S. 2007. Serogroups of Escherichia coli from drinking water. Environ Monit Assess 130:215–230. Regli S, Rose JB, Hass CN, Gerba CP. 1991. Modeling the risk from Giardia and viruses in drinking water. J Am Water Works Assoc 83:76–84. Reynolds KA, Mena KD, Gerba CP. 2008. Risk of waterborne illness via drinking water in the United States. Rev Environ Contam Toxicol 192:117–158. Ritchie JM, Campbell GR, Shepherd J, Beaton Y, Jones D, Killham K, Artz RRE. 2003. A stable bioluminescent construct of Escherichia coli O157:H7 for hazard assessments of long-term survival in the environment. Appl Environ Microbiol 69:3359–3367.

142

Section II. Preharvest Strategies

Robertson LJ, Gjerde B. 2001. Occurrence of parasites on fruits and vegetables in Norway. J Food Protect 64:1793–1798. Rose JB, Gerba CP, Jakubowski W. 1991. Survey of potable water supplies for Cryptosporidium and Giardia. Environ Sci Technol 25:1393–1400. Sails AD, Bolton FJ, Fox AJ, Wareing DRA, Greenway DLA. 2002. Detection of Campylobacter jejuni and Campylobacter coli in environmental waters by PCR enzyme-linked immunosorbent assay. Appl Environ Microbiol 68:1319–1324. Schets FM, During M, Italiaander R, Heijnen L, Rutjes SA, van der Zwaluw WK, de Roda Husman AM. 2005. Escherichia coli O157:H7 in drinking water from private water supplies in the Netherlands. Water Res 39:4485–4493. Shannan L. 1998. Irrigation development: proactive planning and interactive management. pp. 251–276. H. Bruins and L. Harvey (ed.). The Arid Frontier. London: Kluwer Academic Press. Smith EM, Gerba CP, Melnick JL. 1978. Role of sediment in persistence of enteroviruses in estuarine sediment. Appl Environ Microbiol 35:685–689. Smith HV, Grimason AM. 2003. Giardia and Cryptosporidium in water and wastewater. In The Handbook of Water and Wastewater Microbiology, edited by Duncan M, Horan N, pp. 695–756. San Diego: Academic Press. Solo-Gabriele HM, Wolfert M, Desmarais TR, Palmer CJ. 2000. Sources of Escherichia coli in a coastal subtropical environment. Appl Environ Microbiol 66:230–237. Solomon EB, Pang HJ, Matthews KR. 2003. Persistence of Escherichia coli O157:H7 on lettuce plants following spray irrigation with contaminated water. J Food Protect 66:2198–2202. Stine SW, Song I, Choi CY, Gerba CP. 2005a. The effect of relative humidity on preharvest survival of bacterial and viral pathogens on the surface of cantaloupe, lettuce, and bell peppers. J Food Protect 68:1352–1358. ———. 2005b. Application of microbial risk assessment to the development of standards for enteric pathogens in water used to irrigate fresh produce. J Food Protect 68:913–918. Thomas KM, Charron DF, Waltner-Toews D, Schuster C, Maarouf AR, Holt JD. 2006. A role of high impact weather events in waterborne disease outbreaks in Canada, 1975–2001. Int J Environ Hlth Res 16:167–180. Thurston-Enriquez JA, Watt P, Dowd SC, Enriquez R, Pepper IL, Gerba CP. 2002. Detection of protozoan parasites and microsporidia in irrigation waters used for crop production. J Food Protect 65:378–382. Tyrell SF, Knox JW, Weatherhead EK. 2006. Microbiological water quality requirements for salad irrigation in the United Kingdom. J Food Protect 69:2029–2035. Van Donsel DJ, Geldreich, EE. 1971. Relationships of salmonellae to fecal coliforms in bottom sediments. Water Res 5:1079–1087. Vereen E, Lowrance RR, Cole DJ, Lipp EK. 2007. Distribution and ecology of Campylobacters in coastal plain streams (Georgia, United States of America). Appl Environ Microbiol 73:1395–1403. Wang G, Doyle MP. 1998. Survival of enterohemorrhagic Escherichia coli O157:H7 in water. J Food Protect 61:662–667. WHO (World Health Organization). 2006. Guidelines for the safe use of wastewater, excreta and greywater. Volume 2: Wastewater use in Agriculture. Geneva: WHO. Zaleski KJ, Josephson KL, Gerba CP, Pepper IL. 2005. Survival, growth, and regrowth of enteric indicator and pathogenic bacteria in biosolids, compost, soil and land applied biosolids. Residuals Sci Technol 2:49–63.

8 The Role of Manure and Compost in Produce Safety Xiuping Jiang and Marion Shepherd

Introduction Fresh fruits and vegetables are vital components of a healthy and balanced diet. Most fruits and vegetables are considered as ready-to-eat (RTE) food with minimal or no terminal treatments before consumption. However, foodborne disease outbreaks in recent years have been increasingly associated with the consumption of fresh produce in the United States. The majority of these outbreaks were due to pathogens transmitted via fecal-oral routes (Johnston and others 2006). Animal wastes are an important source of nutrients for crop production, but may contain a variety of human pathogens such as Esherichia coli O157:H7, Salmonella spp., and Listeria monocytogenes (Pell 1997). In the past, concerns over animal manure were focused primarily on the nutrient’s effect on the environment and the inefficient use in agricultural systems, with little effort on characterizing the fate and reduction of pathogens (Sobsey and others 2001). In addition to the contaminated irrigation or runoff waters, raw or improperly composted manure used as organic fertilizer and soil amendments have been identified as potential vehicles for contamination of vegetables during preharvest (FDA 2001; Buck and others 2003). Numerous cases of foodborne illnesses have been associated with the consumption of fresh vegetables in part contaminated by manure from ruminants and poultry (Schlech and others 1983; Morgan and others 1988; Cieslak and others 1993; CDC 1996, 2007). Recently, several large produce-related outbreaks of E. coli O157:H7 and Salmonella spp. have led to major concerns about the contamination of vegetables with fecal pathogenic bacteria in the agricultural environment (CDC 2006; Doyle and Erickson 2008). Land spreading of animal manure can lead to the introduction of enteric pathogens into the food chain. Foodborne pathogens such as E. coli O157:H7, Salmonella, and L. monocytogenes can survive in soil for several months following the manure application (Kudva and others 1998; Fenlon 2000; Jiang and others 2002, 2004; Natvig and others 2002), and contaminate the vegetables grown in the soil fertilized with manure (Ingham and others 2004, 2005; Islam and others 2004a–d, 2005). Therefore, controlling pathogen populations in the animal wastes used for agricultural production should help in reducing pathogen contamination of the preharvest environment. Due to the large production of animal manure by animal production systems, how to dispose of these wastes safely can be a very challenging task (US SAC 1998). Composting is used by farmers to treat various types of animal waste, and the compost can then be utilized as the value-added fertilizer and soil amendment. Composting as a practical way for pathogen inactivation is largely due to a self-heating process carried 143

144

Section II. Preharvest Strategies

out by microorganisms (Rynk 1992). Due to the outdoor nature, diversity of waste materials, contamination levels, and waste handling practices, the outcome of composting varies significantly between farms, and even between compost heaps on the same farm. Furthermore, there is growing interest in organic foods because consumers believe they are healthier than conventionally grown foods. According to a United States Department of Agriculture (USDA) report, consumption of organic products has increased by more than 20% annually since 1990, with organic vegetables and fruits leading the way (Dimitri and Greene 2002). It is expected that the rapid growth of organic foods, especially fresh produce and fruits, will continue in the future. However, organic foods may be less safe than conventionally grown produce because organic farmers largely use animal manure or compost as a fertilizer for their crops (Mukherjee and others 2004). Because of the increased economic importance of organic produce and less treatments on these products to inactivate any pathogen that may be present, it is important for organic growers to use scientifically valid methods for treating and using manure safely. This chapter focuses on the role of animal manure and compost in produce safety.

Fresh Produce Outbreaks Due to Fecal Contamination of Produce in the Field The number of cases of foodborne diseases annually that are associated with the consumption of raw vegetables and fruits has doubled in the United States during the past 2 decades prior to 1999 (NACMCF 1999). Recent data revealed that about 13% of outbreaks with an identified food source from 1990 to 2005 were linked to fresh produce (DeWaal and Bhuiya 2007). Many varieties of bacteria, viruses, and parasites have been epidemiologically linked to fresh produce-related outbreaks. These microorganisms include various serotypes of Salmonella spp., E. coli O157:H7, enterotoxigenic E. coli, Vibrio cholera, L. monocytogenes, Shigella spp., Campylobacter spp., Bacillus cereus, Clostridium botulinium, Cryptosporidium spp., Cyclospora spp., Hepatitis A, and Norwalk/Norwalklike viruses (FDA 2001; Johnston and others 2006; NACMCF 1999). The most common produce items associated with outbreaks include leafy greens–based salads (such as lettuce), potatoes, tomatoes, and sprouts (Buck and others 2003; Sivapalasingam and others 2004; DeWaal and Bhuiya 2007). Table 8.1 lists some foodborne disease outbreaks with possible fecal contamination of fresh produce with human pathogens. For example, an outbreak of L. monocytogenes in 1981 in Nova Scotia was traced back to the coleslaw that had been made from cabbage grown in a field fertilized with manure from Listeria-infected sheep (Schlech and others 1983). Most recently, a large E. coli O157 outbreak associated with bagged baby spinach was linked to potential sources such as wild pig feces present in and around spinach fields, nearby irrigation wells and surface waterways exposed to feces from cattle, and wildlife on one organic ranch in California (CalFERT 2007). However, the exact source of spinach contamination could not be confirmed due to the complexity associated with spinach production and processing; this was also true for many other outbreaks listed in the table.

Table 8.1. Summary of produce-associated foodborne disease outbreaks with possible fecal contamination Pathogen E. coli O157:H7

Product Contaminated Potatoes

Lettuce

Number of Cases 24 8

Year, Location 1985, U.K. 1995, U.K.

92

1995, U.S.

Alfalfa sprouts

108

1997, U.S.

Mesclun lettuce

49 120

1996, U.S. 2005, Sweden

Spinach

205

2006, U.S.

4

1992, U.S.

23

1991, U.S.

56

1996, U.S.

14

1998, Canada

Vegetables Apple juice (unpasteurized)

7

1999, U.S.

Comments

References

Cow manure contamination of potatoes was implicated. Manure was used as the fertilizer for growing potatoes; however, the pathogen was not detected on the implicated potatoes. Possible routes of contamination included fertilizing fields with improperly aged cattle manure–based compost, application of cattle manure–contaminated irrigation water, and direct or indirect exposure to ruminant feces. Seeds were possibly contaminated by cattle manure from nearby feedlots, deer feces, or fecally contaminated irrigation water. Cattle were found near the fields where lettuce was grown. Fecal contaminated irrigation water application to lettuce crop was implicated as transmission vehicle. Direct and indirect contamination with animal feces; outbreak strain from spinach matched those from cattle and feral swine feces at one ranch suspected of producing contaminated spinach. Vegetable garden was fertilized with cow manure and the pathogen was isolated from the manured soil. Dropped apples were used in making cider, and cattle raised near where apples were picked. Dropped apples in an orchard where cattle and deer feces were present were used in juice production. Dropped apples were used in juice production after cattle had been kept in the orchard before harvesting. Dropped apples contaminated with domestic and/or wild animal manure were used in juice production.

Morgan and others 1988 Chapman and others 1997 Ackers and others 1998

Breuer and others 2001 Hilborn and others 1999 Söderström and others 2005 CDC 2006; FDA 2006

Cieslak and others 1993 Besser and others 1993 CDC 1996; Cody and others 1999 Tamblyn and others 1999 Farber 2000

145

146

Table 8.1. Continued Pathogen

Product Contaminated Alfalfa Sprouts

Number of Cases 242

Year, Location 1995, U.S.

Alfalfa and clover sprouts Cantaloupes

60

1997, U.S.

24

1997, U.S.

47

2000, U.S.

50 58 82

2001, U.S. 2002, U.S. 2005, U.S.

Orange juice (unpasteurized)

62

1995, U.S.

L. monocytogenes

Coleslaw

41

1981, Canada

Y. pseudotuberculosis

Lettuce Carrots

47 76

1998, Finland 2003, Finland

V. cholerae

Cabbage

71

1991, Peru

C. parvum

Unpasteurized apple juice Salad items

160

1993, U.S.

Salmonella

Tomatoes

Hepatitis A

30

1996, Finland

Comments

References

Exposure to feces from rodents, birds, or other animals during pre- or postharvest was suspected. Contaminated seeds due to the use of contaminated fertilizers or grazing livestock in seed harvesting areas were suspected. The imported products contaminated via birds, rodents, or untreated manure were suspected. Irrigation water contaminated with sewage was implicated for cantaloupe contamination. As above. As above. Contaminated irrigation water or animal droppings were suspected; animal feces were present in and around drainage ditches where tomatoes were grown. Possible contamination routes of dropped oranges included direct contact with animals, soil, contaminated irrigation water, or improperly treated manure used as fertilizer. Cabbage used in the coleslaw was grown in field where raw and composted sheep manure were applied. Deer fecal matter was heavily present in lettuce fields. Contamination was thought to have occurred due to exposure to wildlife feces during storage. Farmers commonly irrigated crops with untreated sewage.

Mahon and others 1997

Manure contamination was likely; Cryptosporidium oocysts were detected in calf feces on farm. Two separate outbreaks occurred during the same time interval. Imported salad products were treated with contaminated irrigation water identified as transmission vehicle.

Mohle-Boetani and others 2001 Mohle-Boetani and others 1999 CDC 2002 CDC 2002 CDC 2002 CDC 2007

Cook and others 1998

Schlech and others 1983 Nuorti and others 2004 Jalava and others 2006 Swerdlow and others 1992 Millard and others 1994 Pebody and others 1998

The Role of Manure and Compost

147

Human Pathogens Associated with Animal Wastes Epidemiological investigations of produce contamination on farms have frequently looked into animal feces as potential sources and have focused on the use of raw manure or compost as fertilizer; adjacent land use with livestock or poultry operations; and presence of wild animals such as deer, other mammals, and birds (Beuchat 2006; CalFERT 2007). In addition, irrigation water and foliar application of compost tea are important sources for produce contamination in the field. Therefore, fully understanding the prevalence, survival, and transmission of human pathogens in animal feces is essential for developing intervention strategies for controlling produce contamination on farms. Animal Wastes According to the 2002 census of agriculture, the livestock and poultry industries produced ca. $106 billion worth of products sold in the U.S., which included over 74 million cattle and calves, ca. 185 million hogs and pigs, and ca. 8.5 billion broilers and other meat-type chickens (USDA 2002). With such a large number of livestock and poultry raised in this country, these animals also generate substantial volumes of manure commonly called wastes. An estimated 1.36 billion tons of manure are produced annually in the United States, of which approximately 90% is generated by cattle, 5% by poultry, and the rest by swine (US SAC 1998). Wastes generated from animal-based agricultural enterprises come from cattle feedlots, dairy farms, poultry, swine, pastures, and meat and poultry processing plants. Human Pathogens in Animal Wastes Manure is the mixture of animal excreta such as feces and urine, bedding materials, and other secretions from the animal (Himathongkham and others 1999). Many species of microorganisms are present in animals and their feces, including Aeromonas hydrophila, Arobacter butzleria, Bacillus anthracis, Brucella abortus, Campylobacter jejuni, Chlamydia psittaci, Clostridium perfringens, Clostridium botulinium, Coxiella burneti, E. coli, Erysipelothrix rhusiopathiae, Francisella tularensis, Leptospira spp., L. monocytogenes, Mycobacterium tuberculosis, Mycobacterium paratuberculosis, Salmonella spp., and Yersinia spp. (Sobsey and others 2001). Animal manure frequently contains enteric microorganisms that are human pathogens (Zhao and others 1995; Pell 1997; Kudva and others 1998). Among those 11 agents associated with produce-borne outbreaks, as identified by the National Advisory Committee on Microbiological Criteria for Foods (NACMCF 1999), E. coli O157:H7, Salmonella spp., and L. monocytogenes are responsible for the majority of outbreaks. E. coli O157:H7 are carried by ruminants, especially cattle, and are shed in their feces (Meng and others 2001). Salmonella spp. inhabits the intestinal tracts of a variety of animals with poultry and eggs remaining as a predominant reservoir (Wray and Davies 2003). L. monocytogenes is widely distributed in the environment and is associated with animal feces, decaying vegetation, soil, silage, water, and fresh vegetables (Swaminathan 2001). Fecal excretion of pathogens by animals can be affected by various factors, including species of animals, age, health status, diets, seasonal effect, and farm practices

148

Section II. Preharvest Strategies

(Diez-Gonzalez and others 1998; Keener and others 2004). A significant number of animals may be asymptomatic carriers of enteric pathogens but shedding the organisms in feces (Meng and others 2001). As a primary reservoir of E. coli O157:H7, cattle, especially weaned calves and heifers, shed this microorganism intermittently with events lasting 2 to 4 months (Zhao and others 1995). The populations of E. coli O157:H7 ranged from enrichment positive to 105 CFU/g feces. Epidemiological surveys revealed the prevalence of E. coli O157 in feces from less than 6% to 28% depending on seasons, geographic locations, and herds (Hancock and others 1994; Zhao and others 1995; Elder and others 2000). Dietary change has also been shown to affect fecal excretion patterns of E. coli O157:H7 in cattle and sheep (Kudva and others 1995; Diez-Gonzalez and others 1998). A recent study suggests that the use of dried distillers’ grain as feed additive may contribute to the higher prevalence of E. coli O157 in cattle (Jacob and others 2008). The authors hypothesized that the decreased starch in the feed may alter the ecology in the rumen and favor the growth of E. coli O157:H7. The populations of Salmonella spp. have been detected in amounts up to 107 CFU/g in feces of healthy animals (Pell 1997; Himathongkham and others 1999). Campylobacter spp. is present in poultry wastes, and average populations have been reported to be near 105 CFU/g in feces (Stern and Robach 2003).

Persistence of Enteric Pathogens in Preharvest Environment In Animal Feces and Slurry The intestinal tracts of animals are the natural habitats for enteric pathogens. Immediately after being defecated in feces, these pathogens are exposed to a hostile environment with numerous microorganisms, from both animals and the environment, to compete for nutrients. Studies have demonstrated that some enteric pathogens such as E. coli O157:H7 and L. monocytogenes were still able to multiply for ca. 1∼3 logs CFU/g in fresh feces within 2–3 days at warm temperatures, followed by steady decline in bacterial populations during storage (Wang and others 1996; Himathongkham and Riemann 1999). Even under field conditions, both indicator bacteria and Salmonella multiplied initially for ca. 1.5 logs in bovine feces on pasture (Sinton and others 2007; Van Kessel and others 2007). Therefore, in order to eliminate pathogens from feces effectively, the initial treatment of feces should be emphasized. During storage, exponential inactivation of enteric pathogens in feces was reported in several studies. When stored at various temperatures, E. coli O157:H7 and S. Typhimurium were inactivated in cattle manure following a first-order reaction with most rapid die-off at 37 °C (Wang and others 1996; Himathongkham and others 1999). The survival of other serotypes of shiga-toxin producing E. coli (STEC) such as O26 and O111 in naturally infected cow feces ranged from 1 to 18 weeks with extended survival at 15 °C (Fukushima and others 1999). Prolonged survival of E. coli O157:H7 was observed in manure heaps for up to 47 days, 4 months, and 21 months, in bovine, aerated ovine, and nonaerated ovine manure, respectively (Kudva and others 1998). L. monocytogenes survived in pig manure, soil, and cattle manure for 3∼4, 6∼8, and 6∼8 weeks, respectively, at 15 °C (van Renterghem and others 1991). When compared with cattle manure, high ammonia content (0.2–0.4%) and alkaline pH (8.6) in poultry

The Role of Manure and Compost

149

manure was considered an important factor in rapid inactivation of Salmonella, E. coli O157, and L. monocytogenes (Himathongkham and Riemann 1999). Reduction in water activity of feces apparently is another factor for pathogen inactivation (Wang and others 1996; Himathongkham and others 1999). These results indicate the potential for a robust population of pathogenic bacteria in manure, especially cattle manure, during storage and a potential vehicle for transmitting the pathogens to animals, food, and the environment. Himathongkham and others (1999), based on their research findings, recommended that cow manure should be held for 105 days at 4 °C or 45 days at 37 °C to achieve a 5-log10 reduction of both E. coli O157:H7 and S. Typhimurium. It is a most common practice on dairy farms to wash animal feces, urine, and feed debris into a slurry mixture, which is then held in a settling lagoon undergoing aerobic or anaerobic degradation for a period of time before disposal (Meyer and others 1997). During storage, both E. coli O157:H7 and S. Typhimurium had initial growth at 20 and 37 °C followed by population decline with decimal reduction times ranging from 2 days to 5 weeks in manure slurry (Himathongkham and others 1999). Fast pathogen reduction was observed when storage temperature was at 37 °C. Two other studies reported a faster population decline of E. coli O157:H7 in cattle manure slurry than in solid manure at temperatures below 23 °C, suggesting the influence of solids in slurries on the pathogen survival (Maule 1999; Kudva and others 1998). In addition to many studies on E. coli O157:H7, the fate of STEC O26 was investigated in cow slurry (Fremaux and others 2007b). The pathogen survived for at least 88 days in cow slurry without any genetic change. In Manure-Amended Soil Animal wastes in the form of raw or composted manure are routinely applied to the agricultural land as a fertilizer for the crops and soil amendment. Application of organic wastes can increase plant nutrients and organic matter and enhance the biological, chemical, and physical attributes of soil (Bulluck and others 2002; Gagliardi and Karns 2002). However, if not properly applied or disposed of, animal waste can leach out of the soil or be transported by runoff to pollute the water sources and the environment (Gagliardi and Karns 2000). Most importantly, the application of manure and manure slurries contaminated with enteric pathogens to soil provides an opportunity for pathogens to contaminate produce, drinking water, and irrigation water. When animal manure is incorporated into the soil, the antagonistic effect of soil microorganisms and the hostile environment of soil microcosm are likely factors in killing enteric bacteria. Properties of soil such as soil composition and texture, indigenous soil microorganisms, protozoal grazing, pH, water activity, oxidation-reduction potential, presence of a rhizosphere and microbial interactions can influence the survival or inactivation of pathogenic bacteria (Fenlon and others 2000; Brimecombe and others 2007). Under field conditions, other variables, such as solar radiation, temperature fluctuation, manure types, manure application rates, and desiccation, may also affect the persistence of enteric pathogens (Jones 1999). Effect of Soil Temperature on the Persistence of Enteric Pathogens Microbial activities are affected by the temperature to which the microorganisms are exposed. For example, E. coli O157:H7 survived in manure-amended sandy loam soil

150

Section II. Preharvest Strategies

for at least 56, 152, and 193 days at 5 °, 15 °, and 21 °C, respectively (Jiang and others 2002). Salmonella persisted in hog manure−amended loamy sand and clay soils for more than 180 days during the simulated summer-winter season (25, 10, 4, −18 °C) as compared with for less than 160 days either in a spring-summer temperature (4, 10, 25, 30 °C) or winter to summer (−18, 4, 10, 25 °C) regimens (Holley and others 2006). Similar results for S. Typhimurium in manure-amended soil were reported by Natvig and others (2002), who recommended applying manure in late fall to ensure that harvested vegetables were not contaminated with S. Typhimurium since the repeated freeze-thaw cycles could inactivate the pathogen. Influence of Microbial Activity in Soil Studies have revealed that soil with less microbial activity allows the extended survival of manure-borne enteric pathogens due to less competition from the indigenous soil microflora (FDA 2001; Jiang and others 2002, 2004). You and others (2006) found that Salmonella Newport in soil contaminated with dairy manure were detectable through direct plating for 107 and 158 days in manure mixed with nonsterilized soil and manure mixed with sterilized soil, respectively. Nutrient levels vary among different types of soil and so do the diversity and numbers of soil microbial populations. The number of E. coli increased by ca. 1.5–2 logs CFU/g during the first 2 weeks of manure incorporation but declined significantly greater in loamy sand soil than in silty clay loam soil (Lau and Ingham 2001). Presence of the Rhizosphere The rhizosphere, the soil environment surrounding the root, is a complex ecosystem where the interactions among soil, roots, and microbes take place (Brimecombe and others 2007). The rhizosphere is rich in organic compounds released by plant roots and microorganisms, which may affect the survival and growth of enteric bacteria introduced from manure. Using a soil core model, Gagliardi and Karns (2002) have shown that E. coli O157:H7 was able to survive for 25–47 days, 47–96 days, and 92 days, in fallow soil, on rye roots, and on alfalfa roots, respectively. The authors also showed that the metabolic activity of microbial community in soil microcosms was enhanced from rye and alfalfa roots after 14 days and by 63 days, respectively. Recently, a study by Klerks and others (2007) demonstrated that S. enterica serovars moved toward to the root exudates produced by the lettuce cultivar Tamburo via chemotaxis. Apparently, the persistence of enteric pathogens in the rhizosphere is the outcome of interaction among pathogens, soil microorganisms, soil, and plant roots. Types of Animal Wastes Animal wastes of different sources vary significantly in organic loads, pH, texture, microbial population, and solid contents (CAST 1996). Studies have revealed that the changes in diet can affect the shedding of pathogens by animals and subsequent persistence of the pathogens in manure and in manure-amended soil (Franz and others 2005). The populations of both E. coli O157:H7 and S. Typhimurium declined faster in the manure derived from straw diet than in the manure from grass silage plus maize silage (Franz and others 2005). Several studies investigated the persistence of E. coli and L. monocytogenes pathogens in soil-fertilized with a variety of wastes such as

The Role of Manure and Compost

151

sheep, pig, and cattle feces; ovine stomach content; abattoir houses; sewage treatment plants; or commercial creameries (Van Renterghem and others 1991; Dowe and others 1997; Avery and others 2004, 2005). The persistence of the pathogens varied among different wastes and even the wastes of similar origins, highlighting the variable nature of organic wastes. Rates and Ways for Manure Application Manure is abundant with microorganisms and nutrients. After incorporation into soil, the introduced nutrients become available for existing soil microorganisms to grow. The inhibitory effect of soil microorganisms on manure-borne pathogens was reported by a few studies. Lazarovits (2001) demonstrated that organic manure application to soil increased the overall populations of soil microorganisms by up to thousandfold and reduced populations of plant pathogens. The intensive application of manure to soil (e.g., 1 part manure to 10 parts soil vs. 1 : 25, 1 : 50, or 1 : 100) generally results in greater inactivation of E. coli O157:H7 at both 15 and 21 °C but not that much at 5 °C (Jiang and others 2002). The application methods used for animal wastes can affect the persistence of manure-borne pathogens. Hutchison and others (2004) have shown that the populations of Salmonella, L. monocytogenes, Campylobacter, and E. coli O157:H7 declined significantly slower in samples with animal waste incorporated into the soil immediately than in samples with the waste left on the soil surface. In contrast, another study reported that Salmonella survival was significantly longer when the hog slurry was surface-spread as compared with results from thoroughly mixed manure treatment (Holley and others 2006). Results from Gagliardi and Karns (2000) indicated that E. coli O157:H7 can travel below the top layers of soil for more than 2 months after manure application to a soil core, regardless of disturbed (tilled) or intact (untilled) soil core. In considering the extended survival of pathogens in manure-amended soil, fresh animal manure should not be applied to the land before adequate treatments are applied to reduce the bacterial populations significantly. The Length of Time Pathogens Persisted in Manure-Amended Soil Due to the declining nature of enteric pathogens in soil, initial bacterial load is a key factor to predict the length of pathogen persistence. Fenlon and others (2000) applied cattle slurry inoculated with 30 CFU E. coli O157:H7/100 ml to arable and grass plots on a clay loam soil. They could detect E. coli O157:H7 only in the soil and on the grass during the first week after application. In contrast, E. coli O157:H7 with very high application rates survived for at least 130 days on manure-amended soil cores with a grass cover at 18 °C (Maule 1999). Under field conditions, E. coli O157:H7, Salmonella, and Campylobacter persisted for up to 1 month, and L. monocytogenes for more than 1 month in both sandy arable and clay loam grassland soil fertilized with livestock manure. Survival times for Salmonella spp. were up to 300 days in soils spread with cattle slurry and 259 days for soils amended with animal feces (Jones 1986). Survival of Naturally Occurring Pathogens in Manure-Amended Soil Most pathogen survival studies used bacterial cultures cultivated in laboratory media for soil inoculation. Due to the sudden change of environment, the behavior of those

152

Section II. Preharvest Strategies

newly introduced cultures may not represent the natural fitness of enteric pathogens in feces. A few studies investigated the survival of naturally occurring E. coli O157 in sheep feces and cowpats on pasture, and raw cattle manure on garden soil. The survival of the pathogens ranged from 30 days to 15 weeks (Ogden and others 2002; Mukherjee and others 2006; Van Kessel and others 2007). Due to the uneven distribution of pathogens in the fecal samples collected, the results may not be suitable to validate the previous pathogen survival studies in manureamended soil. Current studies reveal that the length of pathogen persistence in manure-amended soil varied from a few days to more than 300 days, which is likely attributed to variation in types of wastes and environmental parameters used by different studies. Therefore, any recommendations of the land application of manure need to be based on studies that examined a wide range of organic wastes under various environmental conditions (Avery and others 2005). Pathogen Contamination on Fresh Produce Fertilized with Manure and Compost Major sources of fresh fruit and vegetable contamination on the farm include feces, raw manure, or inadequately treated compost, soil, irrigation water, water used for fungicides and insecticide applications, dust, wild animals, and human handling (Beuchat and Ryu 1997; Doyle and Erickson 2008). Enteric pathogens can survive in feces and manure-amended soil for extended periods of time, which may serve as the potential inoculum onto plants in the field. Both plant surfaces and the roots are the locations where pathogens can possibly grow or survive (Brandl 2006). The phyllosphere of plants is considered as a hostile environment for enteric microorganisms due to temperature fluctuations, low water activity, limitation in nutrients, and UV radiation, whereas the rhizosphere of plants provides a moist and low oxygen tension environment with large numbers of microorganisms and organic nutrients (Brandl 2006). The plant pathogens contaminate produce by splashing onto the leaves when heavy rain or water-gun irrigation occurs, being internalized from roots and other openings of the plants, imbedding into phylloplane biofilms, and adhering to the roots via soil particles (Fett 2000; Heaton and Jones 2008). Growing evidence has demonstrated that enteropathogenic bacteria, fitted for the intestinal habitat, have the ability to colonize the entire plant including roots, flowers, and seeds (Guo and others 2002; Brandl 2006). Application of raw manure, compost, or irrigation water containing enteric pathogens facilitates the movement of enteric pathogens from animal feces to the phyllosphere and the rhizosphere of plants in the field. Recently, considerable research efforts have been focused on the pathogen persistence in vegetables grown in soil fertilized with animal manure or compost, or irrigated with contaminated water (Natvig and others 2002; Islam and others 2004a–d, 2005). These studies were conducted either under controlled environments such as growth chambers or greenhouses, or in fields by examining various factors such as types of vegetables, time of pathogen introduction, soil type, pathogen inoculation levels, and different weather conditions.

The Role of Manure and Compost

153

Under Controlled Conditions Studies have been done on the contamination of both leafy vegetables and root crops grown in manure-amended soil with pathogens such as E. coli O157:H7 and Salmonella. Either in a growth chamber or greenhouse, the contamination of onions and carrots by E. coli O157 and radishes and arugula by S. Typhimurium occurred on the vegetable surface at harvest time ranging from 7 to 17 weeks after manure application (Natvig and others 2002; Islam and others 2004d). Franz and others (2005) found only one root sample of lettuce grown in loam soil amended with manure positive with E. coli O157:H7. However, Johannessen and others (2005) failed to detect the pathogen in the roots, outer leaves, and edible parts of the lettuce, but found the pathogen survived in soil for at least 8 weeks but not 12 weeks. In that study, analyzing lettuce samples for pathogen was done only at the harvest, which may have missed the detection of early lettuce contamination by manure-amended soil. In addition, a low inoculation level of pathogen may be another contributing factor for failing to detect pathogen on lettuce, because low inocula may reduce the probability for the bacteria to compete with indigenous microorganisms and colonize the rhizosphere (Solomon and others 2002). Most importantly, as long as the pathogen can be detected in soil, there is a possibility for produce being contaminated, especially when the washing step is not thoroughly performed. Under Field Conditions Using two avirulent strains, Islam and others (2004a–c, 2005) performed a series of experiments investigating the persistence of E. coli O157:H7 and S. Typhimurium on various plants grown in manure compost–amended soil or irrigated with contaminated water under field conditions. At an initial inoculation level of ca. 107 CFU/g, E. coli O157:H7 was detected at harvest on carrots, lettuce, and parsley for 126, 77, and 177 days, respectively, whereas S. Typhimurium was detectable at harvest on radishes, carrots, lettuce, and parsley for 84, 203, 63, and 231 days, respectively. These studies demonstrated that the extent of pathogen survival on produce varied among types of plants, bacterial species, types of soil, and compost used for growing vegetables although the contamination route via irrigation water or manure compost did not affect the survival curves of both strains. Van Renterghem and others (1991) reported that L. monocytogenes was detected 3 months later in 3 out of 6 radish samples, which were sown in soil inoculated with the pathogen in pig manure. However, a field study using manure naturally contaminated with E. coli O157:H7 failed to prove that pathogen contamination occurred from manured soil to lettuce (Johannessen and others 2004). As suggested by the researchers, heavy rainfall soon after fertilizer application possibly washed away E. coli O157:H7 from the lettuce field. Further research on determining the minimal level of pathogen in soil required for produce contamination may help to assess the risk of using contaminated manure or compost for produce production in the field. Internalization via Roots A number of studies have explored the possibility of enteric bacteria colonizing and being internalized in plants such as lettuce, tomato, spinach, and sprouts (Guo and

154

Section II. Preharvest Strategies

others 2002; Itoh and others 1998). Current studies suggest that the enteric pathogens originating from animal manure fertilizer gain entry into the growing plants through the root systems and then migrate throughout the edible portion of the plants (Solomon and others 2006). The successful internalization of human pathogens depends on a number of factors such as strain specificity, mobility of microorganisms, emergence of lateral roots, indigenous microorganisms, temperature, humidity, plant variety and defense mechanisms, and presence of root exudation (Cooley and others 2003, 2006; Solomon and others 2002; Warriner and others 2003; Klerks and others 2007). Both E. coli O157 and Salmonella have been reported to gain entry to the plant under laboratory conditions using either hydroponic or soil models systems. Under confocal laser scanning microscope, Solomon and others (2002) observed the internalization of E. coli O157:H7 inside lettuce, 45 μm from the outer surface. The specific interactions between S. enterica and the lettuce cultivar Tamburo are triggered by the root exudates of lettuce, which is correlated for the degree of colonization but not prevalence (Klerks and others 2007). However, two separate field studies reported that both E. coli O157:H7 and Listeria innocua failed to be internalized in the crisphead lettuce and parsley, respectively (Johannessen and others 2005; Girardin and others 2005). Several other field studies on the persistence of pathogen on produce did not differentiate the sites of pathogen contamination (Islam and others 2004a–c, 2005; Ingram and others 2004, 2005). Therefore, further study is needed to understand the mechanisms for enteric pathogens interacting with plant roots and to assess whether pathogen internalization occurs under field conditions. This information is extremely useful to develop intervention strategies for preventing produce contamination during plant growth. Impact of Wildlife on Farms Wildlife, such as birds, rodents, pigs, and deer, can be a vector for transmitting fecal pathogens to produce fields without the grower ’s knowledge and, at various times during the growing seasons, serving as sources of pathogen contamination of fresh vegetables (Nielsen and others 2004). Wild pig fecal droppings were found near the spinach-growing field linked to the 2006–2007 E. coli O157:H7 outbreak in California (CalFERT 2007). Detection of low levels of indigenous E. coli for the extended period of time after manure application to soil might be due to birds or mammal activity in the field (Ingham and others 2004). Organic Production Practices Organic produce may be more susceptible to fecal contamination due to the use of animal manure as organic fertilizer. According to the guidelines set by the USDA National Organic Program (NOP), raw animal manure must either be composted, applied to land used for crops not intended for human consumption, or incorporated into the soil at least 90 days for edible crops not in contact with soil, and 120 days prior to the harvest of edible crops in contact with soil (NOSB 2002). There were a few studies conducted to validate these rules scientifically. Although a greenhouse study suggested that applying raw manure to soil in early spring and late fall for at least 120 days prior to vegetable harvest should ensure that the vegetables would be free of pathogen contamination (Natvig and others 2002), two field studies by Ingham

The Role of Manure and Compost

155

and others (2004, 2005) concluded that the NOP 120-day minimum for manure application to harvest interval cannot guarantee the absence of E. coli on vegetables such as carrots and lettuce grown in Wisconsin. Their results also emphasized that the fertilization-to-planting interval is a factor affecting produce contamination more important than the fertilization-to-harvest interval because seedling emergence is more susceptible to pathogen internalization.

Control of Pathogens in Animal Manure by Composting Animal manure, if not properly contained or treated, can be a major source of waterborne and foodborne illnesses. Therefore, the question of how to eliminate the persistence of pathogens in animal manure or manure-amended soil should be a major concern for animal waste management practices. Animal production establishments may handle the animal wastes in different ways, depending on the types of manure, farm facilities, and physical condition of the farm. Treatment of animal wastes generally falls into three categories: physical, chemical, and biological (CAST 1996; Sobsey and others 2001). Types of waste treatment facilities include lagoons, composters, solid separators, settling basins, and vegetative treatment systems (Meyer and others 1997; Berry and others 2007). The following section focuses on the process of composting to inactivate pathogens in animal manure. Composting Composting is frequently done to convert agricultural wastes into organic fertilizers, soil amendments, or growth medium for greenhouse plants (Rynk 1992). Compost amendments reduce the bulk density of soil, improve soil tilth, facilitate more vigorous root growth, and suppress plant diseases by directly or indirectly affecting plant pathogens or host capacity for growth (Boulter and others 2002). Composting is a managed process in which manure and other organic materials are digested aerobically and anaerobically by microbial action. The composting process generally goes through four phases: mesophilic, thermophilic, cool down, and maturing. At each phase, microbial populations vary significantly as affected by changing environmental conditions. The heat (50 ° to 70 °C) generated in the process can kill many microorganisms (including human and plant pathogens), weed seeds, and fly larvae. Additionally, composting has been explored as a potentially safe process for preventing horizontal resistant gene transfer by inactivating genetically modified organisms or degrading antibiotics in animal wastes (Singh and others 2006; Arikan and others 2007). According to a US SAC report (1998), there are no national standards for composting animal manure. USDA prescribes to organic growers that composting operations maintain a temperature in the range of 55–70 °C for a minimum of 3 days for static aerated pile or in-vessel systems and 15 days with 5 turns for windrow systems, provided the carbon-to-nitrogen ratio (C : N) is in the range of 25 : 1 to 40 : 1 (NOSB 2002). However, there is limited information regarding the fate of foodborne pathogens during the composting process in compost, or scientifically based data regarding the optimal conditions for composting manure to kill pathogens on the farm.

156

Section II. Preharvest Strategies

The most impactful mechanism of pathogen inactivation during composting is through temperature elevation caused by metabolically active microorganisms, although ammonia gas (Himathongkham and others 1999; Nicholson and others 2005), desiccation (Redlinger and others 2001), and microbial antagonism (Ichida and others 2001) are additional factors that contribute to pathogen reduction during composting. Table 8.2 summarizes some recent studies on pathogen inactivation during composting. Several studies have revealed that composting can effectively inactivate E. coli O157 or S. Enteritidis in manure ranging from 3 to less than 14 days when the initial pathogen populations were ca. 107 CFU/g, provided the composting temperatures were maintained for at least 45 °C (Lung and others 2001; Jiang and others 2003a; Hess and others 2004). D-values for E. coli O157:H7 in dairy compost at typical composting temperatures of 50, 55, and 60 °C were 135, 35.4, and 3.9 min, respectively (Jiang and others 2003b). Therefore, composting temperatures of 50, 55, and 60 °C for at least 14 h, 4 h, and 24 min, respectively, should be sufficient to provide a 6-D reduction of E. coli O157:H7 during composting. Composting is an outdoor process for which there are many variables affecting the fate of pathogens. Field studies under different environmental conditions should be used to validate the results generated in laboratories. Recently, in three separate field composting studies, E. coli O157:H7, non-O157 STEC, Salmonella, and Listeria survived for from 8 to less than 14, 9–20, 4, and 4 days, respectively, in dairy manure compost (Nicholson and others 2005; Fremaux and others 2007a; Shepherd and others 2007). Both Fremaux and others (2007a) and Shepherd and others (2007) have shown the uneven pathogen inactivation throughout the compost heaps with longer survival on the surface or around the periphery of heaps due to lower temperatures there. The results from these studies indicate that the enteric pathogens at a population of 107 CFU/g or less inside the active composting heap should be inactivated rapidly within 3 weeks during composting in field. However, many variables, such as the types of animal manure, C : N ratios, pH, moisture content of compost mix, degradation extent of agricultural wastes, digestibility, size of heaps, ambient temperature, frequency of turning, initial cell numbers of target pathogens, strain variation, season of the year, and geographical locations can affect the exact length of time required for pathogen inactivation in compost. Identifying Improper Composting Practices That Allow the Extended Survival of Pathogens Composting techniques may range from “passive,” with little or no input, to a windrow system that is turned and watered routinely with expensive, specialized equipment. Large-scale, properly managed, on-farm composting has a proven track record of consistently producing high-quality compost that went through an adequate thermophilic phase during the composting process. However, for small farmers and organic growers, some composting practices are less effective than others, in part because compost piles are infrequently turned and the C : N ratio, moisture content, or pH of compost materials are inadequate for optimal microbial activity within the compost heaps (Granberry and others 2003). Such practices may cause slow heat-up of the compost pile, which can enable low populations of pathogens to become acclimatized

Table 8.2. Recent studies on the inactivation of foodborne pathogens during composting of animal manure Composting Trial

Pathogens

Under controlled environment Cow manure and dried sawdust (Forced air)

E. coli O157:H7

7

S. Enteritidis

7

E. coli O157:H7

7

Cow manure, wheat straw, cottonseed meal (forced air) In field Ovine and bovine manure (turned and unturned) Cattle, sheep or pig wastes mixed with bedding (turned and unturned) Dairy cattle and pig farmyard manure, broiler litter (turned and unturned) Dairy manure, sawdust, feed, bedding (turned) Cow manure (turned and unturned)

Initial Count (log CFU/g)

Temp. Range (°C)

Length of Survival (Day)

References

45–48 22–33 45–48 22–33 50–69

<3 No change in counts <2 No change in counts >7 but <14

Lung and others 2001

21–68

>36 >47 in aerated bovine manure heap >120 in aerated sheep manure heap >630 unaerated sheep manure heap 93 in beef cattle compost 93 in beef cattle and sheep compost 93 in beef cattle compost 62 in beef cattle compost <4 <8 <4 <4 >14 but <21 (center of heap) >5 but <7 (center of heap) >120 on the surface of two trials >42 in turned heaps >90 in unturned heap D-values of 0.48 and 2.39 days in the center and the base, respectively

E. coli O15:H7

ca. 3–6

Not recorded

Salmonella E. coli O15:H7 Listeria Campylobacter Salmonella E. coli O15:H7 Listeria Campylobacter E. coli O157:H7

6 6 6 6 3.2–4.5 2.7–5.2 2.2–4.9 2.1–4.2 7 5

15–>50 15–>50 15–>50 15–>50 30–>55 30–>55 30–>55 30–>55 25–65 12–62

Shiga-toxin producing E. coli

5.45–6.81

20–68

Jiang and others 2003a

Kudva and others 1998 Huchison and others 2005

Nicholson and others 2005

Shepherd and others 2007 Fremaux and others 2007a

157

158

Section II. Preharvest Strategies

to heat and substantially increase before significant inactivation temperatures are reached. Under low-key composting conditions such as composting in a small heap or with a low C:N ratio, the transition from the mesophilic to thermophilic phase of composting may be very slow or not occur at all, especially during winter months. Using a bioreactor, Lung and others (2001) and Jiang and others (2003a) have demonstrated that the mesophilic composting either caused the initial growth of pathogen or maintained the pathogen populations for at least 4 days. Hutchison and others (2005) reported the extended survival of E. coli O157 in heaped mixtures of dairy and beef cattle manure for 32 and 93 days, respectively, probably due to a longer mesophilic phase occurring in the field. Droffner and Yamamoto (1991) isolated thermal-resistant E. coli, S. Typhimurium, and Pseudomonas mutants from compost samples, and those mutants were capable of growth at 54 °C. Both E. coli and S. Typhimurium mutants can survive for at least 56 and 44 days, respectively, in an outdoor industrial compost and for at least 9 days in laboratory compost at ca. 60 °C when ca. 107 CFU/g were inoculated into food waste and municipal biosolids (Brinton and Droffner 1994). Most composting studies so far used nonstressed cultures, not representative of the microorganisms in complex composting systems, for pathogen challenge studies (Lung and others 2001; Jiang and others 2003a; Huchinson and others 2005; Nicholson and others 2005). Therefore, future studies should examine pathogen survival during composting using a range of conditions typically encountered during the composting. Other improper composting practices may permit the extended survival of pathogens, including adding fresh manure to the compost that has gone through the thermophilic cycle, lack of temperature and moisture monitoring, less turnings of heaps, changes in composting raw materials, manure deepstacking, cross-contamination of equipment handling raw materials and finished compost, and the presence of wildlife around the farm (FDA 2001; Rangurajan and others 2002). Therefore, risks associated with the practices above should be assessed scientifically, which can help to develop practical composting practices being easily adopted by small farmers to produce compost of acceptable quality free of viable human pathogens. Pathogen Regrowth Even though proper composting can effectively inactivate most enteric pathogens, due to the heterogeneity of composting heaps, cold spots inside the heaps, and surface contamination by wildlife and the environment, the cured or finished compost may allow a few survived or newly introduced pathogenic cells to regrow to higher numbers under favorable environmental conditions. Studies have shown that pathogen repopulation (growth) in compost is affected by moisture level, temperature, and nutrient content of the composted solids (Soares and others 1995; Hay 1996; Tiquia 2005). Studies on the microbiological safety of biosolid composting revealed that Salmonella and E. coli can survive the composting process and regrow in compost and stored biosolids when held under favorable conditions (Russ and Yanko 1988; Sidhu and others 1999). E. coli can regrow to substantial levels in compost samples, which were initially very dry, ca. 19% moisture content (Soares and others 1995). To

The Role of Manure and Compost

159

ensure the microbiological safety of composted manure, environmental factors supporting pathogen regrowth need to be identified and controlled.

Education on Safe Use of Raw Manure and Compost Because most fruits and vegetables grow in an open environment, sources for contamination can be multiple and variable. Complete elimination of enteric pathogens in fresh produce growing on a farm is not practical. However, reducing the risks of produce contamination with pathogens can be achieved by a systematic prevention approach (Heaton and Jones 2008). The FDA developed Good Agricultural Practices (GAPs) to provide guidelines and recommendations for fruit and vegetable growers to minimize microbial contamination by addressing various agricultural activities involved in the preplanting, planting, growing, harvesting and postharvest handling of fresh produce (FDA 2001). As illustrated by Bihn and Gravani (2006), produce safety assurance can be built on a strong foundation of GAPs with emphasis on microbial water quality, manure use and composting, worker health and hygiene, and other areas. Guidelines have been developed for farmers on how to treat and use animal waste properly. According to USDA National Organic Program (NOP) standards, raw manure must either be composted or applied to produce fields at least 120 and 90 days before harvesting the produce where edible portions do and do not have contact with the surface soil, respectively (USDA AMS 2000). The compost of plant or animal materials must be produced through a process with a temperature of between 55 and 77 °C for a minimum of 3 to 15 days depending on types of composting system used (NOSB 2002). Additionally, several field studies have suggested that the time between manure/compost application to fields and produce harvesting should be maximized, and that properly composted manure, instead of raw manure, should be applied as fertilizer (Ingram and others 2005; Islam and others 2004a–c). Clearly, proper application of raw manure to produce fields, appropriate waiting periods, as well as effective composting are essential to the successful implementation of GAPs. Although they are not mandatory regulations, GAPs should be implemented by produce growers to reduce the contamination of their products (Bihn and Gravani 2006). It is estimated that in 1997 in the United States, nearly a third of “mixed vegetable” (mostly vegetables) and herb crops were grown organically on a small farm or parcel (Dimitri and Greene 2002). A survey of New York fruit and vegetable growers revealed the lack of food safety concepts on waste management practices for manure and compost application on the farm among small farmers (Rangurajan and others 2002). Therefore, food safety education programs, especially for small growers, should focus on proper composting strategies to reduce pathogen numbers in animal waste and on using compost safely for fruit and vegetable production.

Conclusion The use of raw or improperly composted manure has been identified as an either direct or indirect source of many foodborne disease outbreaks due to consumption of fresh produce contaminated with human pathogens. As a common agricultural practice,

160

Section II. Preharvest Strategies

incorporation of animal manure into agricultural land can introduce the enteric pathogens into the soil. Many studies have demonstrated that manure-borne enteric pathogens can survive for an extended period of time in manure-amended soil, and subsequently contaminate fresh produce growing in the field. Due to the slow decline of pathogen populations in manure-amended soil, reducing initial bacterial load is a key factor to reducing the length of pathogen persistence in soil. Studies have demonstrated that composting can be a safe and effective process to inactivate enteric pathogens in manure if it is properly operated. However, further research is needed to address the safety of those compost products produced under suboptimal conditions or using improper composting practices. Furthermore, a few pathogenic cells in the finished compost, either surviving composting or newly introduced from the environment, have the potential to regrow to high populations when the conditions are conducive for growth. Therefore, it is important to identify the possible environmental factors that could contribute to the extended survival or regrowth of foodborne pathogens in compost. In addition, the use of manure in the production of produce should be carefully managed by following GAP and NOP guidelines. By eliminating pathogens in manure and compost, the likelihood of contamination of vegetables, fruits, water, and the environment by human pathogens will be diminished and thereby enhance the safety of fresh produce production in preharvest environment.

References Ackers ML, Mahon BE, Leahy E, Goode B, Damrow T, Hayes PS, Bibb WF, Rice DH, Barrett TJ, Hutwagner L, Griffin PM, Slutsker L. 1998. An outbreak of Escherichia coli O157:H7 infections associated with leaf lettuce consumption. J Infect Dis 177:1588–93. Arikan OA, Sikora LJ, Mulbry W, Khan SU, Foster GD. 2007. Composting rapidly reduced levels of extractable oxytetracycline in manure from therapeutically treated beef calves. Biores Tech 98:169–76. Avery LM, Hill P, Killham K, Jones DL. 2004. Escherichia coli O157 survival following the surface and sub-surface application of human pathogen contaminated organic waste to soil. Soil Biol Biochem 36:2101–03. Avery LM, Killham K, Jones DL. 2005. Survival of Escherichia coli O157:H7 in organic wastes destined for land application. J Appl Microbiol 98(4):814–22. Berry ED, Woodbury BL, Nienaber JA, Eigenberg RA, Thurston JA, Wells JE. 2007. Incidence and persistence of zoonotic bacterial and protozoan pathogens in a beef cattle feedlot runoff control-vegetative treatment system. J Environ Qual 36(6):1873–82. Besser RE, Lett SM, Weber JT, Doyle MP, Barrett TJ, Wells G, Griffin PM. 1993. An outbreak of diarrhea and hemolytic uremic syndrome from Escherichia coli O157:H7 in fresh-pressed apple cider. J Am Med Assoc 269:2217–20. Beuchat LR. 2006. Vectors and conditions for preharvest contamination of fruits and vegetables with pathogens capable of causing enteric diseases. Br Food J 108(1):38–53. Beuchat LR, Ryu JH. 1997. Produce handling and processing practices. Emerg Infect Dis 3(4):459–465. Bihn EA, Gravani RB. 2006. Role of good agricultural practices in fruit and vegetable safety. In: Matthews KR, editor. Microbiology of Fresh Produce. Washington D.C.: ASM Press. Boulter JI, Boland GJ, Trevors JT. 2002. Assessment of compost for suppression of fusarium patch (Microdochium nivale) and typhula blight (Typhula ishikariensis) snow molds of turfgrass. Biol Control 25:162–72. Brandl MT. 2006. Fitness of human enteric pathogens on plants and implications for food safety. Annu Rev Phytopathol 44:367–92. Breuer T, Benkel DH, Shapiro RL, Hall WH, Winnett MM, Linn MJ, Neimann J, Barrett TJ, Dietrich S, Downes FP, Toney DM, Pearson JL, Rolka H, Slutsker L, Griffin PM. 2001. A multistate outbreak of

The Role of Manure and Compost

161

Escherichia coli O157:H7 infections linked to alfalfa sprouts grown from contaminated seeds. Emerg Infect Dis 7:977–82. Brimecombe MJ, De Leij FAAM, Lynch JM. 2007. Rhizodeposition and microbial populations. In: Pinton R, Varanini Z, Nannipieri P, editors. The Rhizosphere: Biochemistry and Organic Substances at the Soil-Plant Interface. Boca Raton, FL: CRC Press. pp. 73–109. Brinton WF, Droffner MW. 1994. Microbial approaches to characterization of composting processes. Comp Sci & Util 94(summer):12–17. Buck JW, Walcott RR, Beuchat LR. 2003. Recent trends in microbiological safety of fruits and vegetables. Online. Plant Health Progress 10.1094/PHP-2003-0121-01-RV. Bulluck LR, Brosius M, Evanylo GK, Ristaino JB. 2002. Organic and synthetic fertility amendments influence soil microbial, physical and chemical properties on organic and conventional farms. Appl Soil Ecol 19(2):147–60. CalFERT (California Food Emergency Response Team). 2007. Investigation of an Escherichia coli O157:H7 outbreak associated with Dole pre-packaged spinach. Final Report, 21 March 2007. CAST (Council for Agricultural Sciences and Technology). 1996. Integrated animal waste management, Task Force Report, No. 128. CAST, Ames, IA. CDC (Centers for Disease Control and Prevention. 1996. Outbreak of Escherichia coli O157:H7 infections associated with drinking unpasteurized commercial apple juice—British Columbia, California, Colorado and Washington, October 1996. Morb Mortal Wkly Rep 45:945. ———. 2002. Multistate outbreaks of Salmonella serotype Poona infections associated with eating cantaloupe from Mexico—United States and Canada, 2000–2002. Morb Mortal Wkly Rep 51:1044–7. ———. 2006. Ongoing multistate outbreak of Escherichia coli serotype O157:H7 infections associated with consumption of fresh spinach—United States, September 2006. Morb Mortal Wkly Rep 38:1045–6. ———. 2007. Multistate outbreaks of Salmonella infections associated with raw tomatoes eaten in restaurants—United States, 2005–2006. Morb Mortal Wkly Rep 56:909–11. Chapman PA, Siddons CA, Manning J, Cheetham C. 1997. An outbreak of infection due to verocytotoxinproducing Escherichia coli O157 in four families: the influence of laboratory methods on the outcome of the investigation. Epidemiol Infect 119:245–50. Cieslak PR, Barrett TJ, Griffin PM, Gensheimer KF, Beckett G, Buffington J, Smith MG. 1993. Escherichia coli O157:H7 infection from a manured garden. Lancet 342:367–71. Cody SH, Glynn K, Farrar JA, Cairns LC, Griffin PM, Kobayashi J, Fyfe M, Hoffman R, King AS, Lewis JH, Swaminathan B, Bryant RG, Vugia DJ. 1999. An outbreak of Escherichia coli O157:H7 infection from unpasteurized commercial apple juice. Ann Intern Med 130:202–9. Cook KA, Dobbs TE, Hlady WG, Wells JG, Barrett TJ, Puhr ND, Lancette GA, Bodager DW, Toth BL, Genese CA, Highsmith AK, Pilot KE, Finelli L, Swerdlow DL. 1998. Outbreak of Salmonella serotype Hartford infections associated with unpasteurized orange juice. J Am Med Assoc 280:1504–9. Cooley MB, Chao D, Mandrell RE. 2006. Escherichia coli O157:H7 survival and growth on lettuce is altered by the presence of epiphytic bacteria. J Food Prot 69(10):2329–35. Cooley MB, Miller WG, Mandrell RE. 2003. Colonization of Arabidopsis thaliana with Salmonella enterica and enterohemorrhagic Escherichia coli O146:H7 and competition by Enterobacter asburiae. Appl Environ Microbiol 69:4915–26. DeWaal CS, Bhuiya F. 2007. Outbreaks by the numbers: fruits and vegetables 1990–2005. International Association for Food Protection Poster Presentation P3-03, July 8–11, Orlando, Florida. Posted at http:// www.cspinet.org/foodsafety/IAFPPoster.pdf. Accessed on March 1, 2008. Diez-Gonzalez T, Calloway TR, Kizoulis MG, Russell JB. 1998. Grain feeding and the dissemination of acid-resistant Escherichia coli from cattle. Science 281:1666–8. Dimitri C, Greene C. 2002. Recent growth patterns in the U.S. organic foods market. U.S. Department of Agriculture, Economic Research Services, Market and Trade Economics Division and Resource Economics Division. Agriculture Information Bulletin Number 777. Dowe MJ, Jackson ED, Mori JG, Bell CR. 1997. Listeria monocytogenes survival in soil and incidence in agricultural soils. J Food Prot 60(10):1201–7. Doyle MP, Erickson M. 2008. Summer meeting 2007—the problems with fresh produce: an overview. J Appl Microbiol (online early article at http://www.blackwell-synergy.com/doi/abs/10.1111/ j.1365-2672.2008.03746.x. Accessed on March 2, 2008.

162

Section II. Preharvest Strategies

Droffner ML, Yamamoto N. 1991. Prolonged environmental stress via a two step process selects mutants of Escherichia, Salmonella and Pseudomonas that grow at 54 °C. Arch Microbiol 156:307–11. Elder RO, Keen JE, Siragusa GR, Barkocy-Gallagher GA, Koohmaraie M, Laegreid WW. 2000. Correlation of enterohemorrhagic Escherichia coli O157:H7 prevalence in feces, hides and carcasses of beef cattle during processing. Proc Nat Acad Sci 97:2999–3003. Farber JM. 2000. Qualitative risk assessment unpasteurized fruit juice/cider. Ottawa, Ontario: Health Canada, Food Directorate, Health Products and Food Branch, p. 27. FDA (Food and Drug Administration). 2001. Chapter II: Production practices as risk factors in microbial food safety of fresh and fresh-cut produce. In: Analysis and evaluation of preventive control measures for the control and reduction/elimination of microbial hazards on fresh and fresh-cut produce. Posted on http://www.cfsan.fda.gov/∼comm/ift3-2a.html. Accessed in January 2008. Fenlon DR, Ogden ID, Vinten A, Svoboda I. 2000. The fate of Escherichia coli and E. coli O157 in cattle slurry after application to land. J Appl Microbiol 88(Suppl.):149S–156S. Fett WF. 2000. Naturally occurring biofilms on alfalfa and other types of sprouts. J Food Prot 63:625–32. Food and Drug Administration (FDA). 2006. Update: E. coli O157:H7 outbreak at Taco Bell restaurants likely over. FDA traceback investigation continues. http://www.fda.gov/bbs/topics/NEWS/2006/ NEW01527.html. Accessed on March 2, 2008. Franz E, van Diepeningen AD, de Vos OJ, van Bruggen AHC. 2005. Effects of cattle feeding regimen and soil management type on the fate of Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium in manure, manure-amended soil, and lettuce. Appl Environ Microbiol 71:6165–74. Fremaux B, Delignette-Muller ML, Prigent-Combaret C, Gleizal A, Vernozy-Rozand C. 2007a. Growth and survival of non-O157:H7 Shiga-toxin-producing Escherichia coli in cow manure. J Appl Microbiol 102:89–99. Fremaux B, Prigent-Combaret C, Delignette-Muller ML, Dothal M, Vernozy-Rozand C. 2007b. Persistence of shiga toxin-producing Escherichia coli O26 in cow slurry. Lett Appl Microbiol 45(1):55–61. Fukushima H, Hoshina K, Gomyoda M. 1999. Long-term survival of shiga toxin-producing Escherichia coli O26, O111, and O157 in bovine feces. Appl Environ Microbiol 65(11):5177–81. Gagliardi JV, Karns JS. 2000. Leaching of Escherichia coli O157:H7 in diverse soils under various agricultural management practices. Appl Environ Microbiol 66(3):877–83. ———. 2002. Persistence of Escherichia coli O157:H7 in soil and on plant roots. Environ Microbiol 4:89–96. Girardin H, Morris CE, Albagnac C, Dreux N, Glaux C, Nguyen-the C. 2005. Behaviour of the pathogen surrogates listeria innocua and Clostridium sporogenes during production of parsley in fields fertilized with contaminated amendments. FEMS Microbiol Ecol 54(2):287–95. Granberry DM, Germishuizen PJ, Diaz JC, Das KC. 2003. Passive composting fails food safety test. Georgia Vegetable Extension-Research Report, The University of Georgia CAES, Cooperative ResearchExtension Publication No. 5-2003. Guo X, van Iersel MW, Chen J, Brackett RE, Beuchat LR. 2002. Evidence of association of salmonellae with tomato plants grown hydroponically in inoculated nutrient solution. Appl Environ Microbiol 68:3639–43. Hancock DD, Besser TE, Kinsel ML, Tarr PI, Rice DH, Paros MG. 1994. The prevalence of Escherichia coli O157:H7 in dairy and beef-cattle in Washington-state. Epidemiol Infect 113(2):199–207. Hay JC. 1996. Pathogen destruction and biosolids composting. BioCycle 37(6):67–76. Heaton JC, Jones K. 2008. Microbial contamination of fruit and vegetables and the behavior of enteropathogens in the phyllosphere: a review. J Applied Microbiol 104(3):613–26. Hess TF, Grdzelishvili I, Sheng HQ, Hovde CJ. 2004. Heat inactivation of Escherichia coli during manure composting. Compost Sci Util 12(4):314–22. Hilborn ED, Mermin JH, Mshar PA, Hadler JL, Voetsch A, Wojtkunski C, Swartz M, Mshar R, LambertFair MA, Farrar JA, and others. 1999. A multistate outbreak of Escherichia coli 0157:H7 infections associated with consumption of mesclun lettuce. J Am Med Assoc 159:1758–64. Himathongkham S, Bahari S, Riemann H, Cliver D. 1999. Survival of Escherichia coli O157:H7 and Salmonella Typhimurium in cow manure and cow manure slurry. FEMS Microbiol Lett 178:251–7.

The Role of Manure and Compost

163

Himathongkham S, Riemann H. 1999. Destruction of Salmonella Typhimurium, Escherichia coli 0157:H7 and Listeria monocytogenes in chicken manure by drying and/or gassing with ammonia. FEMS Microbiol Letts 171(2):179–82. Holley RA, Arrus KM, Ominski KH, Tenuta M, Blank G. 2006. Salmonella survival in manure-treated soils during simulated seasonal temperature exposure. J Environ Qual 35:1170–80. Hutchison ML, Walters LD, Avery SM, Moore A. 2005. Decline of zoonotic agents in livestock waste and bedding heaps. J Appl Microbiol 99(2):354–62. Hutchison ML, Walters LD, Moore A, Crookes KM, Avery SM. 2004. Effect of length of time before incorporation on survival of pathogenic bacteria present in livestock wastes applied to agricultural soil. Appl Environ Microbiol 70(9):5111–8. Ichida JM, Krizova L, LeFevre CA, Keener HM, Elwell DL, Burtt EH. 2001. Bacterial inoculum enhances keratin degradation and biofilm formation in poultry compost. J Microbiol Meth 47:199–208. Ingham SC, Fanslau MA, Engel RA, Breuer JR, Breuer JE, Wright TH, Reith-Rozelle JK, Zhu J. 2005. Evaluation of fertilization-to-planting and fertilization-to-harvest intervals for safe use of noncomposted bovine manure in Wisconsin vegetable production. J Food Prot 68:1134–42. Ingham SC, Losinski JA, Andrews MP, Breuer JE, Breuer JR, Wood TM, Wright TH. 2004. Escherichia coli contamination of vegetables grown in soils fertilized with noncomposted bovine manure: gardenscale studies. Appl Environ Microbiol 70(11):6420–7. Islam M, Doyle MP, Phatak SC, Millner P, Jiang XP. 2004a. Persistence of enterohemorrhagic Escherichia coli O157: H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. J Food Prot 67(7):1365–70. Islam M, Doyle MP, Phatak SC, Millner P, Jiang XP. 2005. Survival of Escherichia coli O157:H7 in soil and on carrots and onions grown in fields treated with contaminated manure composts or irrigation water. Food Microbiol 22(1):63–70. Islam M, Morgan J, Doyle MP, Jiang XP. 2004d. Fate of Escherichia coli O157:H7 in manure compostamended soil and on carrots and onions grown in an environmentally controlled growth chamber. J Food Prot 67:574–578. Islam M, Morgan J, Doyle MP, Phatak SC, Millner P, Jiang XP. 2004b. Fate of Salmonella enterica serovar Typhimurium on carrots and radishes grown in fields treated with contaminated manure composts or irrigation water. Appl Environ Microbiol 70:2497–502. ———. 2004c. Persistence of Salmonella enterica serovar Typhimurium on lettuce and parsley and in soils on which they were grown in fields treated with contaminated manure composts or irrigation water. Foodborne Pathogens Dis 1:27–35. Itoh Y, Sugita-Konishi Y, Kasuga F, Iwaki M, Hara-Kudo Y, Saito N, Noguchi Y, Konuma H, Kumagai S. 1998. Enterohemorrhagic Escherichia coli O157:H7 present in radish sprouts. Appl Environ Microbiol 64(4):1532–5. Jacob ME, Fox JT, Drouillard JS, Renter DG, Nagaraja TG. 2008. Effects of dried distillers’ grain on fecal prevalence and growth of Escherichia coli O157 in batch culture fermentations from cattle. Appl Environ Microbiol 74(1):38–43. Jalava K, Hakkinen M, Valkonen M, Nakari U-M, Palo T, Hallanvuo S, Ollgren, Siitonen A, Pekka Nuorti J. 2006. An outbreak of gastrointestinal illness and erythema nodosum from grated carrots contaminated with Yersinia pseudotuberculosis. J Infect Dis 194:1209–16. Jiang XP, Islam M, Morgan J, Doyle MP. 2004. Fate of Listeria monocytogenes in bovine manure-amended soil. J Food Prot 67(8):1676–81. Jiang XP, Morgan J, Doyle MP. 2002. Fate of Escherichia coli O157:H7 in manure-amended soil. Appl Environ Microbiol 68(5):2605–9. ———. 2003a. Fate of Escherichia coli O157:H7 during composting in a laboratory-scale bioreactor. J Food Prot 66:25–30. ———. 2003b. Thermal inactivation of Escherichia coli O157:H7 in cow manure compost. J Food Prot 66(10):1771–7. Johannessen GS, Bengtsson GB, Heier BT, Bredholt S, Wasteson Y, Rørvik LM. 2005. Potential uptake of Escherichia coli O157:H7 from organic manure into crisphead lettuce. Appl Environ Microbiol 71(5):2221–5.

164

Section II. Preharvest Strategies

Johannessen GS, Frøseth RB, Solemdal L, Jarp J, Wasteson Y, Rørvik LM. 2004. Influence of bovine manure as fertilizer on the bacteriological quality of organic Iceberg lettuce. J Appl Microbiol 96:787–794. Johnston LM, Moe CL, Moll D. Jaykus L. 2006. The epidemiology of produce-associated outbreaks of foodborne disease. In: James J, editor. Microbial Hazard Identification in Fresh Fruits and Vegetables. Hoboken, NJ: John Wiley & Sons, Inc. pp. 37–72. Jones DL. 1999. Potential health risks associated with the persistence of Escherichia coli O157 in agricultural environments. Soil Use Manage 15(2):76–83. Jones PW. 1986. Sewage sludge as a vector of salmonellosis. In: Block JC, Haielaar AH, L’Hermite P, editors. Epidemiological Studies of Risks Associated with the Agricultural Use of Sewage Sludge. London: Elsevier. pp. 21–33. Keener KM, Bashor MP, Curtis PA, Sheldon BW, Kathariou S. 2004. Comprehensive review of Campylobacter and poultry processing. Comp Rev Food Sci Food Safety 3:105–16. Klerks MM, Franz E, van Gent-Pelzer M, Zijlstra C, van Bruggen AHC. 2007. Differential interaction of Salmonella enterica serovars with lettuce cultivars and plant-microbe factors influencing the colonization efficiency. Isme J 1(7):620–31. Kudva IT, Blanch K, Hovde CJ. 1998. Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Appl Environ Microbiol 64:3166–74. Kudva IT, Hatfield PG, Hovde CJ. 1995. Effect of diet on the shedding of Escherichia coli O157:H7 in a sheep model. Appl Environ Microbiol 61:1363–70. Lau MM, Ingham SC. 2001. Survival of fecal indicator bacteria in bovine manure incorporated into soil. Lett Appl Microbiol 33:131–6. Lazarovits G. 2001. Management of soil-borne plant pathogens with organic soil amendments: a disease control strategy salvaged from the past. Can J Plant Pathol 23:1–7. Lung AJ, Lin C–M, Kim JM, Marshall MR, Nordstedt R, Thompson NP, Wei CI. 2001. Destruction of Escherichia coli O157:H7 and Salmonella Enteritidis in cow manure composting. J Food Prot 64:1309–14. Mahon BE, Ponka A, Hall WN, Komatsu K, Dietrich SE, Siitonen A, Cage G, Hayes PS, Lambert-Fair MA, Bean NH, Griffin PM, Slutsker L. 1997. An international outbreak of Salmonella infections caused by alfalfa sprouts grown from contaminated seeds. J Infect Dis 175:876–82. Maule A. 1999. Environmental aspects of E. coli O157:H7. Intern Food Hygiene 9:21–3. Meng J, Doyle MP, Zhao T, Zhao S. 2001. Enterohemorrhagic Escherichia coli. In: Doyle MP, Beuchat LR, Montville TJ, editors. Food Microbiology: Fundamentals and Frontiers, 2nd ed. Washington D.C.: ASM Press. pp. 193–213. Meyer DM, Garnett I, Guthrie JC. 1997. A survey of dairy manure management practices in California. J Dairy Sci 80:1841–45. Millard PS, Gensheimer KF, Addiss DG, Sosin DM, Beckett GA, Houck-Jankoski A, Hudson A. 1994. An outbreak of cryptosporidiosis from fresh-pressed apple cider. J Am Med Assoc 272:1592–96. Mohle-Boetani JC, Farrar JA, Werner SB, Minassian D, Bryant R, Abbott S, Slutsker L, Vugia DJ. 2001. Escherichia coli O157:H7 and Salmonella infections associated with sprouts in California, 1996–1998. Ann Internal Med 135:239–47. Mohle-Boetani JC, Reporter R, Werner SB, Abbott S, Farrar J, Waterman SH, Vugia DJ. 1999. An outbreak of Salmonella serogroup Saphra due to cantaloupes from Mexico. Ann Internal Med 180:1361–4. Morgan GM, Newman C, Palmer SR, Allen JB, Shepherd W, Rampling AM, Warren RE, Gross RJ, Scotland SM, Smith HR. 1988. First recognized community outbreak of haemorrhagic colitis due to verotoxinproducing Escherichia coli O157:H7 in the UK. Epidemiol Infect 101:83–91. Mukherjee A, Cho S, Scheftel J, Jawahir S, Smith K, Diez-Gonzalez F. 2006. Soil survival of Escherichia coli O157:H7 acquired by a child from garden soil recently fertilized with cattle manure. J Appl Microbiol 101(2):429–36. Mukherjee A, Speh D, Dyck E, Diez-Gonzalez F. 2004. Preharvest evaluation of coliforms, Escherichia coli, Salmonella, and Escherichia coli O157:H7 in organic and conventional produce grown by Minnesota farmers. J Food Prot 67:894–900. NACMCF (National Advisory Committee on Microbiological Criteria for Foods). 1999. Microbiological safety evaluations and recommendations on fresh produce. Food Control 10:117–43.

The Role of Manure and Compost

165

National Organic Standards Board). 2002. Compost Task Force Recommendations, as amended by the NOP. http://www.ams.usda.gov/nosb/NOSBrecommendations/Compost.pdf. Accessed on January 8, 2008. Natvig EE, Ingham SC, Ingham BH, Cooperband LR, Roper TR. 2002. Salmonella enterica serovar Typhimurium and Escherichia coli contamination of root and leaf vegetables grown in soils with incorporated bovine manure. Appl Environ Microbiol 68(6):2737–44. Nicholson FA, Groves SJ, Chambers BJ. 2005. Pathogen survival during livestock manure storage and following land application. Bioresour Technol 96(2):135–43. Nielsen EM, Skov MN, Madsen JJ, Lodal J, Jespersen JB, Baggesen DL. 2004. Verocytotoxin-producing Escherichia coli in wild birds and rodents in close proximity to farms. Appl Environ Microbiol 70:6944–7. Nuorti, JP, Niskanen T, Hallanvuo S, Mikkola J, Kela E, Hatakka M, Fredriksson-Ahomaa M, Lyytikäinen, Siitonen A, Korkeala H, Ruutu P. 2004. A widespread outbreak of Yersinia pseudotuberculosis O:3 infection from Iceberg lettuce. J Infect Dis 189:766–74. Ogden ID, Hepburn NF, MacRae M, Strachan NJ, Fenlon DR, Rusbridge SM, Pennington TH. 2002. Longterm survival of Escherichia coli O157 on pasture following an outbreak associated with sheep at a scout camp. Lett Appl Microbiol 34:100–4. Pebody RG, Leino T, Ruutu P, Kinnunen, Davidkin I, Nohynek H, Leinikki P. 1998. Foodborne outbreaks of hepatitis A in a low endemic country: an emerging problem? Epidemiol Infect 120:55–9. Pell AN. 1997. Manure and microbes: Public and animal health problem? J Dairy Sci 80(10):2673–81. Rangurajan A, Pritts MP, Reiners S, Pedersen LH. 2002. Focusing food safety training based on current grower practices and farm scale. Hort Technol 12(1):126–31. Redlinger T, Graham J, Corella-Barud V, Avitia R. 2001. Survival of fecal coliforms in dry-composting toilets. Appl Environ Microbiol 67:4036–40. Russ CF, Yanko WA. 1988. Factors affecting Salmonella population in composted sludges. Appl Environ Microbiol 41:597–602. Rynk R. 1992. On-farm composting handbook. Northeast Regional Agricultural Engineering Service, Ithaca, NY. Schlech WI, Lavigne PM, Bortolussi RA, Allen AC, Haldane EV, Won AJ, Hightower AW, Johnson SE, King SH, Nicholls ES, Broome CV. 1983. Epidemic listeriosis—evidence for transmission by food. N Engl J Med 308:203–6. Shepherd MW, Liang PF, Jiang XP, Doyle MP, Erickson MC. 2007. Fate of Escherichia coli O157:117 during on-farm dairy manure-based composting. J Food Prot 70(12):2708–16. Sidhu J, Gibbs RA, Ho GE, Unkovich I. 1999. Selection of Salmonella Typhimurium as an indicator for pathogen regrowth potential in composted biosolids. Lett Appl Microbiol 29:303–7. Singh A, Billingsley K, Ward O. 2006. Composting: a potentially safe process for disposal of genetically modified organisms. Crt Rev Biotech 26:1–16. Sinton LW, Braithwaite RR, Hall CH, Mackenzie ML. 2007. Survival of indicator and pathogenic bacteria in bovine feces on pasture. Appl Environ Microbiol 73:7917–25. Sivapalasingam S, Friedman CR, Cohen L, Tauxe RV. 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J Food Prot 67:2342–53. Soares HM, Cardenas B, Weir D, Switzenbaum MS. 1995. Evaluating pathogen regrowth in biosolids compost. BioCycle 36(6):70–6. Sobsey MD, Khatib LA, Hill VR, Alocilja E, Pillai S. 2001. Pathogens in animal wastes and the impacts of waste management practices on their survival, transport and fate. In: White Papers on Animal Agriculture and the Environment. MidWest Plan Service (MWPS), Iowa State University, Ames, IA (Chapter 17). Söderström A, Lindberg A, Anderson Y. 2005. EHEC O157 outbreak in Sweden from locally produced lettuce, August–September 2005. Eurosurveillance Weekly. 10:E050922.1. http://www.eurosurveillance. org/ew/2005/050922.asp#1. Solomon EB, Brandl MT, Mandrell RE. 2006. Biology of foodborne pathogens on produce. In: Matthews KR, editor. Microbiology of Fresh Produce. Washington D.C.: ASM Press. pp. 55–83. Solomon EB, Yaron S, Matthews KR. 2002. Transmission of Escherichia coli O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalization. Appl Environ Microbiol 68:397–400.

166

Section II. Preharvest Strategies

Stern NJ, Robach MC. 2003. Enumeration of Campylobacter spp. in broiler feces and in corresponding processes carcasses. J Food Prot 66:1557–63. Swaminathan B. 2001. Listeria monocytogenes. In: Doyle MP, Beuchat LR, Montville TJ, editors. Food Microbiology: Fundamentals and Frontiers, 2nd ed. Washington D.C.: ASM Press. pp. 383–409. Swerdlow DL, Mintz ED, Rodriquez M, Tejada E, Ocampo C, Espejo L, Greene KD, Saldana W, Seminario L, Tauxe RV. 1992. Waterborne transmission of epidemic cholera in Trujillo, Peru: lessons for a continent at risk. Lancet 340:28–33. Tamblyn S, de Grosbois J, Taylor D, Stratton J. 1999. An outbreak of Escherichia coli O157:H7 infection associated with unpasteurized non-commercial, custom-pressed apple cider—Ontario, 1998. Can Commun Dis Rep 25:113–7. Tiquia SM. 2005. Microbiological parameters as indicators of compost maturity. J Appl Microbiol 99:816–28. USDA (United States Department of Agriculture). Agricultural marketing Service (AMS). 2000. National Organic Program. 7 CFR part 205.203. United States Department of Agriculture, Washington, D.C. ———. 2002. Census of agriculture—United States data. Posted at http://www.agcensus.usda.gov/ publications/2002/index.asp. Accessed on March 1, 2008. US SAC (United States Senate Agriculture Committee). 1998. Animal waste pollution in America: An emerging national problem. Washington, D.C.: U.S. Government Printing Office. Van Kessel JS, Pachepsky YA, Shelton DR, Karns JS. 2007. Survival of Escherichia coli in cowpats in pasture and in laboratory conditions. J Appl Microbiol 103:1122–7. Van Renterghem B, Huysman F, Rygole R, Verstraete W. 1991. Detection and prevalence of Listeria monocytogenes in the agricultural ecosystem. J Appl Bacteriol 71:211–7. Wang G, Zhao T, Doyle MP. 1996. Fate of enterohemorrhagic Escherichia coli O157:H7 in bovine feces. Appl Environ Microbiol 62:2567–70. Warriner K, Ibrahim F, Dickinson M, Wright C, Waites WM. 2003. Interaction of Escherichia coli with growing salad spinach plants. J Food Prot 66:1790–7. Wray C, Davies RD. 2003. Epidemiology and ecology of Salmonella in meat-producing animals. In: Torrence, ME, Isaacson, RE, editors. Microbial Food Safety in Animal Agriculture: Current Topics. Ames, Iowa: Iowa State Press. pp. 73–82. You YW, Rankin SC, Aceto HW, Benson CE, Toth JD, Dou ZX. 2006. Survival of Salmonella enterica serovar Newport in manure and manure-amended soils. Appl Environ Microbiol 72(9):5777–83. Zhao T, Doyle MP, Shere J, Garber L. 1995. Prevalence of Escherichia coli O157:H7 in a survey of dairy herds. Appl Environ Microbiol 61:1290–3.

Section III Postharvest Interventions

9 Aqueous Antimicrobial Treatments to Improve Fresh and Fresh-Cut Produce Safety Joy Herdt and Hao Feng

Introduction Most fresh produce is subjected to washing after harvest, and it is usually done by processors using flume transport systems, batch tanks, or water sprays. The washing of fresh produce is an important step for removing soil and debris, improving the appearance of the commodity, lowering the produce’s temperature, and limiting the development of physiological changes. Washing also reduces the microbial load on the surface of incoming produce, which impacts the product’s quality, shelf life, and safety. With increasing concerns about the safety of fresh and fresh-cut produce, the latter benefit of washing is becoming increasingly more important. In most cases, antimicrobials are added to the washing systems to enhance the control of microorganisms that are found on the surface of incoming fresh produce. Water alone will provide the basic objectives of washing; however, it is very common for processors to recycle the wash water for conservation. Reusing wash water increases the risk of spreading contamination, because water can remove pathogens from the surface of produce and then transfer microbial contaminants to other batches of produce during subsequent washing processes. Proper sanitation of the wash water is critical to prevent the transmission of pathogens or other bacteria from the contaminated to uncontaminated produce. This requires diligent monitoring and controlling efforts to ensure that proper levels of antimicrobial agents are maintained in the recycled water. The washing step is critical to remove the level of contamination from the surface of the produce to the maximum extent possible, and the antimicrobial in the water eliminates the microorganism. Antimicrobial treatments vary in their effectiveness based on the chemical, physical, and mechanical action applied during treatment. Parameters such as concentration, time, temperature, pH, washing dynamics, soil load on the produce, buildup of organic material in the wash water, target microorganism, how and where the microorganism is attached to the produce surface, the level of microorganisms on the surface of the produce, and the produce surface topography must be considered when optimizing the washing system. It is the combination of these factors that determines the efficacy of an antimicrobial treatment in reducing microbial populations on the surface of produce. Summarized in this chapter are several antimicrobial agents that have been cleared by the FDA for washing fruits and vegetables. 169

170

Section III. Postharvest Interventions

Factors That Influence Antimicrobial Activity in Produce Washing Washing Conditions Antimicrobial Concentration The efficacy of a produce wash operation is primarily determined by the concentration of the antimicrobial. High concentrations typically result in high antibacterial activity when pH, temperature, and organic content are held constant, but it may also damage produce tissue. The impact of the antimicrobial on the environment at high concentrations is another factor that has to be taken into consideration. Each antimicrobial and produce combination has an optimal concentration with which the quality degradation of the produce will be minimal. Additionally, there are regulatory limitations on the level of chemical agents that are allowed for use on specific food products, and each country has its own regulatory body. In the United States, the Food and Drug Administration (FDA) and the Environmental Protection Agency (EPA) regulate the level of the antimicrobial agents that is allowed for use on fruits and vegetables. It is also important to maintain a consistent effective concentration of antimicrobial during the course of washing. Time In an industrial produce wash, a two-stage strategy is normally used, first to remove debris and soil in stage one, followed by stage two, where reducing the microbial load is the main target. The time used in each stage can range from 30 sec to 1 or 2 min and will be determined by throughput, produce type, harvest location and time, and other pertinent factors. Generally, an increase in washing time will result in an increased removal of bacteria. However, extending washing time is not always effective in a produce wash. In a washing study using a laboratory flow-through wash system, Wang and others (2007) reported a fast E. coli O157:H7 population reduction in the first 2–3 min of washing cantaloupe and fresh-cut apples with aqueous peroxyacetic acid (POAA). Prolonging the washing beyond 3 min, however, did not induce significantly more reduction in the microbial population. This dual-phase microbial inactivation where a fast microbial count reduction in the first few minutes of washing is followed by a slow reduction region was also documented by Wang and others (2006) in a produce wash test with chlorine, electrolyzed water, and POAA. In produce washing operations, washing in the slow reduction stage is ineffective and should be avoided. Temperature An increase in the temperature of an antimicrobial solution generally produces an increase in bactericidal activity. However, to minimize degradation of the produce quality, especially during storage, currently the produce industry employs a lowtemperature wash practice. Normally the entire supply chain, including washing in the plant is maintained at 4 °C. In addition, because the maximum solubility of chlorine in water occurs at about 4 °C, this cooler water temperature is more effective in retaining its antimicrobial activity. However, in order to establish a temperature difference that minimizes product uptake of wash water (which may introduce microorganisms into the produce), ideally the temperature of the wash water should be at least 10 °C higher than that of the fruits or vegetables (Beuchat 1998).

Aqueous Antimicrobial Treatments

171

pH Beside concentration, the pH of a solution can impact the bactericidal effect of certain antimicrobials. An increase in pH substantially decreases the biocidal activity of chlorine; a low pH environment favors the formation of hypochlorous acid (HClO), which increases the bactericidal activity of the wash solution. The same is true for acid sanitizers. The pH of the solution must be at or below the dissociation constant (or pKa) of the acid in order for it to be effective. Other chemical oxidants can be influenced by pH, but not to the same extent as chlorine or acids. Sanitizer Flow Hydrodynamics Generally, high shear rates on produce surfaces during a wash operation increase detachment of attached bacteria (Chang and others 1991). Mechanical forces due to liquid flow enhance the removal of attached bacteria from tissues and food contact surfaces. In addition to shear forces generated by liquid flow and turbulence, other physical methods, such as pressurized water jet and ultrasonic treatment, have also been used to increase the shear rate on produce surfaces (De Zuniga and others 1991). Increasing the turbulence of the wash water is important to ensure adequate contact to produce surfaces and removal of debris and microorganisms. Longley (1978) proposed that bacterial cells form protective clusters during exposure to a sanitizer. Increased turbulence in a washing operation may aid in enhancing the activity of the antimicrobial by breaking up the clusters and increasing the accessibility of the antimicrobial to individual bacterial cells (Wang 2006). Water jets, air bubbling or injection, and gravity flow have all been used by the produce industry as mechanical means to increase turbulence in a wash process. Water Quality Organic Loading Because most of the antimicrobial agents used in washing fruits and vegetables are also oxidizing agents, they often react chemically with organic materials, which causes the depletion of the effective component of the antimicrobial in the wash water (Table 9.1). The organic materials encountered in produce washing may include fruit and vegetable tissue, cellular fluids exuded by cutting, soil particles, insects, and microbes. Because wash water is generally reused, the accumulation of organic matter in the wash water causes a decrease in the effective concentration of the antimicrobial, and thereby contributes to an increased risk of contamination of foodborne pathogens. It is therefore important to understand the effect of organic loading on the antimicrobial

Table 9.1. Oxidation capacity of antimicrobials Oxidizing Agent Ozone Peracetic acid Hydrogen Peroxide Chlorous acid (ASC) Chlorine dioxide Sodium hypochlorite

Electron Volts (eV) 2.07 1.81 1.78 1.58 1.57 1.36

172

Section III. Postharvest Interventions

and to replenish the antimicrobial agent in a timely manner and maintain a constant, effective concentration. Most studies on the effect of organic loading have been conducted with chlorine. The concept of chlorine demand has been intensively studied in the field of water sanitation and poultry processing. The chlorine demand is the difference between the amount of chlorine applied to the washing tank and the amount of free, combined, or total chlorine residual remaining at the end of the washing. The chlorine demand is considered an extrinsic property of the water that is affected by pH, temperature, and dissolved compounds in the water (Tsai and others 1992). When the interaction of chlorine with microbes is considered, the chlorine concentration is assumed to decay exponentially, following a first-order kinetic model (Virto and others 2004). Studies also reported bilinear behavior of chlorine decay, which has been attributed to the existence of two distinct reaction types in the chlorination of microorganisms (Shang and Blatchley 2001). In a test using chlorine to wash sliced romaine lettuce, the rate of depletion of both free chlorine and total chlorine followed a first-order kinetics model at three produce-to-solution ratios (Zhou and others 2008). The effect of produce-to-solution ratio on chlorine decay (ppm/min) was also characterized by a semilog relation. In a produce washing test, Garg and others (1990) reported that total aerobic bacteria counts were not reduced on fresh-cut vegetables such as carrot, red cabbage, and lettuce washed with 300 mg/l chlorine, which was attributed to the rapid depletion of chlorine by increased levels of organic loading. Water pH and Chemistry Because chlorine works most effectively in the pH range 5–7, water supplies that are buffered or are highly alkaline (higher pH values) often need their pH adjusted for chlorine to be effective. With acid sanitizers in buffered water, higher concentrations of acids may be required to reduce the pH to effective levels. Other components of water, such as organic and inorganic compounds, will react with and consume oxidizers. Components of hard water such as inert magnesium and calcium ions generally do not affect most oxidizing chemistries. Microbial Type, Level, and Attachment Vegetative cells are less resistant to chlorine than the spore-forming organisms. E. coli was found to be generally more resistant to chlorine than other vegetative bacteria. Although a bacterium is usually readily killed when directly exposed to a sanitizer, the inactivation becomes much more difficult when the microorganism is attached to produce surfaces. Lund (1983) suggested that contact with host tissue may inactivate hypochlorite and that the adhesion of bacteria to plant tissues generally increases their resistance to sanitizers. The population reduction in cell suspensions of E. coli O157:H7, L. monocytogenes, and Shigella treated with acidic electrolyzed water (AEW) reached over 8.0 log CFU/ml in 1 min (Kim and others 2000). However, less than 2.7 log CFU/g reduction was achieved in a 15-min treatment with AEW when those bacteria were attached to produce surfaces and air-dried for 1 hr (Park and others 2001) or 7 hrs (Stan and Daeschel 2003) after inoculation. Microbial biofilms also could develop on produce surfaces when the right growth conditions are satisfied. The removal of biofilm from produce surfaces remains a major challenge to the food

Aqueous Antimicrobial Treatments

173

industry. In addition, the microorganisms entrapped in pores or stems in the calyx of fruits will also be difficult to remove during a sanitation operation. Produce Surface Properties The chemical, physical, and topographical surface properties of fruits and vegetables play an important role in the interaction between produce and microorganisms (Wang 2006). For example, the cuticular waxes not only influence plant surface wetability, but also alter the way the plant and microorganism interact (Beattie and Marcell 2002). Stomata, lenticels, broken trichomes, and scars on the plants represent natural ways of entry for microorganisms. Many live E. coli O157:H7 cells were found in the stomata and on cut edges of lettuce after treatment with a chlorine solution, or in the inner tissues and stomata of cotyledons of radish sprouts (Seo and Frank 1999; Itoh and others 1998). The ability of water and hypochlorite wash to effectively remove bacteria is limited when the microbes stay in protective hydrophobic pockets and folds on leaf surfaces (Adams and others 1989) or in crevices, pits, and pores (Frank and Koffi 1990; Holah and Thorpe 1990). Han and others (2000) observed that E. coli O157:H7 attached more readily to coarse, porous, and injured surfaces than to uninjured surfaces of green peppers after a chlorine dioxide treatment. Liao and Sapers (2000) suggested that the Salmonella Chester attached to injured tissue and calyx cavities of the apple more easily than to unbroken skin due to their different topographical structures and specific physicochemical properties. Surface topography is an important physical property of fresh fruits and vegetables, influencing not only their visual and sensorial aspects but also microorganism attachment and removal (Liao and Sapers 2000; Han and others 2000). Published work on quantifying the effect of surface topography on bacterial removal from fresh produce is limited, maybe because of the difficulties in the measurement of surface roughness of fresh produce with suitable technologies. Traditional methods used to determine surface roughness cannot be used to measure the soft surface of produce. A new noncontact method was recently developed by Wang and others (2005) with confocal laser scanning microscopy (CLSM). The roughness of Golden Delicious apples, navel oranges, and cantaloupes was measured with the CLSM method; and a positive correlation among roughness, adhesion strength of E. coli O157:H7, and the efficacy of inactivation by POAA was reported.

Aqueous Antimicrobial Treatments Chlorine and Chlorine Compounds As the most widely used antimicrobial for washing fresh fruits and vegetables in the food industry (Nguyen-the and Carlin 1994), chlorine is active against a wide spectrum of microorganisms, including viruses, non–acid-fast vegetative bacteria, acid-fast bacilli, bacterial spores, fungi, algae, and protozoa, with bacterial spores being the most resistant form of microbial life (Trueman 1971). When elemental chlorine or hypochlorites are added to water, they undergo the following reactions (Dychdala 2001): Cl 2 + H 2 O → HClO + H + + Cl −

174

Section III. Postharvest Interventions

Ca ( ClO)2 + H 2 O → Ca 2+ + H 2 O + 2ClO − Ca ( ClO)2 + 2H 2 O → Ca( OH )2 + 2HClO HClO ↔ H + + Claa − The term free available chlorine consists of chlorine gas (Cl2), hypochlorous acid (HClO), or hypochlorite ions (ClO−). Hypochlorous acid is the main active ingredient formed. The dissociation of HClO depends on the pH and equilibrium between HClO and ClO−, which is maintained even when HClO is constantly consumed through its antimicrobial activity (Beuchat 1992). The mode of action of hypochlorous acid destroying microorganisms has not been fully understood. It is thought that HClO allows oxygen to emerge, which in turn supposedly combines with components of cell protoplasm, destroying the organism. Thomas (1979) reported that the lethality of HClO is attributed to its interactions with cell membrane proteins to form nitrogenchlorine (N-Cl) derivatives (chloramines or chloramides), which would interfere with cell metabolism. Beuchat (1992) proposed that because of the low chlorine level required for bactericidal action, chlorine must inhibit some key enzymatic reactions in the cell. The inhibition of essential cytoplasmic metabolic reactions should be largely responsible for the destruction of both bacterial and fungal cells. The typical concentration of chlorine used in a sanitation treatment is 200 mg/l at a pH of <8.0 and with a contact time of 1–2 min for raw fruits and vegetables processing (Beuchat 1996). Brackett (1987) reported that when the free chlorine concentration was <50 mg/l, there was no observed antimicrobial effect on L. monocytogenes (initial concentration of about 108 CFU/g). Exposing the microbes to chlorine at ≥50 mg/l for a contact time ≥20 sec, however, resulted in no recoverable cells. Park and Beuchat (1999) used a chlorine washing solution at concentration of 2000 mg/l and reduced aerobic bacteria by over 2 logs in 3 min on honeydew melons, but washing with chlorine at 200 mg/l reduced the population by only 1 log. Mazollier (1988) observed that a further increase of chlorine concentration above 50–200 mg/l did not reduce the population of the total aerobic count on lettuce more than a concentration of 50 mg/l of free chlorine treatment. The effectiveness of chlorine on produce is dependent on the produce type. For instance, leafy produce with surfaces having a lot of folds can provide better protection to bacteria than relatively smooth-surfaced honeydew melons (Adams and others 1989). For leafy greens, a moderate increase in chlorine concentration or treatment time may not be effective in microbial inactivation. The efficacy of chlorine wash is affected by the pH of the chlorine solution. The time required to achieve a 99% reduction in spore populations was shorter at pH 4–5; the times required to reduce the spore population at higher pH values were longer (Sykes 1965). Adam and others (1989) observed that an adjustment of pH from 9.0 to 4.5 with inorganic or organic acids introduced an additional 1.5- to 4.0-log reduction in the number of microbes on lettuce leaves. Sykes (1965) reported an equivalent activity of a chlorine solution with 50 mg/l at pH 9 to that with free chlorine of ≥100 mg/l at pH 10, indicating that pH is sometimes more important than the chlorine concentration. However, a low pH (4–5) treatment is often corrosive to equipment surface, and a pH of 6.5–7.0 (at which the percentage of HClO is near 97%) is recommended for use in washing fresh produce (Beuchat 1996).

Aqueous Antimicrobial Treatments

175

Chlorine wash is the most widely used industrial washing method in the produce industry, mainly because of its efficacy, availability, relatively low cost, and ease of operation. However, current chlorine-based sanitizers used to wash fresh produce do not provide satisfactory microbial reduction in industrial scale sanitizing treatments. Within the FDA approved concentration, a chlorine wash can only achieve a 1–2 log CFU/g reduction in microbial populations (Hegenbart 2002). The use of chlorine also brings up health and environmental concerns due to the generation of chlorinated by-products with some carcinogenic properties such as chloroform and trihalomethane (THM) during a washing treatment when precursor organic compounds, such as humic and fulvic acids are present (Richardson and others 1998; Menzer and Nelson 2001). Additional limitations of chlorine include its rapid depletion under high organic loading, the requirement for pH adjustment and off-gassing during processing. Therefore, the fresh produce industry constantly seeks alternative sanitizers—which are safe, environmentally friendly, and more effective—to replace chlorine. Electrolyzed Water Electrolyzed water is a technique developed in Japan. It is produced by electrolysis of water containing a low concentration of sodium chloride (0.1%) in an electrolysis chamber where anode and cathode are separated by a bipolar membrane. The water from the anode normally has a pH of <2.6, and an oxidation-reduction potential (ORP) of >1,100 mV, and hypochlorous acid is present. The water produced from the cathode has a pH of >11.4 and ORP of <−795 mV. The chemical reactions that occur at the anode and cathode are listed in Table 9.2. The strong bactericidal ability of AEW was first reported in 1982 in Japan, which led to the development of a variety of applications in food, medicine, and agriculture. AEW was approved in Japan as a bactericidal agent for food products in 2000. In the U.S., AEW has been used for sanitation of fertile hatching eggs (Yoshida 2004; Russell 2004) and in the brewing industry (Hopkins and Kindred 2006). AEW has shown a strong bactericidal effect against a wide spectrum of microorganisms, including pathogens such as E. coli, Salmonella, L. monocytogenes, Shigella, Bacillus cereus, and Campylobacter jejuni (Bari and others 2003; Park and others 2002; Kim and others 2000) and spoilage microorganisms such as Penicilium expansum and Aspergillus parasiticus (Okull and Laborde 2004). The biosphere of microorganisms (bacteria and viruses) is defined by a region with pH from 3 to 10 and ORP from +900 to −400 mV. Because the AEW has a pH of <2.6 and an ORP of >1100 mV,

Table 9.2. Chemical reactions at the anode and cathode of an electrolyzing unit Anode

Cathode

2H2O → O2 + 4H+ + 4e 2OH− − 2e → 2OH· 2OH. → (O) + H2O (O) + O2 → O3 2Cl− → Cl2 + 2e Cl2 + H2O → HOCl + HCl

2H2O + 4e → H2 + 2OH− Na+ + OH− → NaOH

176

Section III. Postharvest Interventions

it creates conditions difficult for microorganisms to survive. Furthermore, the disinfecting effect of AEW can be strengthened by the residual chlorine in it, which may function to decompose fat and protein of the cell membrane. The synergistic effect of low pH, high ORP, and free chlorine was found to inactivate cytoplasmic enzymes (Kiura and others 2002), make the membrane potential of organelles exceed its stabilizing limit, and inhibit energy metabolism and respiration (Anonymous 2004). AEW has been examined for its efficacy in food safety and agricultural applications. Venkitanarayanan and others (1999) used AEW to treat cell suspensions of E. coli O157:H7, S. enteritidis, and L. monocytogenes, and reported a reduction of >7.0 log CFU/ml for all three organisms in a few minutes. AEW was also used to treat surfaces of fresh vegetables (Kolseki and others 2004). Koseki and others (2001) used AEW to wash lettuce in comparison with ozone and sodium hypochlorite. They found that AEW failed to inactivate microbes inside the cellular tissue. Park and others (2001) conducted inactivation tests with lettuce inoculated with E. coli O157:H7 and L. monocytogenes and examined the quality of treated lettuce during 2 weeks of storage. The effectiveness of AEW for inactivating L. monocytogenes biofilms was studied by Kim and others (2001). They reported that the adherent cell population on stainless steel coupons was reduced by about 9 log CFU/ml after 300 sec of treatment. The fungicidal effect of AEW on pear fruit was demonstrated by Al-Haq and others (2002). The effect of frozen AEW on lettuce during storage was examined by Koseki and others (2002) and a 1.5 log CFU/g reduction of aerobic bacteria was achieved. In the tests using AEW to inactivate Salmonella on alfalfa seeds and sprouts, Kim and others (2003) reported that 10 min of treatment with AEW can reduce the population of Salmonella by 1.66 log CFU/g for alfalfa seeds. When combining AEW with sonication, they were able to achieve 2.3 log CFU/g greater reduction than with treatments of AEW alone. AEW was found to be very effective in inactivating E. coli O157:H7, Salmonella and L. monocytogenes spot-inoculated on tomato surfaces, with reductions of 7.85, 7.46, and 7.54 log CFU/g after 40 sec of treatment (Bari and others 2003), respectively. AEW wash effectively inhibited natural flora growth on fresh-cut cilantro during a 14-day storage (Wang and others 2004). Other types of electrolysis equipment do not contain the separating membrane and therefore produce electrolyzed neutral water (ENW) with pH close to neutral. ENW was effective for decontaminating fresh-cut vegetables, such as spinach (Izumi and others 2000), and carrots and cucumbers (Izumi 1999). The bactericidal effect of ENW was 2–3 times higher than that of the sodium hypochlorite solution with the same active chlorine concentration (Izumi 1999). The advantages of AEW and ENW include their lack of production of chlorinated by-products and their nontoxicity in humans. The pH, ORP, and chlorine ion concentration of ENW do not change much during storage in a closed dark-brown glass bottle at room temperature for 21 days; the total free available chlorine and dissolved oxygen concentration in ENW significantly decrease in storage (Hsu and Kao 2004). Similar to chlorinated water, the efficacy of electrolyzed water can decrease in the presence of organic matter that contributes to diminishing the amount of freely available chlorine (Oomori and others 2000).

Aqueous Antimicrobial Treatments

177

Chlorine Dioxide Chlorine dioxide is a small, volatile, and highly energetic molecule. In contrast to the hydrolysis of chlorine gas in water, chlorine dioxide in water does not hydrolyze to any appreciable extent but remains in solution as a dissolved gas (Aieta and Berg 1986). Chlorine dioxide cannot be stored commercially or shipped as a gas because it is highly explosive under pressure; therefore, it must be generated at the point of use. There are multiple ways to generate chlorine dioxide. Most conventional generators react sodium chlorite (NaClO2) with gaseous chlorine (Cl2), hypochlorous acid (HClO), or hydrochloric acid (HCl) to generate chlorine dioxide gas (ClO2). The reactions are as follows (Anonymous 1999): Gaseous chlorine-chlorite solution: 2 NaClO2 + Cl2 = 2ClO2 + 2 NaCl Aqueous chlorine-chlorite solution: 2 NaClO2 + HClO = 2ClO2 + NaCl + NaOH Acid-chlorite solution: 5NaClO2 + 4HCl = 4ClO2 + 5NaCl + 2H 2 O Electrochemical generators cycle sodium chlorite solutions through an electrolyte cell to generate chlorine dioxide. Compared to the amount of information available on chlorine as an antimicrobial for fresh produce, much less is available about aqueous chlorine dioxide on the effectiveness as an antimicrobial for fresh and fresh-cut produce. One benefit of using chlorine dioxide instead of chlorine is that chlorine dioxide produces fewer potentially carcinogenic chlorinated by-products such as trihalomethanes in the presence of organic material (Richardson and others 1998). Chlorine dioxide is also less affected by pH and is typically used between pH 6–10 (Dychdala 1991). However, like chlorine, the antimicrobial activity of chlorine dioxide is diminished in the presence of organic matter that reacts and reduces chlorine dioxide’s effectiveness (Dychdala 1991; Beuchat and others 2004b). As a dissolved gas in water, chlorine dioxide will off-gas if the concentration in solution or the water temperature is too high or if agitation is added to the wash system (Kelley 2004). From a safety perspective, levels as low as 3 ppm chlorine dioxide used in recirculated wash water led to off-gassing and caused respiratory discomfort for workers (Reina and others 1995; Roberts and Reymond 1994). Thus, in indoor situations ventilation equipment would be necessary to prevent worker exposure to excessive chlorine dioxide vapors. Chlorine dioxide is approved for use as an antimicrobial agent in water used to wash fruits and vegetables that are not raw agricultural commodities in the amount not to exceed 3 ppm residual chlorine dioxide (Anonymous 2007a). In addition, the treatment of produce with chlorine dioxide must be followed by a potable water rinse or blanching, cooking, or canning. Although the inactivation mechanism of chlorine dioxide is not well understood, it is thought that chlorine dioxide disrupts the outer

178

Section III. Postharvest Interventions

membrane of the cell wall (Aieta and Berg 1986). The efficacy of chlorine dioxide varies based on concentration, exposure time, and temperature, and to some extent based on pH, type of fruit or vegetable, and microbial species (Zoffoli and others 2005; Zhang and Farber 1996; Benarde and others 1965). Roberts and Reymond (1994) evaluated the use of chlorine dioxide in a commercial apple presize dump tank and concluded that 3–5 ppm of chlorine dioxide effectively controlled fungal spores in the recirculated process water. Reina and others (1995) concluded that 1.33 ppm chlorine dioxide effectively controlled the microbial buildup in the hydro-cooling water of cucumbers, but had little effect on the microorganisms on the fruit. Other studies have shown that exposure to 10 ppm of chlorine dioxide for 20 min was required to inactivate Monilinia laxa on infected wounds on nectarines (Mari and others 1999), and only 30 sec was necessary to control fungal pathogens associated with pears (Spotts and Peters 1980). Zhang and Farber (1996) found that a 10-min exposure of shredded lettuce or cabbage to 5 ppm chlorine dioxide caused a 1.1-log reduction of L. monocytogenes, while Rodgers and others (2004) achieved a 4-log reduction using 5 ppm for 5 min against L. monocytogenes on shredded lettuce. The difference in results from these studies is likely due to variations in the methods employed. More substantial reductions of pathogens on the surface of produce have been achieved by treatments with gaseous chlorine dioxide (Sy and others 2005; SunYoung and others 2004; Du and others 2003; Han and others 2000). Acidified Sodium Chlorite Acidified sodium chlorite (ASC) forms by combining a weak acid such as citric acid with sodium chlorite in aqueous solution. It is done under strictly controlled conditions of concentration and pH to minimize the production of chlorine dioxide and maximize the formation of chlorous acid (HClO2). It is characterized by the following reaction:

( Weak Acid) H + + NaClO2 ↔ HClO2 It is hypothesized that the uncharged chlorous acid is able to penetrate bacterial cell walls and disrupt protein synthesis by virtue of its reaction with sulfhydryl, sulfide, and disulfide containing amino acids and nucleotides. The undissociated acid is hypothesized to facilitate proton leakage into cells and thereby increase the energy output of the cells to maintain their usual internal pH. Disruption of membrane activity adversely affects amino acid transport (Warf and Kemp 2001). Acidified sodium chlorite is approved for washing fruits and vegetables at levels that result in chlorite concentration of 500 ppm and 1200 ppm in combination with any generally recognized as safe (GRAS) acid at levels sufficient to achieve a pH between 2.3–2.9 and must be followed by a potable water rinse or by blanching, cooking, or canning (Anonymous 2007c). In literature reports, ASC has shown promising results on pathogen populations inoculated onto produce surfaces and appears to be more tolerant to organic loading. However, some published studies have reported organoleptic effects to certain produce when using high levels on cut surfaces. The treatment of Chinese cabbage with 500 ppm ASC at pH 2.2–3.1 using various organic acids and a treatment time of 15 min significantly reduced levels of E. coli O157:H7 by 2.5–3.0 log CFU/g without causing apparent changes in color (Inatsu and

Aqueous Antimicrobial Treatments

179

others 2005a). Additional enhancements in microbial reduction were achieved using mild heat (50 °C) in combination with the ASC solutions; however, this combination was detrimental to product quality. In another study by Inatsu and others (2005b), washing Chinese cabbage with ASC reduced natural aerobic bacteria populations by 2.0 log CFU/g and artificially inoculated pathogen populations by >2.0 log, and populations remained relatively constant through the incubation period of 8 days at 10 °C with the exception of L. monocytogenes, which gradually increased. The efficacy of sanitizers to inactivate E. coli O157:H7 on shredded carrots was compared when using tap water and simulated process water with a chemical oxygen demand of 3500 mg/l (Gonzalez and others 2004). ASC eliminated E. coli O157:H7 to undetectable levels (5.25-log reduction) under both tap water and process water scenarios when used at 1000 ppm for 2 min, and no viable counts were recovered from the samples treated with ASC after enrichment. Ruiz-Cruz and others (2006) found that high levels of ASC caused tissue damage to shredded carrots. By reducing the concentration of ASC to 100–500 ppm, Ruiz-Cruz and others (2007) were still able to reduce pathogen populations to undetectable levels, achieving reductions of 4.81, 4.84, and 2.5 log CFU/g of E. coli O157:H7, Salmonella, and L. monocytogenes, respectively, under both tap and process water conditions. Park and Beuchat (1999) reported reductions of comparable magnitudes when they evaluated ASC at 800– 1200 ppm on cantaloupes, honeydew melons, and asparagus spears. When reviewing the efficacy of ASC in eliminating E. coli O157:H7 from alfalfa seeds, Taormina and Beuchat (1999) found 500 ppm ASC was effective in reducing populations of the pathogen by more than 2 logs. Weissinger and Beuchat (2000) reported slightly lower reductions against Salmonella on alfalfa seeds when using similar concentrations of ASC solutions. Peroxyacetic Acid Peroxyacetic acid (POAA or peracetic acid) has been used for several years as a sanitizer and a disinfectant in the food, dairy, and beverage industry (Cords 1994). Peracetic acid (C2H4O3) is produced by the reaction of acetic acid (CH2COOH) with hydrogen peroxide (H2O2). The reaction is allowed to equilibrate for several days before the maximum level of peracetic acid is generated, as shown in this equation: CH 3COOH + H 2 O2 ↔ CH 3COOOH + H 2 O The primary mode of action for peracetic acid is oxidation. Oxidation is the transfer of electrons, and the stronger the oxidizer the faster the electrons are transferred to the microorganism rendering them inactivated or killed. It has been suggested that POAA disrupts the sulfydryl and sulfur bonds in proteins, enzymes and other metabolites causing rupturing of the outer cell walls (Block 1991). POAA has a stronger oxidizing potential than chlorine, chlorine dioxide, chlorous acid, and hydrogen peroxide, but less than ozone. Peracetic acid is approved for fresh and fresh-cut produce as an antimicrobial process water additive (Anonymous 2007b,f) at a maximum concentration of 80 ppm. One advantage of POAA over other oxidizing treatments is its higher tolerance to organic materials, which allows the active ingredient to maintain effectiveness with the varying degree of soil loading in the recycled wash water (Herdt and others 2007).

180

Section III. Postharvest Interventions

It is typically not necessary to adjust the pH of the source water to maintain the effectiveness of POAA; however, it works best below a pH of 8 (Block 1991). POAA is completely soluble in water and does not have the potential to off-gas like chlorine at pH below 5, or like chlorine dioxide and ozone with increasing water temperatures or agitation. Additionally, POAA does not react with organic matter to form toxic residues, and the breakdown components (hydrogen peroxide, oxygen, acetic acid) are innocuous (Fraser 1987). On apple surfaces inoculated with E. coli O157:H7, 80 ppm of POAA reduced the population by 2.6 logs with a 2-min exposure (Wright and others 2000), which were consistent with the findings of Park and Beuchat (1999) when they reviewed reductions on cantaloupes and honeydew melons. Rodgers and others (2004) achieved much higher reductions against L. monocytogenes and E. coli O157:H7 (4.3 and 4.5 log, respectively) using 80 ppm peracetic acid for 5 min on various commodities. With increased concentrations and exposure times (<160 ppm for 15 min), Wisniewsky and others (2000) achieved a 5-log reduction against E. coli O157:H7 inoculated on whole apples using POAA. When treating fresh-cut apples with 80 ppm POAA, E. coli O157:H7 populations were reduced by 2.7 log with a 5-min treatment and POAA was more effective than chlorine at 80 ppm and AEW at 70 ppm; however, it caused more quality degradation than the other treatments. POAA can also be used to control postharvest fruit decay (Mari and others 1999; Brown and Schubert 1987). The use of POAA on leafy vegetables and carrots has also been reviewed. When initial contamination levels were ≥100 CFU/g, reductions obtained with a 60 ppm POAA solution (1.8 log CFU/g) were significantly larger than those obtained with water (0.8 log CFU/g) on lettuce when used by food-service employees (Smith and others 2003). Hellström and others (2006) found that a 500 ppm POAA solution reduced the number of L. monocytogenes significantly more than a 2500 ppm citric acid–based produce wash or a 100 ppm chlorine solution at 1.7, 1.0, and 0.7 log CFU/g, respectively, on shredded lettuce. However, at lower concentrations Beuchat and others (2004a) found lettuce treated with 80 ppm POAA or 100 ppm chlorine was not significantly different. On shredded carrots POAA reduced levels of Salmonella, E. coli O157:H7, and L. monocytogenes by 2.1, 1.24, and 0.83 log CFU/g, respectively, (Ruiz-Cruz and others 2007); however, these results were not as good as ASC, which reduced populations to undetectable levels. Gonzalez and others (2004) found that when treating shredded carrots in process water, the effectiveness of POAA and ASC were not affected by the high organic content. The efficacy of a peroxyacetic/octanoic acid mixture was compared to peracetic acid alone for use in fruit and vegetable process water (Hilgren and Salverda 2000). Results suggest that the combination of peroxyacetic acid and peroxyoctanoic acid are more effective against yeast and mold population over peracetic acid alone when used in a commercial vegetable processing facility in celery, cabbage, and potato process water. Peracetic acid has also been reviewed against viruses. Lukasik and others (2003) found 100 ppm POAA to be as effective as 200 ppm ASC against bacteria and viruses on fresh strawberries, and both chemicals were better than stabilized chlorine dioxide (100, 200 ppm). Allwood and others (2004) concluded that 200 ppm sodium hypochlo-

Aqueous Antimicrobial Treatments

181

rite, 30,000 ppm hydrogen peroxide, 80 ppm peracetic acid, and 100,000 ppm sodium bicarbonate were all ineffective against F-specific coliphage MS2 and Feline calicivirus (as surrogates to norovirus). Ozone Ozone has been used for many years as a disinfectant in water and wastewater treatment. Its first use in water disinfection dated back to 1893 in Oudshourn, Netherlands (Rice and Farquhar 1982). Currently, more than 200 water and wastewater treatment plants employ ozone treatments for water supplies in the United States. In 1997, ozone as an aqueous disinfectant was declared to be generally recognized as safe (GRAS) for food contact applications (Guzel-Seydim and others 2004). Ozone is cleared by the FDA for use on raw agricultural commodities (Anonymous 2007d). The primary advantages of ozone include strong oxidizing capabilities and fast decomposition in water without leaving any chemical residues. Efficacy of ozone is not affected by the solution pH (Graham 1997). Richardson and others (1998) indicated that the byproducts of ozone are less likely to cause deleterious health effects than the byproducts of chlorine treatment. A drawback of ozonated water wash is its sensitivity to organic loading and potential off-gassing. Ozone is formed by the excitation of molecular oxygen (O2) into atomic oxygen (O) in an energizing environment that allows the recombination of atoms into O3, as shown in the following reaction: 3O2 + electricity → 2O3 The half-life of molecular ozone in air is relatively long (approximately 12 hr); however, the half-life of ozone in aqueous solutions is significantly shorter and depends on temperature, pH, UV light, initial O3 concentration, and concentration of radical scavengers. Ozone has shown a strong bactericidal effect against a wide spectrum of microorganisms, including the chlorine-resistant parasites Cryptosporidium parvum and Giarda lambla oocysts, both of which have invaded food and water supplies and caused deaths in recent years (Kim and others 1999a). Exposure to 1 mg/l ozone for 5 min achieved >90% inactivation of Cryptosporidium parvum oocysts (Korich and others 1990). In contrast, a nearly 90-min exposure to 80 mg/l of chlorine was required to achieve the same results. Ozonated water (0.1–0.2 mg/l) was reported to instantaneously kill 5 logs S. typhimurium and E. coli, 3 logs Pseudomonas aeruginosa and Yersinia enterocolitica and 4.5 log Candida albicans and Zygosaccharomyces bacilli and 1 log Aspergillus niger after 5 min (Restaino and others 1995). Ozone can also destroy pesticides and chemical residues, such as chlorinated by-products, and convert nonbiodegradable organic materials into biodegradable forms (Kim and others 1999b). The primary mode of action of ozone is oxidation. Ozone acts by oxidizing the outer cell membrane of vegetative bacterial cells, endospores, yeast, and mold spores (Graham 1997). The oxidation of cell membrane results in a change in cell permeability, eventually leading to cell lysis and death (Murray and others 1965). Ozone may inactivate microorganisms by causing damage to their DNA (Scott 1975). Ozone is a more efficient sanitizer than chlorine for inactivating microorganisms on meat, poultry, eggs, fish, fruits, vegetables, and dry fruits (Kim and others 1999b).

182

Section III. Postharvest Interventions

Aqueous ozone has been tested for its efficacy in the decontamination of lettuce (Kim and others 1999b; Montecalvo 1998), and alfalfa seeds and sprouts (Sharma and others 2002a,b; Wade and others 2003). Kim and others (1999a) reported a 4-log reduction of mesophilic bacteria on lettuce when exposed to 1.3 mg/l ozone. Similarly, Montecalvo (1998) reported 4-log reduction of E. coli O157:H7 on lettuce when exposing lettuce to 3 mg/l ozone for 3 min Sharma and others (2002a,b), and Wade and others (2003) both reported that extending the exposure time of sprouts to ozone increased log reduction of L. monocytogenes and E. coli O157:H7. Spotts and Cervantes (1992) and Kim and others (1999b) proposed that bacterial survival of ozone treatments was not attributed to a lack of effectiveness of ozone, but actually to a lack of contact with ozone when bacteria were entrapped in the surface wounds. Treating vegetables with ozone gas at 170 mg/l at 25 °C for 20 min and then immersing vegetables in an aqueous solution of ozone at 20 mg/l at 20 °C provided a wider antimicrobial spectrum than treatment with ozone gas or solution alone. Ozone treatment was reported to extend the shelf life of fresh produce by removing ethylene gas to slow down ripening (Rice and Farquhar 1982). Produce tested for this purpose included blackberries (Barth and others 1995), black pepper (Zhao and Cranston 1995), grapes (Sarig and others 1996), broccoli, carrots, and tomatoes (Hampson and Fiori 1997). Barth and others (1995) reported that fungal development on blackberries was suppressed during storage at 2 °C in air with 0.3 mg/l ozone. Baranovskaya and others (1979) reported that the shelf life of potatoes was extended to 6 months with good quality at 6–14 °C and 93–97% relative humidity with 2 mg/l ozone. Hydrogen Peroxide Hydrogen peroxide is listed as GRAS for use in specific food products as a bleaching agent, an oxidizing and reducing agent, and an antimicrobial agent (Anonymous 2007e). Hydrogen peroxide is a well-established antimicrobial agent used as an antiseptic on wounds, for dental and medical instrument disinfection, and as a sterilizing agent for aseptic packaging containers (Block 1991). As an antimicrobial agent, it is allowed for use in cheese making, during the preparation of modified whey, and during thermophile-free starch production. However, it is not specifically approved by the FDA for use on minimally processed fruits and vegetables unless it is used in combination with acetic acid to form peroxyacetic acid (Anonymous 2007b). The EPA, on the other hand, exempts from the requirement of tolerance application rates of ≤1% hydrogen peroxide on all food commodities when it is applied to growing or postharvest crops (Anonymous 2007g). Although vaporized hydrogen peroxide provides some promise for some commodities (Sapers and Simmons 1998; Simmons and others 1997), the use of hydrogen peroxide for washing fresh and fresh-cut fruits and vegetables has been investigated by several researchers. The treatment of lettuce leaves with 2% hydrogen peroxide at 50 °C for 90 sec reduced pathogenic populations and maintained good sensory qualities for up to 15 days. Specifically, 4.5-, 4.7-, and 2.7-log reductions of Salmonella enteritidis, E. coli O157:H7, and L. monocytogenes, respectively, were obtained (Lin and others 2002). Larger reductions in pathogen populations were observed with a combination of lactic acid and hydrogen peroxide; however, the quality of the lettuce was compromised.

Aqueous Antimicrobial Treatments

183

Inoculated apples washed with 5% hydrogen peroxide alone or in combination with acidic surfactants showed 3–4-log reduction of E. coli when treated for 1 min (Sapers and others 1999). Liao and Sapers (2000) achieved 3.3-log reduction of Salmonella Chester on apple skin and 1.52-log reduction in the calyx cavity when washing in a 6% hydrogen peroxide solution for 5 min. The use of 3% hydrogen peroxide for 1 min reduced levels of E. coli O157:H7 inoculated on the surface of strawberries by 2.2 log CFU/g, which was significantly higher than the other chemical treatments tested (Yu and others 2001). The treatments of cantaloupes, honeydew melons, and asparagus with 1% hydrogen peroxide for 3 min was not as effective as chlorine, ASC, or POAA (Park and Beuchat 1999). However, using 5% hydrogen peroxide for 5 min, Ukuku and Sapers (2001) observed a 3.2-log reduction of Salmonella Stanley populations that were inoculated on the surface of cantaloupe. Sapers and others (2001) reported that applying 5% hydrogen peroxide at 50 °C for 1 min to the rind of melons prior to cutting improved the microbial quality and shelf life of fresh-cut cantaloupe. The use of hydrogen peroxide on alfalfa seeds has also been investigated. Taormina and Beuchat (1999) reported significant reductions of E. coli O157:H7 on alfalfa seeds using ≥1% hydrogen peroxide for 3 min with little or no effect on seed germination. When alfalfa seeds inoculated with Salmonella were treated with 10% hydrogen peroxide for 30 sec, counts were reduced from 3.57 log CFU/g to less than 1 log CFU/g (Beuchat 1997). Weissinger and Beuchat (2000) achieved similar results against Salmonella on alfalfa seeds using 8% hydrogen peroxide for 10 min. Organic Acids Many types of produce, especially fruit, naturally contain organic acids. A number of organic acids are generally recognized as safe (GRAS) and have broad regulatory approvals for use in direct food applications. Organic acids are commonly used as antimicrobials in food preservation. Citric and acetic acids are often used in the freshcut industry to adjust the pH of water in chlorine applications, and lactic acid has been studied extensively and is used widely in the meat industry as a postevisceration carcass rinse. In order for an organic acid to provide microbial control in produce washing it must be used at or below the dissociation constant for the specific acid type. The dissociation constants (or pKa) of most organic acids are between 3 and 5 (Beuchat 1992). Organic acids are very stable in the presence of organic material and generally present no objectionable odor. A major disadvantage of using organic acids is the relatively high cost, because it takes large amounts of acid to adjust the system pH, especially for high-alkaline or buffered water sources. Additionally, due to the low pH required, they can compromise the organoleptic properties of some produce. The use of acetic acid to reduce natural microbial populations on lettuce was studied by Nascimento and others (2003). Populations of aerobic mesophilic bacteria, yeast and mold, and total coliforms were reduced by 3.91, >3.58, and >2.25 log CFU/g, respectively, after washing for 15 min in a solution of 4% acetic acid. Washing parsley leaves in a solution of 2% acetic acid for 15 min reduced populations of Yersinia enterocolitica by 7 logs (Karapinar and Gonul 1992). Wright and others (2000) reported that a 2-min treatment with 5% acetic acid was the most effective of several

184

Section III. Postharvest Interventions

treatments in reducing populations of E. coli O157:H7 inoculated on apple surfaces, achieving >3-log reductions. The treatment of strawberries with 5% acetic acid for 1 min reduced populations of E. coli O157:H7 by 1.6 log units (Yu and others 2001). Escudero and others (1999) reviewed the effectiveness of 0.5% acetic and lactic acid alone and in combination with 100 ppm chlorine and found that the combination of lactic acid and chlorine was more effective than any either organic acid individually when tested against Yersinia enterocolitica on lettuce. Similar results were reported by Zhang and Farber (1996) when they reviewed the effectiveness of lactic and acetic acids alone or in combination with chlorine on L. monocytogenes inoculated onto shredded lettuce. A commercial citric acid-based product reduced L. monocytogenes by 1.0 log CFU/g when tested at 0.25% for 1 min on shredded lettuce (Hellström and others 2006). These results were similar to a 100 ppm chlorine wash, which reduced the Listeria population by 0.8 logs. Burnett and others (2004) compared 200 ppm chlorine to a 0.5% citric acid/surfactant–based product against L. monocytogenes on cut lettuce. They reported both products caused similar reduction in L. monocytogenes populations of 1.76 and 1.51 log CFU/piece, respectively. When tested against E. coli O157:H7 on shredded carrots, a different citric acid–based product used at 0.66% was only as effective as water, providing a 0.79-log reduction over the unwashed control. The treatment of lettuce with 50 mM of fumaric acid reduced the native flora bacterial number by 1.4 logs, which was more effective than 200 ppm sodium hypochlorite; however, fumaric acid caused browning of the treated lettuce (Kondo and others 2006).

Conclusion There are many chemical agents that present viable options to reduce microbial populations in process water; however, most of these antimicrobials do not eliminate all pathogens from the surface of produce. A multiple intervention approach throughout the supply chain must be considered to enhance the safety of fresh and fresh-cut produce. A better understanding of the interactions among produce, microorganisms, and antimicrobial agents will provide insight into finding means to improve the efficacy of produce washing, including enhancing the contact time of antimicrobials with pathogens adhered to or entrapped in structures and tissues of fruit and vegetable surfaces.

References Adams MR, Hartley AD, Cox LJ. 1989. Factors affecting the efficacy of washing procedures used in the production of prepared salads. Food Microbiology 6:69–77. Aieta E, Berg JD. 1986. A review of chlorine dioxide in drinking water treatment. Journal of the American Water Works Association 78(6):62–72. Al-Haq MI, Seo Y, Oshita S, Kawagoe Y. 2002. Disinfection effects of electrolyzed oxidizing water on suppressing fruit rot of pear caused by Botryosphaeria berengeriana. Food Research International 35:657–664. Allwood PB, Malik YS, Hedberg CW, Goyal SM. 2004. Effect of temperature and sanitizers on the survival of Feline Calicivirus, Escherichia coli, and F-Specific Coliphage MS2 on leafy salad vegetables. Journal of Food Protection 67(7):1451–1456.

Aqueous Antimicrobial Treatments

185

Anonymous. 1999. Chlorine Dioxide. In: Alternative Disinfectants and Oxidants Guidance Manual. EPA 815-R-99-014. United States Environmental Protection Agency. Chapter 4. ———. 2004. Cleantop® WM-S. Available at: http://www.cbc-de.com/cbc-med/uk/html/uk_main.html. Accessed April 17, 2008. ———. 2007a. Chlorine Dioxide. Code of Federal Regulations 21 CFR 173.300. Office of the Federal Register, U.S. Government Printing Office. Washington D.C. ———. 2007b. Chemicals used in washing or to assist in the peeling of fruits and vegetables. Code of Federal Regulations 21 CFR 173.315. Office of the Federal Register, U.S. Government Printing Office. Washington D.C. ———. 2007c. Acidified Sodium Chlorite. Code of Federal Regulations 21 CFR 173.325. Office of the Federal Register, U.S. Government Printing Office. Washington D.C. ———. 2007d. Ozone. Code of Federal Regulations 21 CFR 173.368. Office of the Federal Register, U.S. Government Printing Office. Washington D.C. ———. 2007e. Hydrogen peroxide. Code of Federal Regulations 21 CFR 184.1366. Office of the Federal Register, U.S. Government Printing Office. Washington D.C. ———. 2007f. Peroxyacetic acid; exemption from the requirement of tolerance. Code of Federal Regulations 40 CFR 180.1196. Office of the Federal Register, U.S. Government Printing Office. Washington D.C. ———. 2007g. Hydrogen peroxide; exemption from the requirement of tolerance. Code of Federal Regulations 40 CFR 180.1197. Office of the Federal Register, U.S. Government Printing Office. Washington D.C. Baranovskaya V, Zapol’skii O, Ovrutskaya N, Yushkevich O. 1979. The use of ozone gas sterilization during storage of potatoes and vegetables. Konserun, Ovoshesus. Promst 4:10–12. Bari ML, Sabina Y, Isobe S, Uemura T, Isshiki K. 2003. Effectiveness of electrolyzed acidic water in killing Escherichia coli O157:H7, Salmonella enteritidis, and Listeria monocytogenes on the surfaces of tomatoes. Journal of Food Protection 66:542–548. Barth MM, Zhou C, Mercier M, Payne FA. 1995. Ozone storage effects on anthocyanin content and fungal growth in blackberries. Journal of Food Science 60:1286–1287. Beattie GA, Marcell LM. 2002. Effect of alterations in cuticular wax biosynthesis on the physicochemical properties and topography of maize leaf surfaces. Plant Cell Environment 25:1–16. Benarde MA, Israel BM, Olivieri VP, Granstrom ML. 1965. Efficiency of chlorine dioxide as a bactericide. Applied Microbiology 13(5):776–780. Beuchat LR. 1992. Surface disinfection of raw produce. Dairy Food Environment Sanitation 12(1):6–9. ———. 1996. Pathogenic microorganisms associated with fresh produce. Journal of Food Protection 59:204–216. ———. 1997. Comparison of chemical treatments to kill Salmonella on alfalfa seeds destined for sprout production. International Journal of Food Microbiology 34:329–333. ———. 1998. Surface decontamination of fruits and vegetables eaten raw: a review. Food Safety Unit, World Health Organization. Beuchat LR, Adler BB, Lang MM. 2004a. Efficacy of chlorine and a peroxyacetic acid sanitizer in killing Listeria monocytogenes on iceberg and romaine lettuce using simulated commercial processing conditions. Journal of Food Protection 67(6):1238–1242. Beuchat LR, Pettigrew CA, Trembaly ME, Roselle BJ, Scouten AJ. 2004b. Lethality of chlorine, chlorine dioxide, and a commercial fruit and vegetable sanitizer to vegetative cells and spores of Bacillus cereus and spores of Bacillus thuringiensis. Journal of Food Protection 67(8):1702–1708. Block SS. 1991. Peroxygen compounds. In: S.S. Block, editor. Disinfection, Sterilization and Preservation, 4th edition. Philadelphia: Lea & Febiger Co. p 167–181. Brackett RE. 1987. Antimicrobial effect of chlorine on Listeria monocytogenes. Journal of Food Protection 50: 999–1003. Brown GE, Schubert TS. 1987. Use of Xanthomonas campestris pv. vescatoria to evaluate surface disinfectants for canker quarantine treatment of citrus fruit. Plant Disease 71(4):319–323. Burnett AB, Iturriaga MH, Escartin EF, Pettigrew CA, Beuchat LR. 2004. Influence of variations in methodology on populations of Listeria monocytogenes recovered from lettuce treated with sanitizers. Journal of Food Protection 67(4):742–750. Chang HT, Rittmann BE, Amar D, Heim R, Ehrlinger O, Lesty Y. 1991. Biofilm detachment mechanisms in a liquid fluidized bed. Biotechnology and Bioengineering 38:499–506.

186

Section III. Postharvest Interventions

Cords BR. 1994. New peroxyacetic acid sanitizer. In: Twenty-third Convention Proceedings. March 1994. Sydney Australia: Institute of Brewing. p 165–169. De Zuniga AG, Anderson ME, Marshall RT, Ianotti EL. 1991. A model system for studying the penetration of microorganisms into meat. Journal of Food Protection 54:256–258. Du J, Han Y, Linton RH. 2003. Efficacy of chlorine dioxide gas in reducing Escherichia coli O157:H7 on apple surfaces. Food Microbiology 20:583–591. Dychdala GR. 1991. Chlorine and chlorine compounds. In: S.S. Block, editor. Disinfection, Sterilization and Preservation, 4th edition. Philadelphia: Lea & Febiger Co. p 131–151. ———. 2001. Chlorine and chlorine compounds. In: S.S. Block, editor. Disinfection, Sterilization, and Preservation, 5th edition. New York: Lippincott Williams & Wilkins. p 135–157. Escudero ME, Velazquez L, Di Genaro MA, De Guzman AMS. 1999. Effectiveness of various disinfectants in the elimination of Yersinia enterocolitica on fresh lettuce. Journal of Food Protection 62(6):665–669. Frank JF, Koffi RA. 1990. Surface adherent growth of Listeria monocytogenes is associated with increased resistance to surfactant sanitizer and heat. Journal of Food Protection 53:550–554. Fraser JAL. 1987. Novel applications of peracetic acid in industrial disinfection. Specialty Chemicals 7:178–186. Garg N, Churey JJ, Splittstoesser DF. 1990. Effect of processing conditions on the microflora of fresh-cut vegetables. Journal of Food Protection 53:701–703. Gonzalez RJ, Luo Y, Ruiz-Cruz S, Cevoy ALM. 2004. Efficacy of sanitizers to inactivate Escherichia coli O157:H7 on fresh-cut carrot shreds under simulated process water conditions. Journal of Food Protection 67(11):2375–2380. Graham DM. 1997. Use of ozone for food processing. Food Technology 51:72–75. Guzel-Seydim Z, Greene AK, Seydim AC. 2004. Use of ozone in the food industry. LebensmittelWissenschaft und Technologie 37:453–460. Hampson BC, Fiori SR. 1997. Applications of ozone in food processing operations. Proceedings of 1997 IOA PAG Conf. Lake Tahoe, Nevada. p 261–267. Han Y, Sherman DM, Linton RH, Nielsen SS, Nelson PE. 2000. The effects of washing and chlorine dioxide gas on survival and attachment of Escherichia coli O157:H7 to green pepper surfaces. Food Microbiology 17:521–533. Hegenbart S. 2002. Ozone—another layer of food safety. Food Product Design Feb:76–79. Hellström S, Kervinen R, Lyly M, Ahvenainen-Rantala R, Korkeala H. 2006. Efficacy of disinfectants to reduce Listeria monocytogenes on precut iceberg lettuce. Journal of Food Protection 69(7):1565– 1570. Herdt JG, Warf CC, Foong-Cunningham S, Swanson KMJ. 2007. Critical factors that influence the effectiveness of aqueous antimicrobial interventions for fresh produce. Presented at Institute of Food Technologists, 2007 Annual Meeting. Chicago, Illinois. July 28–Aug 1. Hilgren JD, Salverda JA. 2000. Antimicrobial efficacy of a peroxyacetic/octanoic acid mixture in fresh-cut vegetable process water. Journal of Food Science 65(8):1376–1379. Holah JT, Thorpe RH. 1990. Cleanability in relation to bacterial retention on unused and abraded domestic sink materials. Journal of Applied Bacteriology 69:599–608. Hopkins J, Kindred DW. 2006. Electrolyzed water treatment for poultry products. WO/2006/107760. Hsu S-Y, Kao H-Y. 2004. Effects of storage conditions on chemical and physical properties of electrolyzed oxidizing water. Journal of Food Engineering 65:465–471. Inatsu Y, Bari MDL, Kawasaki S, Isshiki K, Kawamoto S. 2005a. Efficacy of acidified sodium chlorite in reducing Escherichia coli O157:H7 on Chinese cabbage. Journal of Food Protection 68(2):251– 255. Inatsu Y, Maeda Y, Bari ML, Kawasaki S, Kawamoto S. 2005b. Prewashing with acidified sodium chlorite reduces pathogenic bacteria in lightly fermented Chinese cabbage. Journal of Food Protection 68(5):999–1004. Itoh Y, Sugita-Konishi Y, Kasuga F, Iwaki M, Hara-kudo Y, Saito N, Noguchi Y, Konuma H, Kumagai S. 1998. Enterohaemorrhagic Escherichia coli O157:H7 present in radish sprouts. Applied and Environmental Microbiology 64:1532–1535. Izumi H. 1999. Electrolyzed water as a disinfectant for fresh-cut vegetables. Journal of Food Science 64:536–539.

Aqueous Antimicrobial Treatments

187

Izumi H, Kiba T, Hashimoto. 2000. Efficacy of electrolyzed water as a disinfectant for fresh-cut spinach. Quality assurance in agricultural produce, Australian Center for International Agricultural Research (ACIAR) Proceedings 100:216–221. Karapinar M, Gonul SA. 1992. Removal of Yersinia enterocolitica from fresh parsley by washing with acetic acid or vinegar. International Journal of Food Microbiology 16:261–264. Kelley DG. 2004. Chlorine disinfectants and ORP control. WSU-TFREC Postharvest Information Network. Available at http://postharvest.tfrec.wsu.edu/PC2004B.pdf, accessed December 4, 2007. Kim C, Hung Y-C, Brackett RE. 2000. Efficacy of electrolyzed oxidizing (EO) and chemically modified water on different types of foodborne pathogens. International Journal of Food Microbiology 61:199–207. Kim C, Hung Y-C, Brackett E, Frank JF. 2001. Inactivation of Listeria monocytogenes biofilms by electrolyzed oxidizing water. Journal of Food Processing and Preservation 25:91–100. Kim C, Hung Y-C, Brackett E, Lin C-S. 2003. Efficacy of electrolyzed oxidizing water in inactivating Salmonella on alfalfa seeds and sprouts. Journal of Food Protection 66:208–214. Kim J, Yousef AE, Chism GW. 1999a. Use of ozone to inactivate microorganisms on lettuce. Journal of Food Safety 19:17–34. Kim JG, Yousef AE, Dave S. 1999b. Application of ozone for enhancing the microbiological safety and quality of foods: a review. Journal of Food Protection 62:1071–1078. Kiura H, Sano K, Morimatsu S, Nakano T, Morita C, Yamaguchi M, Maeda T, Katsuoka Y. 2002. Bacterial activity of electrolyzed acid water from solution containing sodium chloride at low concentration, in comparison with that at high concentration. Journal of Microbiology Methods 49:285–293. Kolseki S, Yoshida K, Isobe S, Itoh K. 2004. Efficacy of acidic electrolyzed water for microbial decontamination of cucumbers and strawberries. Journal of Food Protection 67(6):1247–1251. Kondo N, Murata M, Isshiki K. 2006. Efficiency of sodium hypochlorite, fumaric acid, and mild heat in killing native microflora and Escherichia coli O157:H7, Salmonella Typhimurium DT104, and Staphylococcus aureus attached to fresh-cut lettuce. Journal of Food Protection 69(2):323–329. Korich DG, Mead JR, Madore MS, Sinclair NA, Sterling CR. 1990. Effects of ozone, chlorine dioxide, chlorine and monoamine of Cryptosporidium parvum oocyst viability. Applied and Environmental Microbiology 56:1423–1428. Koseki S, Fujiwara K, Itoh K. 2002. Decontaminative effect of frozen acidic electrolyzed water on lettuce. Journal of Food Protection 65(2):411–414. Koseki S, Yoshida K, Isobe S, Itoh K. 2001. Decontamination of lettuce using acidic electrolyzed water. Journal of Food Protection 64:652–658. Liao CH, Sapers GM. 2000. Attachment and growth of Salmonella Chester on apple fruits and in vivo response of attached bacteria to sanitizer treatments. Journal of Food Protection 63(7):876–883. Lin CM, Moon SS, Doyle MP, McWatters KH. 2002. Inactivation of Escherichia coli O157:H7, Salmonella enterica serotype enteritidis, and Listeria monocytogenes on lettuce by hydrogen peroxide and lactic acid and by hydrogen peroxide with mild heat. Journal of Food Protection 65(8):1215–1220. Longley KE. 1978. The role of mixing in wastewater disinfection. American Chemical Society Journal of the American Water Works Association 74:376–390. Lukasik J, Bradley ML, Scott TM, Dea M, Ko A, Hsu W, Bartz JA, Farrah SR. 2003. Reduction of Poliovirus 1, bacteriophages, Salmonella Montevideo and Escherichia coli O157:H7 on strawberries by physical and disinfectant washes. Journal of Food Protection 66(2):188–193. Lund BM. 1983. Bacterial spoilage. In: Dennis C, editor. Post-Harvest Pathology of Fruits and Vegetables. London: Academic Press. p 218–257. Mari M, Cembali T, Baraldi E, Casalini L. 1999. Peracetic acid and chlorine dioxide for post harvest control of Monilinia laxa in stone fruits. Plant Disease 83(8):773–776. Mazollier J. 1988. Lavage-désinfection des salads. Infros-Crofl 41:19–22. Menzer RE, Nelson JO. 2001. Water and soil pollutants. In: Klassen CD, Amdur MO, Doull J, editors. Casserett and Doull’s Toxicology. The Basic Science of Poisons, 6th edition. Columbus, OH: McGrawHill. pp. 1236. Montecalvo JR. 1998. Ozone research results summary. Department of Food Science and Nutrition, California State Polytechnic University, San Luis Obispo. Murray R, Pamela S, Elson H. 1965. Location of mucopeptide of selection of the cell wall of E. coli and other Gram-negative bacteria. Canadian Journal of Microbiology 11:547–560.

188

Section III. Postharvest Interventions

Nascimento MS, Silva N, Catanozi MPLM, Silva KC. 2003. Effects of different disinfection treatments on the natural microbiota of lettuce. Journal of Food Protection 66(9):1697–1700. Nguyen-the C, Carlin F. 1994. The microbiology of minimally processed fresh fruits and vegetables. Critical Reviews in Food Science and Nutrition 34:371–401. Okull DO, Laborde LF. 2004. Activity of electrolyzed oxidizing water against Penicilium expansum in suspension and on wounded apples. Journal of Food Science 69:23–27. Oomori T, Oka T, Inuta T, Arata Y. 2000. The efficiency of disinfection of acidic electrolyzed water in the presence of organic materials. Analytical Science 16:365–369. Park C, Hung YC, Brackett RE. 2002. Antimicrobial effect of electrolyzed water for inactivating Campylobacter jejuni during poultry washing. International Journal of Food Microbiology 72:77–83. Park C, Hung YC, Doyle MP, Ezeike GOI, Kim C. 2001. Pathogen reduction and quality of lettuce treated with electrolyzed oxidizing and acidified chlorinated water. Journal of Food Science 66:1368–1372. Park CM, Beuchat LR. 1999. Evaluation of sanitizers for killing Escherichia coli O157:H7, Salmonella, and naturally occurring microorganisms on cantaloupes, honeydew melons and asparagus. Dairy Food Environment Sanitation 19(12):842–847. Reina LD, Fleming HP, Humphries EG. 1995. Microbiological control of cucumber hydrocooling water with chlorine dioxide. Journal of Food Protection 58(5):541–546. Restaino L, Frampton EW, Hemphill JB, Palnikar P. 1995. Efficacy of ozonated water against various foodrelated microorganisms. Applied and Environmental Microbiology 61:3471–3475. Rice R, Farquhar J. 1982. Review of the application of ozone for increasing storage times of perishable foods. Ozone Science and Engineering 4:147–163. Richardson SD, Thruston AD, Caughran TV, Collette TW, Patterson KS, Lykins BW. 1998. Chemical byproducts of chlorine and alternate disinfectants. Food Technology 52(4):58–61. Roberts RG, Reymond ST. 1994. Chlorine dioxide for reduction of post harvest pathogen inoculum during handling of tree fruits. Applied and Environmental Microbiology 60(8):2864–2868. Rodgers SL, Cash JN, Siddiq M, Ryser ET. 2004. A comparison of different chemical sanitizers for inactivating Escherichia coli O157:H7 and Listeria monocytogenes in solution and on apples, lettuce, strawberries and cantaloupe. Journal of Food Protection 67(4):721–731. Ruiz-Cruz S, Acedo-Félix E, Díaz-Cinco M, Islas-Osuna MA, González-Aguilar GA. 2007. Efficacy of sanitizers in reducing Escherichia coli O157:H7, Salmonella spp. and Listeria monocytogenes populations on fresh-cut carrots. Food Control 18(11):1383–1390. Ruiz-Cruz S, Luo Y, Gonzalez RJ, Tao Y, Gonzalez GA. 2006. Acidified sodium chlorite as an alternative to control microbial growth on shredded carrots while maintaining quality. Journal of the Science of Food and Agriculture 86(12):1887–1893. Russell SM. 2004. Application of electrolyzed water for the poultry industry. Presented at Institute of Food Technologist, 2004 Annual meeting. Las Vegas, Nevada. Sapers GM, Miller RL, Mattrazzo AM. 1999. Effectiveness of sanitizing agents in inactivating Escherichia coli in golden delicious apples. Journal of Food Science 64(4):734–737. Sapers GM, Miller RL, Pilizota V, Mattrazzo AM. 2001. Antimicrobial treatments for minimally processed cantaloupe melon. Journal of Food Science 66(2):345–349. Sapers GM, Simmons GF. 1998. Hydrogen peroxide disinfection of minimally processed fruits and vegetables. Food Technology 52(2):48–52. Sarig P, Zahavi Y, Zutkhi S, Yannai N, Lisker Benarie R. 1996. Ozone for control of post-harvest decay of table grapes caused by Rhizopus stolonifer. Physiology and Molecular Plant Pathology 48:403–415. Scott D. 1975. The effect of ozone on nucleic acids and their derivatives. In: Blogoslawski WJ, Rice RG, editors. Aquatic applications of ozone. 1st edition. Syracuse, New York: International Ozone Institute. p 226–240. Seo KH, Frank JF. 1999. Attachment of Escherichia coli O157:H7 to lettuce leaf surface and bacterial viability in response to chlorine treatment as demonstrated by using confocal scanning laser microscopy. Journal of Food Protection 62:3–9. Shang C, Blatchley III ER. 2001. Chlorination of Pure Bacterial Cultures in Aqueous Solution. Water Research 35:244–254. Sharma RR, Demirci A, Beuchat LR, Fett WF. 2002a. Inactivation of Escherichia coli O157:H7 on inoculated alfalfa seeds with ozonated water and heat treatment. Journal of Food Protection 65:447–451.

Aqueous Antimicrobial Treatments

189

———. 2002b. Inactivation of Escherichia coli O157:H7 on inoculated alfalfa seeds with ozonated water under pressure. Journal of Food Safety 22:107–119. Simmons GF, Smilanick JL, John S, Margosan DA. 1997. Reduction of microbial populations on prunes by vapor phase hydrogen peroxide. Journal of Food Protection 60:188–191. Smith S, Dunbar M, Tucker D, Schaffner DW. 2003. Efficacy of a commercial produce wash on bacterial contamination of lettuce in a food service setting. Journal of Food Protection 66(12):2359–2361. Spotts RA, Cervantes LA. 1992. Effect of ozonated water on postharvest pathogens of pear in laboratory and packinghouse. Plant Disease 76:256–259. Spotts RA, Peters BB. 1980. Chlorine and chlorine dioxide for control of Anjou pear decay. Plant Disease 64:1095–1097. Stan SD, Daeschel MA. 2003. Reduction of Salmonella enterica on alfalfa seeds with acidic electrolyzed oxidizing water and enhanced uptake of acidic electrolyzed oxidizing water into seeds by gas exchange. Journal of Food Protection 66:2017–2022. Sun-Young L, Costello M, Kang DH. 2004. Efficacy of chlorine dioxide gas as a sanitizer of lettuce leaves. Journal of Food Protection 67(7):1371–1376. Sy KV, Murray MB, Harrison MD, Beuchat LR. 2005. Evaluation of gaseous chlorine dioxide as a sanitizer for killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes, and yeasts and molds on fresh and fresh-cut produce. Journal of Food Protection 68(6):1176–1187. Sykes G. 1965. Disinfection and Sterilization, 2nd edition, Philadelphia: Lippincott Williams Wilkins, pp. 486. Taormina PJ, Beuchat LR. 1999. Comparison of chemical treatments to eliminate enterohaemorrhagic Escherichia coli O157:H7 on alfalfa seeds. Journal of Food Protection 62(4):318– 324. Thomas EL. 1979. Myeloperoxidase, hydrogen peroxide, chloride antimicrobial system: Nitrogen-chlorine derivatives of bacterial components in bactericidal action against Escherichia coli. Infection and Immunity 23:522–531. Trueman JR. 1971. The Halogens. In: W.B. Hugo, editor. Inhibition and Destruction of the Microbial Cell. London: Academic Press. p 137–165. Tsai LS, Schade JE, Molyneux BT. 1992. Chlorination of poultry chiller water: chlorine demand and disinfection efficiency. Poultry Science 71(1):188–196. Ukuku DO, Sapers GM. 2001. Effect of sanitizer treatment on Salmonella Stanley attached to the surface of cantaloupe and cell transfer to fresh-cut tissue during cutting practices. Journal of Food Protection 64(9):1286–1291. Venkitanarayanan KS, Ezeike GOI, Hung YC, Doyle MP. 1999. Efficacy of electrolyzed oxidizing water for inactivating Escherichia coli O157:H7, Salmonella enteritidis, and Listeria monocytogenes. Applied and Environmental Microbiology 65:4276–4279. Virto R, Sanz D, Álvarez I, Condon S, Raso J. 2004. Relationship between inactivation kinetics of a Listeria monocytogenes suspension by chlorine and its chlorine demand. Journal of Applied Microbiology 6:1281–1288. Wade WN, Scouten AJ, McWatters KH, Wick RL, Demirci A, Fett WF, Beuchat LR. 2003. Efficacy of ozone in killing Listeria monocytogenes on alfalfa seeds and sprouts and effects on sensory quality of sprouts. Journal of Food Protection 66:44–51. Wang H. 2006. Investigation of the interactions between sanitizers, surface characteristics, washing conditions, and bacteria for improving microbial safety of fresh produce. Dissertation, University of Illinois at Urbana-Champaign. Wang H, Feng H, Liang W, Luo YG. 2005. Surface topography affects the attachment and inactivation of Escherichia coli O157:H7 on fresh fruits. Institute of Food Technologist, 2005 Annual meeting, New Orleans, Louisiana. Wang H, Feng H, Luo Y. 2006. Dual-phasic inactivation of Escherichia coli O157:H7 with peroxyacetic acid, acidic electrolyzed water, and chlorine on cantaloupes and fresh-cut apples. Journal of Food Safety 26:335–347. ———. 2007. Modeling of the effect of washing solution flow conditions on Escherichia coli O157:H7 population reduction on fruit surfaces. Journal of Food Protection 70:2533–2540. Wang H, Feng H, Luo YG. 2004. Microbial reduction and storage quality of fresh-cut cilantro washed with acidic electrolyzed water and aqueous ozone. Food Research International 37:949–956.

190

Section III. Postharvest Interventions

Warf CC, Kemp GK. 2001. The chemistry and mode of action of acidified sodium chlorite. Presented at Institute of Food Technologists, 2001 Annual Meeting. New Orleans, Louisiana. June 23–27. Weissinger WR, Beuchat LR. 2000. Comparison of chemical treatments to eliminate Salmonella on alfalfa seeds. Journal of Food Protection 63(11):1475–1482. Wisniewsky MA, Glatz BA, Gleason ML, Reitmeier CA. 2000. Reduction of Escherichia coli O157:H7 counts on whole fresh apples by treatment with sanitizers. Journal of Food Protection 63(6):703–708. Wright JR, Sumner SS, Hackney CR, Pierson MD, Zoecklein BW. 2000. Reduction of Escherichia coli O157:H7 on apples using wash and chemical sanitizers treatments. Dairy Food Environment Sanitation 20(2):120–126. Yoshida K. 2004. Application of electrolyzed water for food industry in Japan. Presented at Institute of Food Technologist, 2004 Annual meeting, Las Vegas, Nevada. Yu K, Newman MC, Archbold DD, Hamilton-Kemp TR. 2001. Survival of Escherichia coli O157:H7 on strawberry fruit and reduction of the pathogen population by chemical agents. Journal of Food Protection 64(9):1334–1340. Zhang S, Farber JM. 1996. The effects of various disinfectants against Listeria monocytogenes on fresh-cut vegetables. Food Microbiology 13(4):311–321. Zhao J, Cranston PM. 1995. Microbial decontamination of black pepper by ozone and the effect of the treatment on volatile oil constituents of the spice. Journal of the Science of Food and Agriculture 68:11–18. Zhou B, Feng H, Luo Y. 2008. Decay kinetics of chlorine in fresh produce washing system. Institute of Food Technologist, 2008 Annual meeting, New Orleans, Louisiana. Zoffoli JP, Latorre BA, Daire N, Viertel S. 2005. Effectiveness of chlorine dioxide as influenced by concentration, pH, and exposure time on spore germination of Botrytis cinerea, Penicillium expansum, and Rhizopus stolonifer. Cein Inv Agr 32(2):142–148.

10

Irradiation Enhances Quality and Microbial Safety of Fresh and Fresh-Cut Fruits and Vegetables Brendan A. Niemira and Xuetong Fan

Introduction Foodborne illness (FBI) outbreaks associated with contaminated fruits, vegetables, salads, and juices have risen more than fivefold in recent decades (Sivapalasingam and others 2004). Although preharvest (good agricultural practices, GAPs), postharvest (good manufacturing practices, GMPs) and supply-chain (good handling practices, GHPs) controls can help to reduce risk, they have not been able to prevent repeated FBI outbreaks and product recalls of tomatoes, leafy greens, melons, sprouts, and other fresh produce. It is increasingly recognized that the lack of a broadly applicable antimicrobial process (a “kill step”) is hampering the food safety efforts of the fresh produce industry (UFPA 2007; JIFSAN 2007). Conventional thermal processes cannot be applied to leafy vegetables without unacceptable damage, and existing antimicrobial chemical treatments are insufficient to adequately reduce contamination. An antimicrobial process that has come under increased scrutiny is irradiation. Irradiation is the application of controlled doses of ionizing radiation in the form of electron beams, x-rays, or gamma rays (Table 10.1). Irradiation is a nonthermal process that kills spoilage organisms and pathogenic bacteria in a variety of fruits and vegetables (Thayer and Rajkowski 1999; Lacroix and Vigneault 2007). The safety and wholesomeness of irradiated food has been demonstrated numerous times in the 60+ years that this technology has been studied (Thayer and Rajkowski 1999; FDA 2000; Smith and Pillai 2004). In one recent example, the FDA has investigated the possibility that furan, a putative carcinogen present in thermally processed canned foods such as meats, soups, etc., might also be produced during irradiation. The safety of irradiation was reaffirmed when it was recently shown that a dose of 5 kilogray (10 kGy = 1 Mrad) did not induce detectable levels of furan in most fresh-cut fruits and vegetables (Fan and Sokorai 2008). Whereas furan production was above the limit of detection after irradiation, the levels were shown to be much lower than in thermally processed foods. Therefore, irradiation with moderate doses is as safe as thermal processing and has a potential as one of several “hurdles” in fruit and vegetable processing (Smith and Pillai 2004; Niemira and Deschenes 2005). Fresh produce may be irradiated to inhibit sprouting, to delay ripening, to sterilize or kill insect pests, or to reduce microbial populations. Until recently, in the United States, irradiation has regulatory approval for application to produce only for insect control, sprout inhibition, and shelf-life extension. The highest dose allowed for these purposes is 1 kGy (Table 10.2). However, in 2008 the FDA (2008) approved the use of irradiation up to 4.0 kGy on fresh lettuce and fresh spinach to improve food safety 191

192

Section III. Postharvest Interventions

Table 10.1. Major irradiation technologies—advantages and disadvantages Factors Source

Electron Beam Electric power Electrons are generated using electronics and accelerated to high energy using magnetic fields, 5–10 MeVa. When accelerator is powered off, no radiation is emitted.

X-ray Electric power Created when high-energy electrons (up to 5 MeV) strike a metal plate (e.g., tungsten or tantalum alloys); typical conversion efficiency is 5–10%. When accelerator is powered off, no radiation is emitted.

Gamma Radioisotopes Radioactive decay of 60 cobalt (2.5 MeV) or 137 cesium (0.51 MeV). Radioisotope source is always emitting radiation—shielding of source must be the default position.

Mechanism

High-energy electrons cleave water molecules, creating oxygen and hydroxyl radicals that damage DNA, membranes. Direct cleavage of DNA also occurs.

High-energy photons stimulate atoms within target to release high-energy electrons, which cleave water molecules into radicals. Direct cleavage of DNA also occurs.

High-energy photons stimulate atoms within target to release high-energy electrons, which cleave water molecules into radicals. Direct cleavage of DNA also occurs.

Infrastructure required

Shielding:>2 m concrete or <1 m steel/iron/ lead Cooling: extensive for high-voltage electronics and accelerator Ventilation: for ozone removal while unit is operating

Shielding:>2 m concrete or <1 m steel/iron/lead Cooling: extensive for high-voltage electronics and accelerator; additional cooling systems required for plate target Ventilation: for ozone removal while unit is operating

Shielding: Depending on design, >5 m water or >2 m concrete or <1 m steel/iron/lead. Cooling: moderate for control equipment Ventilation: at all times for ozone removal when source is exposed to air

Speedb

Seconds

Seconds

Minutes (depending on source strength)

Penetrabilityc

6–8 cm, suitable for relatively thin or low-density products

30–40 cm, suitable for all products

30–40 cm, suitable for all products

a

MeV = million electron volts. Speed of dose delivery. The desired dose will vary depending on the target organism and commodity irradiated. c Penetrability in food, avg. density approximately 1 g/cm3. This figure will vary for individual commodities due to localized variation in density associated with bone, voids, fibrous matter, etc. b

and extend shelf life. The FDA is currently considering the approval of additional fruit and vegetable commodities listed in a petition filed by the Food Irradiation Coalition that would allow the produce industry to use irradiation to improve the safety of fresh and fresh-cut fruits and vegetables. If approved, treatments of up to 4.5 kGy could be applied.

Irradiation, Quality, and Microbial Safety

193

Table 10.2. United States Code of Federal Regulations 21CFR179.26: Applications and dose limits for irradiated foods Commodity and Purpose Control of Trichinella in pork Suppression of growth and maturation in fresh foods Disinfestation of insect pests Antimicrobial treatment of dry enzymes Antimicrobial treatment of dry herbs and spices Control of pathogens in fresh and frozen raw poultry Sterilization of foods intended for use by NASA Control of pathogens and extension of shelf life of refrigerated and frozen meats Control of Salmonella in fresh shell eggs Control of pathogens in seeds used to produce sprouts Control of Vibrio species and other foodborne pathogens in fresh or frozen molluscan shellfish Control of food-borne pathogens and extension of shelf-life of Iceberg lettuce and spinach

Dose Limits 0.3–1.0 kGy Maximum dose 1.0 kGy Max. 1.0 kGy Max. 10.0 kGy Max. 30.0 kGy Max. 3.0 kGy Minimum dose 44.0 kGy Max. 4.5 kGy (refrigerated), Max. 7.0 kGy (frozen) Max. 3.0 kGy Max. 8.0 kGy Max. 5.5 kGy Max: 4.0 kGy

Figure 10.1. The radura logo required for labeling on irradiated foods. The logo must be accompanied by the text “treated with radiation” or “treated by irradiation.”

The FDA is also considering an updating of the rules regarding required labeling for foods that have been irradiated (FDA 2000). Under the proposed rule, the radura symbol and associated text (Fig. 10.1) would be required only in those foods in which irradiation causes a material change. In this context, the term material change refers to a change in the organoleptic, nutritional, or functional properties of a food. Also, the FDA would allow the use of the terms pasteurized or pasteurization for a food that has been treated by irradiation, where the irradiation results in the same level of reduction as thermal pasteurization. Under current FDA rules, foods that have been

194

Section III. Postharvest Interventions

irradiated must bear both the radura logo and a statement that the food has been “treated with radiation” or “treated by irradiation.” Effective reduction of microbial load requires doses higher than for the other major purposes, and will therefore exert a larger effect on produce quality and shelf life (Farkas and others 1997). Currently, cold chain integrity and modified atmosphere packaging (MAP) are the primary means of ensuring the quality of fresh produce after it leaves the packing or manufacturing facility (Sumner and Peters 1997; Niemira and others 2005). Improper or excessive treatment of fresh produce can lead to changes in firmness, aroma, color, or taste (Yu and others 1996; Mahrouz and others 2004). Delayed effects on phytoplane microbial ecology, including behavior of contaminating pathogens, is also an important consideration (Prakash and others 2000a; Niemira and others 2004; Lacroix and Vigneault 2007). Irradiated fresh produce must adhere to accepted GAP/GMP guidelines for preservation of quality and food safety.

Quality of Irradiated Produce Many studies have demonstrated that most fresh-cut fruits and vegetables irradiated at doses of 1 kGy or less did not exhibit any significant change in appearance, texture, flavor, or nutrient quality (Fan and Sokorai 2002a,b; Fan and others 2003a–c; Khattak and others 2006; Kim and others 2005; Yu and others 1995). Shelf life of some freshcut fruits and vegetables can be extended by low-dose irradiation due to the reduction of spoilage microorganisms. For example, Koorapati and others (2004) showed that irradiation at doses above 0.5 kGy prevented microbial-induced browning and blotching of sliced mushrooms. Studies have also shown that irradiated fresh produce may have higher antioxidant content than nonirradiated controls as irradiation increased synthesis of phenolic compounds (Fan and others 2005a). In some fresh-cut fruits and vegetables, irradiation may cause softening and loss of ascorbic acid (Fan and others 2008b). However, the adverse effects on texture and ascorbic acid due to irradiation are often small compared to variation among cultivars and the changes in storage (Fan and Sokorai 2002a). Irradiation at higher doses (above 1 kGy) often caused an increase in electrolyte leakage of many fresh-cut fruits and vegetables, an indication of cell membrane damage. The increased electrolyte leakage, which may result in a soggy and wilted appearance of leafy vegetables, varies among vegetables. In a study of thirteen vegetables, Fan and Sokorai (2005) observed that red cabbage, broccoli, and endive had the lowest increases in electrolyte leakage; celery, carrot, and green onion had the highest increase in leakage. The losses in quality due to irradiation can be minimized by combination with other sanitizers or techniques such as MAP, heat treatment, calcium infiltration, and antibrowning agents (Prakash and Foley 2004; Niemira and Fan 2006). For example, Boynton and others (2006) showed that fresh-cut cantaloupes irradiated at 1 kGy in MAP of 4% O2, 10% CO2 had the highest rating in sweetness and cantaloupe flavor intensity and lowest in off-flavor after 17 days of storage compared to the control and 0.5 kGy samples. Foley and others (2004) combined chlorine (200 ppm) with low-dose radiation to eliminate E. coli O157:H7 on cilantro and found that the combined treatment significantly reduced levels of the pathogen on fresh cilantro while maintaining

Irradiation, Quality, and Microbial Safety

195

product quality. In a sequential combination, sanitization of whole produce prior to irradiation has been shown to synergistically lower the microbial load of fresh-cut produce. Fan and others (2008a) surface-pasteurized whole cantaloupes with 76 °C water for 3 min. Fresh-cut cantaloupe pieces prepared from the pasteurized whole fruits were then packaged in clamshell containers and exposed to 0.5 kGy radiation. They found that samples treated with combined heat and low-dose radiation had lower microflora populations than either treatment alone and maintained the quality of the product. Overall, the studies conducted in the last decade demonstrated that most fresh-cut fruits and vegetables can tolerate up to 1 kGy radiation without deleterious sensory impact.

Microbial Safety of Irradiated Produce Irradiation doses sufficient to achieve a 1-log reduction for surface-contaminating bacterial pathogens are typically in the range of 0.2–0.8 kGy. Viruses and fungi are generally more resistant, often requiring 1–3 kGy to achieve the same level of reduction (Niemira and Sommers 2006). Doses required for 3-log reductions of viruses and fungi are deleterious for most types of produce. However, it is important to recognize that the majority of serious foodborne illnesses resulting in hospitalizations and deaths (60% and 72% of the total, respectively) are attributed to bacterial pathogens (Mead and others 1999). Irradiation is therefore suitable for inactivating bacterial pathogens such as E. coli O157:H7, Salmonella, and Listeria, the most serious safety threats for consumers of fruits and vegetables. Relatively low doses of irradiation can result in significant reductions of foodborne pathogens. A 1 kGy dose resulted in a 4-log reduction of total aerobic plate counts (TAPC) and L. monocytogenes on bell peppers (Farkas and others 1997). The same degree of reduction of TAPC was obtained on peeled, ready-to-use carrots (Lafortune and others 2005). In that study, reductions increased to 4.5 log when MAP was substituted for air packaging. A 1 kGy dose also produced a 5-log reduction of E. coli and L. monocytogenes on diced celery (Prakash and others 2000b). These studies yielded a D10 for these pathogens of 0.2–0.3 kGy. D10 values for E. coli O157:H7 on radish, alfalfa, and broccoli sprouts were 0.34, 0.27, and 0.26 kGy, respectively (Rajkowski and Thayer 2000). Bari and others (2004) obtained D10 of 0.3 kGy for both E. coli O157:H7 and Salmonella on radish sprouts, but lower values (0.16–0.18 kGy) on mung bean sprouts. Irradiation of melons at 0.5 or 1.0 kGy reduced the microbial load and improved the keeping quality in storage (Boynton and others 2006). Apple slices treated with the antibrowning agent calcium ascorbate required higher irradiation doses to inactivate inoculated L. monocytogenes (Fan and others 2005b). However, the compound also protected apple slices from the negative sensory effect impact resulting from the higher dose. Irradiation of sliced carrot (2 kGy) reduced E. coli, Yersinia enterocolitica, and L. monocytogenes to undetectable levels (Kamat and others 2005). The D10 values were calculated to be 0.12 kGy for E. coli, 0.26 kGy for Y. enterocolitica, and 0.3–0.5 kGy for L. monocytogenes. In the same study, irradiated carrots showed insignificant losses in sucrose, total carotenes, and ascorbic acid, and had two- to fourfold increases in the refrigerated shelf life. Several isolates of Salmonella inoculated onto

196

Section III. Postharvest Interventions

diced tomatoes showed a D10 of 0.26–0.39 kGy when combined with a 1% calcium chloride dip (Prakash and others 2007). Goularte and others (2004) obtained D10 values of ∼0.11 kGy for E. coli O157:H7 and ∼0.2 kGy for Salmonella on shredded Iceberg lettuce. Niemira and others (2002) determined that D10 for E. coli O157:H7 inoculated on iceberg, Boston, red leaf, or green leaf lettuce was dependent on which type of lettuce was examined. D10 in that study ranged from 0.12–0.14 kGy. In a subsequent study, the D10 for Salmonella was also dependent on the variety of the suspending lettuce and ranged from 0.23–0.35 kGy on the same four varieties (Niemira 2003). In contrast, L. monocytogenes did not show the same type of alteration on irradiation sensitivity. D10 values were invariant on the four lettuce types (0.19–0.20 kGy) (Niemira 2003). A combination of chlorination and irradiation at doses of 0.15–0.5 kGy produced fresh-cut lettuce with a reduced microbial population (Hagenmaier and Baker 1997). Although 0.81 kGy reduced the firmness of lettuce, resulting in lower shear force, 0.5 kGy or less did not affect shear force, and irradiated samples had similar shelf life as the control samples. A later study (Foley and others 2002) found that chlorination plus irradiation (5.5 kGy) reduced TAPC, yeasts, molds, and E. coli O157:H7 by 5.4 logs in shredded iceberg lettuce without softening of tissues. Pathogen regrowth in storage following irradiation is a known phenomenon. Irradiation protocols must therefore be optimized within the context of GHP to ensure lasting suppression of the target pathogens throughout the storage period. Romaine lettuce, inoculated with L. monocytogenes, gave D10 values of 0.16–0.25 kGy and presented no indication of regrowth in refrigerated storage (Mintier and Foley 2006). L. monocytogenes was observed to regrow in refrigerated storage on endive leaves following 0.42 kGy, a dose equivalent to effecting a 2-log reduction (Niemira and others 2003). However, 0.84 kGy, equivalent to effecting a 4-log reduction, suppressed L. monocytogenes throughout the 19-day storage period. Combining irradiation with passive MAP was insufficient to suppress the regrowth of L. monocytogenes to regrow after exposure to doses of irradiation sufficient to achieve 1–3-log reductions (Niemira and others 2004). However, an active MAP using reduced-O2, enhanced-CO2 effectively prevented the pathogen from regrowing after these low irradiation doses. Protected Pathogens Pathogens that are hidden within natural anatomical openings are often protected from chemical sanitizers and other conventional antimicrobial processes (Takeuchi and Frank 2000). Similarly, pathogens in biofilms on produce surfaces are protected from chemical antimicrobial treatments (Stewart and others 2004; Robbins and others 2005). Free-living (planktonic) cells of L. monocytogenes were reduced by 8 logs after a 0.5 min exposure to 10 ppm sodium hypochlorite; in the biofilm habitat, 1,000 ppm sodium hypochlorite for 20 min yielded only a 2-log reduction (Norwood and Gilmour 2000). Compared with their planktonic counterparts, biofilm-associated cells of E. coli O157:H7 (Ryu and Beuchat 2005), Staphylococcus aureus (Luppens and others 2002), and Salmonella (Joseph and others 2001) required orders-of-magnitude increases in treatment severity to effect adequate kill.

Irradiation, Quality, and Microbial Safety

197

These results indicate that conventional processes are inadequate to address contamination by pathogens in protected areas. Irradiation, as a penetrating process, holds more promise in targeting this type of contamination, but the literature on the efficacy of this application is not extensive. Pathogens within Biofilms Research has recently begun to assess the ability of irradiation to inactivate internalized or biofilm-associated pathogens. Irradiation is a penetrating process, but the efficiency of irradiation in killing protected pathogens is not well known. The limited data available suggest that the particular isolate and the biofilm culture conditions (growth temperature, medium, time of cultivation, etc.) can influence irradiation efficacy. Biofilm-associated cells of S. Stanley and S. Enteritidis were significantly more sensitive to ionizing radiation than respective planktonic cells, although S. Anatum showed no increase in radiation sensitivity for biofilm-associated cells (Niemira and Solomon 2005). The antimicrobial efficacy of irradiation against Salmonella is therefore observed to be preserved or enhanced when treating biofilm-associated bacteria based on in vitro evidence. Biofilm-associated cells of E. coli O157:H7 ATCC 43894 grown at 37 °C for 24, 48, or 72 h were significantly more sensitive to irradiation than respective planktonic cells (Niemira 2007). E. coli O157:H7 C9490 biofilms grown for 24 h showed a similar increase in sensitivity; the radiation-sensitivity of biofilmassociated cells of E. coli O157:H7 ATCC 35150 were significantly reduced at 24 h. The response of biofilm and planktonic cells were not different for either of these two E. coli O157:H7 isolates in older biofilms (48 and 72 h). Biofilm-associated cells of E. coli O157:H7 were therefore sometimes more sensitive to irradiation and sometimes less so, with D10 values that varied as much as 27% above or below the D10 values obtained for planktonic cells. The modest amount of information that is available on the relative sensitivity of planktonic and biofilm cells to irradiation suggests a complex difference between the two physiological states of these cells (Niemira 2007). Further research in this emerging field is expected to improve our understanding of how biofilms may alter the efficacy of irradiation. Internalized Pathogens Pathogen internalization in produce and the resulting increase in the risk of FBI is a subject of ongoing research. Studies with lettuce (Solomon and others 2002a,b), barley (Kutter and others 2005), and maize (Bernstein and others 2007) have shown that pathogen internalization can occur when introduced via irrigation water, contaminated soil, or other means. However, other studies using tomatoes have found that internalization via the root system either does not occur, or is extremely inefficient (Jablasone and others 2004, 2005). Additional research will further improve our understanding. Bacterial populations within a leaf, fruit, or vegetable are isolated from conventional antimicrobial treatments. A penetrating process such as irradiation may be suited to addressing this problem, although very few studies have yet investigated this. The inefficient uptake of bacteria via roots and vasculature make microbiological analysis problematic. Nevertheless, although the body of literature is relatively scant

198

Section III. Postharvest Interventions

at the present time, irradiation has been shown to eliminate pathogenic bacteria internalized within leaf tissues as a result of root uptake. Lettuce plants grown in hydroponic solutions inoculated with E. coli O157:H7 contained the pathogen in the leaf tissue. Irradiation effectively killed the pathogen although a treatment with 200 ppm aqueous chlorine was ineffective (Nthenge and others 2007). As an alternative to irrigation with contaminated water, a direct inoculation method that introduces inoculum into the intracellular spaces of leaves has been developed. In these studies, irradiation was similarly effective in eliminating internalized E. coli O157:H7 from baby spinach and various types of lettuce (romaine, iceberg, Boston, green leaf, red leaf); 300 or 600 ppm sodium hypochlorite was generally ineffective (Niemira 2007, 2008). The limited data available suggests that D10 values for internalized cells (0.30–0.45 kGy) are often two- to threefold higher than for surface associated cells (0.12–0.14 kGy) (Niemira 2007). Because pathogen populations within the leaf are expected to be very low in a commercial setting, nearly absolute elimination of internalized pathogens may be practically achieved using irradiation doses that do not cause undue sensory damage. Additional research is needed to more fully understand the influence of internalization on pathogens and on the efficacy of irradiation and other treatments.

Irradiation: Not a “Silver Bullet” but a “High Hurdle” Irradiation has sometimes been mischaracterized as a universal cure-all for microbial contamination. Although irradiation is demonstrably effective in killing bacterial pathogens, this efficacy must be practical when assessed within the context of realworld fruit and vegetable processing. The economic and commercial factors that govern other antimicrobial processes are equally relevant for irradiation of fresh and fresh-cut produce. Factors such as cost, efficiency, throughput, administrative and marketing overhead, and other issues will influence how irradiation may ultimately be used to improve the safety of fresh and fresh-cut produce. The most useful understanding of produce irradiation is to consider the commodities and products for which it is appropriate as well as those for which it may be inappropriate. Therefore, although claims to “silver bullet” status must be viewed as hyperbole, it is reasonable to regard irradiation as a “high hurdle” that can be incorporated as one important step in an overall processing plan. Processing Considerations A number of factors influence the antimicrobial efficacy of irradiation. The pathogen targeted, the commodity and its state of preparation (whole vs. cored, peeled, cut, chopped, etc.), type of MAP, and other product-specific factors can all modify the results of the irradiation process. Like any other industrial food-processing technology, irradiation must undergo process validation for each product being treated. Specific details such as time, temperature, handling, etc., will differ depending on commodity and purpose. For example, an irradiation process designed to eliminate Salmonella from tomatoes may yield unacceptable quality or microbiological results when applied, without modification or validation, for the elimination of E. coli O157:H7 from leafy greens.

Irradiation, Quality, and Microbial Safety

199

Calculating the economic cost/benefit for produce irradiation is complex and will be specific to commodity and circumstances. The extra processing step will necessarily incur additional costs in production. These costs will differ for irradiation as a contract service with an independent irradiation facility vs. treatment with in-plant irradiation equipment. In the first case, contracting fees, shipment/transshipment costs, and time in transit are major factors. In the second case, capital costs, facility footprint, operator training, and seasonal utilization/downtime are some of the key factors. In both instances, the throughput capacity and administrative overhead will be important issues. The market benefits (brand-name protection resulting from a safer product, reduced microbial load, reduction of storage losses, premium prices commanded by specialty markets, etc.), may be offset by ancillary market drawbacks (necessity for increased public education/outreach spending, potential for increased regulatory oversight, etc.). Packaging Several packaging materials are approved for use in irradiation of prepackaged foods (Tables 10.3, 10.4). However, a much wider array of packaging materials currently used by the produce industry are diversified, many of which are not yet approved, such as polylactic acid, novel edible coatings, and biodegradable antimicrobial films. Because fruits and vegetables are living, respiring products, the produce industry has developed many complex packaging systems to preserve the color, texture, flavor, and Table 10.3. United States Code of Federal Regulations 21CFR179.45: Packaging materials approved for irradiated foods Material Nitrocellulose-coated or vinylidene chloride copolymer-coated cellophane Glassine paper Wax-coated paperboard Films of polyolefin or polyethylene terephthalate. These may contain: 1. Sodium citrate, sodium lauryl sulfate, polyvinyl chloride* 2. Coatings comprising a vinylidene chloride copolymer containing a minimum of 85% vinylidene chloride with one or more of the following comonomers: acrylic acid, acrylonitrile, itaconic acid, methyl acrylate, and methyl methacrylate Kraft paper (only as a container for flour) Polystyrene film Rubber hydrochloride film Vinylidene chloride-vinyl chloride copolymer film Nylon 11 Ethylene-vinyl acetate copolymers Vegetable parchments Polyethylene film* Polyethylene terephthalate film* Nylon 6 films* Vinyl chloride-vinyl acetate copolymer film* Acrylonitrile copolymers* * This material may be amended with additional materials, listed in Table 10.4.

Maximum Dose 10 kGy 10 kGy 10 kGy 10 kGy

0.5 kGy 10 kGy 10 kGy 10 kGy 10 kGy 30 kGy 60 kGy 60 kGy 60 kGy 60 kGy 60 kGy 60 kGy

200

Section III. Postharvest Interventions

Table 10.4. United States Code of Federal Regulations 21CFR179.45: Adjuvants and amendments approved for incorporation into certain packaging materials approved for irradiated foods Adjuvant/Amendment Amides of erucic, linoleic, oleic, palmitic, and stearic acid. BHA (butylated hydroxyanisole) BHT (butylated hydroxytoluene) Calcium and sodium propionates Petroleum wax Mineral oil Stearates of aluminum, calcium, magnesium, potassium, and sodium Triethylene glycol Polypropylene, noncrystalline

Limit (by Wt. of Polymer) 1% 1% 1% 1% 1% 1% 1% 1% 2%

aroma of these commodities. New packaging materials and new combinations of existing materials are brought to market each year. This poses a challenge for the regulatory review and approval process for irradiation of packaging. For example, polyethylene terephthalate (PET) films are approved by the FDA under 21 CFR 179.45, but rigid and semirigid PETs are not. In cases where the use of the new packaging material in the food-contact article results in a dietary concentration at or below 0.5 ppb, the FDA will consider requests to expand the permissible packaging materials for irradiated foods. However, the processing conditions that must be met for such exemptions to be granted (max. dose <3 kGy, oxygen-free packaging or vacuumfrozen product) are usually not appropriate for fresh produce. New packaging materials that are not currently approved for irradiation, such as biodegradable and antimicrobial packages, adjuvants (antioxidants, stabilizers, etc.), plasticizers, colorants, and adsorbent pads may need more research before being evaluated and approved by the FDA (Komolprasert 2007). Consumer Acceptance Commercial adoption of food irradiation has been limited. Foods that are approved for irradiation make up a relatively short list. Uncertainties regarding the cost of the process and consumer reluctance contribute to this, despite encouraging market research data. Consumers are more willing to buy irradiated foods after they are provided information about the process, with 50% or more willing to buy irradiated food if given the option (Bhumiratana and others 2007). Consumer education is the most influential factor in the purchasing decision. A recent survey of different elements within the produce industry found differences in acceptance of produce irradiation (Anonymous 2007). A majority (63%) of growers/shippers believe that the produce industry should push for irradiation or similar treatments, as long as product quality can be preserved. Among packers, 40% supported irradiation and 40% were undecided. A minority (30%) of growers/shippers think consumers would currently purchase irradiated leafy greens and other produce. At the retail level, only 25% of retailers believe there is consumer acceptance for irradiated produce and only 7% of retailers actually stock irradiated produce. Education and outreach to retailers and

Irradiation, Quality, and Microbial Safety

201

consumers may be an initial requirement to advance the commercial applications of irradiation for fresh produce.

Summary Irradiation is a nonthermal kill step that has great potential for application to fresh and fresh-cut produce. Used as part of an overall program of GAP, GMP, and GHP, irradiation can serve as an important tool in preserving the safety and quality of produce. The wide and expanding range of fruit and vegetable products on the market presents both a challenge and an opportunity for processors wishing to evaluate irradiation. Process validation, including commodity preparation methods, storage conditions, and market forces will identify the appropriate venues for irradiation to be employed. Ultimately, irradiation can play an important role in the production of safe, high-quality produce.

Acknowledgments The authors thank Drs. D. Geveke and J. Gurtler for their thoughtful reviews of this manuscript. Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.

References Anonymous. 2007. Produce Pulse. The Packer 114(6):A1,A4. Bari ML, Al-Haq MI, Kawasaki T, Nakauma M, Todoriki S, Kawamoto S, Isshiki K. 2004. Irradiation to kill Escherichia coli O157:H7 and Salmonella on ready-to-eat radish and mung bean sprouts. J Food Prot 67(10):2263–2268. Bernstein N, Sela S, Pinto R, Ioffe M. 2007. Evidence for internalization of Escherichia coli into the aerial parts of maize via the root system. J Food Prot 70(2):471–475. Bhumiratana N, Belden LK, Bruhn CM. 2007. Effect of an educational program on attitudes of California consumers toward food irradiation. Food Protect Trends 27(10):744–748. Boynton BB, Welt BA, Sims CA, Balaban MO, Brecht JK, Marshall MR. 2006. Effects of low-dose electron beam irradiation on respiration, microbiology, texture, color, and sensory characteristics of fresh-cut cantaloupe stored in modified-atmosphere packages. J Food Sci 71(2):S149–S155. doi:10.1111/ j.1365-2621.2006.tb08918.x. Fan X, Annous B, Beaulieu J, Sites J. 2008a. Effect of hot water surface pasteurization of whole fruit on shelf life and quality of fresh-cut cantaloupes. J Food Sci (in press). Fan X, Niemira BA, Mattheis JP, Zhuang H, Olson DW. 2005a. Quality of fresh-cut apple slices as affected by low dose ionizing radiation and calcium ascorbate treatment. J Food Sci 70(2):S143–S148. Fan X, Niemira BA, Prakash A. 2008b. Irradiation of fresh and fresh-cut fruits and vegetables. Food Technol 3:36–43. Fan X, Niemira BA, Sokorai KJB. 2003a. Sensorial, nutritional and microbiological quality of fresh cilantro leaves as influenced by ionizing irradiation and storage. Food Res Int 36:713–719. ———. 2003b. Use of ionizing radiation to improve sensory and microbial quality of fresh-cut green onion leaves. J Food Sci 68(4):1478–1483. Fan X, Sokorai KJ, Sommers CH, Niemira BA, Mattheis JP. 2005b. Effects of calcium ascorbate and ionizing radiation on the survival of Listeria monocytogenes and product quality of fresh-cut “gala” apples. J Food Sci M352–M358. Fan X, Sokorai KJB. 2002a. Sensorial and chemical quality of gamma irradiated fresh-cut iceberg lettuce in modified atmosphere packages. J Food Prot 65:1760–1765.

202

Section III. Postharvest Interventions

———. 2002b. Changes in volatile compounds of gamma irradiated fresh cilantro leaves during cold storage. J Agric Food Chem 50:7622–7677. ———. 2005. Assessment of radiation sensitivity of fresh-cut vegetables using electrolyte leakage measurement. Postharv Biol Technol 36:191–197. ———. 2008. Effect of ionizing radiation on furan formation in fresh fruits and vegetables. J Food Sci 73(2)C79–C83. Fan X, Toivonen PMA, Rajkowski KT, Sokorai K. 2003c. Warm water treatment in combination with modified atmosphere packaging reduced undesirable effects of irradiation on the quality of fresh-cut iceberg lettuce. J Agric Food Chem 51:1231–1236. Farkas J, Saray T, Mohacsi-Farkas C, Horti K, Andrassy E. 1997. Effects of low dose gamma radiation on shelf-life and microbiological safety of pre-cut/prepared vegetables. Adv Food Sci 19:111–119. FDA 2000. U.S. Food and Drug Administration—Food Irradiation: A Safe Measure. http://www.fda.gov/ opacom/catalog/irradbro.html. Accessed February 21, 2008. ———. U.S. Food and Drug Administration—Final Rule (73 FR 49593), Irradiation in the Production, Processing and Handling of Food 21 CFR Part 179 http://www.cfsan.fda.gov/~lrd/fr080822.html. Accessed September 2, 2008. Foley DM, Dufour A, Rodriguez L, Caporaso F, Prakash A. 2002. Reduction of Escherichia coli O157:H7 in shredded iceberg lettuce by chlorination and gamma irradiation. Rad Phys Chem 63:391–396. Foley DM, Euper M, Caporaso F, Prakash A. 2004. Irradiation and chlorination effectively reduce Escherichia coli O157:H7 inoculated on cilantro (Coriandrum sativum) without negatively affecting quality. J Food Prot 67:2092–2098. Goularte L, Martins CG, Morales-Aizpurua IC, Destro MT, Franco BDGM, Vizeu DM, Huntzler BW, Landgraf M. 2004. Combination of minimal processing and irradiation to improve the microbiological safety of lettuce (Lactuca sativa L.). Rad Phys Chem 71:155–159. Hagenmaier RD, Baker RA. 1997. Low-dose irradiation of cut iceberg lettuce in modified atmosphere packaging. J Agric Food Chem 45:2864–2868. Jablasone J, Brovko LY, Griffiths MW. 2004. A research note: the potential for transfer of Salmonella from irrigation water to tomatoes. J Sci Food Agric 84(3):287–289. Jablasone J, Warriner K, Griffiths M. 2005. Interactions of Escherichia coli O157:H7, Salmonella typhimurium and Listeria monocytogenes plants cultivated in a gnotobiotic system. Int J Food Microbiol 99(1):7–18. JIFSAN (Joint Institute for Food Safety and Applied Nutrition). 2007. Tomato Safety Research Needs Workshop http://www.jifsan.umd.edu/tomato_wkp2007.htm. Accessed February 6, 2008. Joseph B, Otta SK, Karunasagar I. 2001. Biofilm formation by Salmonella spp. on food contact surfaces and their sensitivity to sanitizers. Int J Food Microbiol 64:367–372. Kamat AS, Ghadge N, Ramamurthy MS, Alur MD. 2005. Effect of low-dose irradiation on shelf life and microbiological safety of sliced carrot. J Sci Food Agric 85(13):2213–2219. Khattak AB, Bibi N, Chaudry MA, Khan M, Khan M, Qureshi MJ. 2006. Shelf life extension of minimally processed cabbage and cucumber through gamma irradiation. J Food Prot 69(11):2648–2663. Kim HJ, Hao F, Toshkov SA, Fan X. 2005. Effect of sequential treatment of warm water dip and low dose gamma irradiation and hot water wash on the quality of fresh-cut green onions. J Food Sci 70:M179–M185. Komolprasert V. 2007. Packaging for foods treated with ionizing radiation. In: Han JH (ed), Packaging for Nonthermal Processing of Food. Blackwell Publishing, Ames, IA. pp. 87–116. Koorapati A, Foley D, Pilling R, Prakash A. 2004. Electron-beam irradiation preserves the quality of white button mushroom (Agaricus bisporus) slices. J Food Sci 69:S25–S29. Kutter S, Hartmann A, Schmid M. 2005. Colonization of barley (Hordeum vulgare) with Salmonella enterica and Listeria spp. FEMS Microbial Ecol 56:262–271. Lacroix M, Vigneault C. 2007. Irradiation treatment for increasing fruit and vegetable quality. Stewart Postharv Rev 3(3):1–8. Lafortune R, Caillet S, Lacroix M. 2005. Combined effects of coating, modified atmosphere packaging and gamma irradiation on quality maintenance of ready-to-use carrots (Daucus carota). J Food Prot 68(2):353–359.

Irradiation, Quality, and Microbial Safety

203

Luppens SBI, Reij MW, van der Heijden RWL, Rombouts FM, Abee T. 2002. Development of a standard test to assess the resistance of Staphylococcus aureus biofilm cells to disinfectants. Appl Environ Microbiol 68:4194–4200. Mahrouz M, Lacroix M, D’Aprano G, Oufedjikh H, Beoubekri C. 2004. Shelf-life and quality evaluation of clementine following a combined treatments with gamma-irradiation. Rad Phys Chem 71:141–143. Mead PS, Slutsker L, Dietz V, McCaig LF, Bresee JS, Shapiro C, Griffin PM, Tauxe RV. 1999. Food-related illness and death in the United States. Emerg Infect Dis 5(5):607–625. Mintier AM, Foley DM. 2006. Electron beam and gamma irradiation effectively reduce Listeria monocytogenes populations on chopped romaine lettuce. J Food Prot 9(3):570–574. Niemira BA. 2003. Radiation sensitivity and recoverability of Listeria monocytogenes and Salmonella on four lettuce types. J Food Sci 68(9):2784–2787. ———. 2007. Irradiation sensitivity of planktonic and biofilm-associated Escherichia coli O157:H7 isolates is influenced by culture conditions. Appl Environ Microbiol 73(10):3239–3244. ———. 2008. Irradiation vs. chlorination for elimination of Escherichia coli O157:H7 internalized in lettuce leaves: influence of lettuce variety. J Food Sci (in press). Niemira BA, Deschenes L. 2005. Ionizing radiation processing of fruits and fruit products. In: Barrett DM, Somogyi L, Ramaswamy H (eds), Processing Fruits: Science and Technology, 2nd ed. CRC Press, Boca Raton, FL. pp. 221–260. Niemira BA, Fan X. 2006. Low dose irradiation of fresh produce: safety, sensory and shelf life. Ch. 10. In: Sommers CH, Fan X (eds), Food Irradiation Research and Technology. Blackwell Publishing, Ames, IA. pp. 169–184. Niemira BA, Fan X, Sokorai KJB. 2004. Irradiation and modified atmosphere packaging of endive influences survival and regrowth of Listeria monocytogenes and product sensory qualities. Rad Phys Chem 72(1):41–48. Niemira BA, Fan X, Sokorai KJB, Sommers CH. 2003. Ionizing radiation sensitivity of Listeria monocytogenes and L. innocua inoculated on endive (Cichorium endiva). J Food Prot 66:993–998. Niemira BA, Solomon EB. 2005. Sensitivity of planktonic and biofilm-associated Salmonella to ionizing radiation. Appl Environ Microbiol 71(5):2732–2736. Niemira BA, Sommers CH. 2006. New applications in food irradiation. In: Heldman DR (ed), Encyclopedia of Agricultural, Food, and Biological Engineering. Taylor & Francis, Group, New York, NY. pp. 1–6. DOI: 10.1081/E-EAFE-120042055 (http://www.dekker.com/sdek/linking~db=enc~content=a713617899, accessed Feb 19, 2008). Niemira BA, Sommers CH, Fan X. 2002. Suspending lettuce type influences recoverability and radiation sensitivity of Escherichia coli O157:H7. J Food Prot 65(9):1388–1393. Niemira BA, Sommers CH, Ukuku D. 2005. Mechanisms of microbial spoilage of fruits and vegetables. In: Lamikanra O, Imam SH, Ukuku DO (eds), Produce Degradation: Reaction Pathways and their Prevention. Taylor and Francis Group, New York. pp. 463–482. Norwood DE, Gilmour A. 2000. The growth and resistance to sodium hypochlorite of Listeria monocytogenes in a steady-state multispecies biofilm. J Appl Microbiol 88:512–520. Nthenge AK, Weese JS, Carter M, Wei C-I, Huang T-S. 2007. Efficacy of gamma radiation and aqueous chlorine on Escherichia coli O157:H7 in hydroponically grown lettuce plants. J Food Prot 70(3):748–752. Prakash A, Foley D. 2004. Improving safety and extending shelf-life of fresh-cut fruits and vegetables using irradiation. In: Komolprasert V, Morehouse KM (eds). Irradiation of Food and Packaging: Recent Developments. American Chemical Society, Washington, D.C. pp. 90–106. Prakash A, Guner AR, Caporaso F, Foley DM. 2000a. Effects of low-dose gamma irradiation on the shelf life and quality characteristics of cut romaine lettuce packaged under modified atmosphere. J Food Sci 65(3):549–553. Prakash A, Inthajak P, Huibregtse H, Caporaso F, Foley DM. 2000b. Effects of low-dose gamma irradiation and conventional treatments on shelf life and quality characteristics of diced celery. J Food Sci 65(6):1070–1075. Prakash A, Johnson N, Foley D. 2007. Irradiation D values of Salmonella spp. in diced tomatoes dipped in 1% calcium chloride. Foodborne Pathogens Dis 4(1):84–88. Rajkowski KT, Thayer DW. 2000. Reduction of Salmonella spp. and strains of Escherichia coli O157:H7 by gamma radiation of inoculated sprouts. J Food Prot 63(7):871–875.

204

Section III. Postharvest Interventions

Robbins JB, Fisher CW, Moltz AG, Martin SE. 2005. Elimination of Listeria monocytogenes biofilms by ozone, chlorine and hydrogen peroxide. J Food Prot 68(3):494–498. Ryu J-H, Beuchat LR. 2005. Biofilm formation by Escherichia coli O157:H7 on stainless steel: effect of exopolysaccharide and curli production on its resistance to chlorine. Appl Envir Microbiol 71:247–254. Sivapalasingam S, Friedman CR, Cohen L, Tauxe RV. 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J Food Prot 67:2342–2353. Smith JS, Pillai S. 2004. Irradiation and food safety and scientific status summary. Food Tech 58(11):48–55. Solomon EB, Potenski CJ, Matthews KR. 2002a. Effect of irrigation method on transmission to and persistence of Escherichia coli O157:H7 on lettuce. J Food Prot 65(4):673–676. Solomon EB, Yaron S, Matthews KR. 2002b. Transmission of Escherichia coli O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalization. Appl Envir Microbiol 68(1):397–400. Stewart PS, Mukherjee PK, Ghannoum MA. 2004. Biofilm antimicrobial resistance. In: Ghannoum MA, O’Toole GA (eds), Microbial biofilms. ASM Press, Washington D.C. pp. 250–268. Sumner SS, Peters DL. 1997. Microbiology of vegetables. In: Smith DS, Cash JN, Nip W-K, Hui YH (eds). Ch. 4. Processing Vegetables: Science and Technology. Technomic Publishing Inc., Lancaster, PA. pp. 87–116. Takeuchi K, Frank JF. 2000. Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and the effect of chlorine treatment on cell viability. J Food Prot 63(4):434–440. Thayer DW, Rajkowski KT. 1999. Developments in irradiation of fresh fruits and vegetables. Food Technol 53(11):62–65. UFPA (United Fresh Produce Association). 2007. Leafy Greens Food Safety Research Conference (http:// www.unitedfresh.org/newsviews/leafy_greens_food_safety_research, accessed Feb 6, 2008). Yu L, Reitmeier CA, Gleason ML, Nonnecke GR, Olson DG, Gladon RJ. 1995. Quality of electron beam irradiated strawberries. J Food Sci 60(5):1084–1087. Yu L, Reitmeier CA, Love MH. 1996. Strawberry texture and pectin content as affected by electron beam irradiation. J Food Sci 61(4):844–846.

11

Biological Control of Human Pathogens on Produce John Andrew Hudson, Craig Billington, and Lynn McIntyre

Introduction Reported outbreaks of foodborne disease associated with the consumption of fresh produce are increasing in frequency, and examples of the particular foods and pathogens involved have already been compiled (Beuchat 2002). Current methods for controlling pathogens in fresh produce are therefore either not sufficiently effective or not being used effectively. Existing methods are based primarily on chemical treatments, such as chlorine applied postharvest, which can have significant disadvantages (Sapers 2006), and consumers are concerned about chemical additions to their food (Food Standards Agency 2007). In response, efforts have been made to find alternatives. One option is to develop treatments that use biocontrol, in which organisms, or products derived from them, are used to control pathogens. It is anticipated that such treatments would be perceived favorably by consumers as “clean, green, and natural” alternatives. The potential for using biocontrol as a means of controlling human pathogens on plant products is shown through observations that the natural microbiota on plants provides some defense against the growth of human pathogens. For example, the growth of Listeria monocytogenes in endive was prevented by the presence of bacteria isolated from the plant. Faster growth of the pathogen occurred when the endive had been chemically sanitized, indicating that the native microbiota inhibited the pathogen (Carlin and others 1996). Similar results have been found for the microbiota on freshcut spinach (Babic and others 1997), baby carrots, and green pepper (Liao 2007). The potential therefore exists for the microbial ecology of produce to be manipulated to control human pathogens, and it has been suggested that an understanding of produce microbial communities is the key to pathogen control (Beuchat 2002). Recently effort has been focused on using pathogen-specific bacteriophages to achieve the control of pathogens in produce. Other biocontrol options include the use of protective cultures, typically pseudomonads or lactic acid bacteria (LAB), or natural antimicrobial compounds such as bacteriocins, naturally occurring plant volatiles, and nonvolatile essential oils. Space limitations preclude a fully comprehensive coverage of all potential biocontrols, so we have focused wherever possible on examples that describe biocontrol in vivo on fresh produce. Other methods (Gould 1996) could be examined in the future. The Connection between Control of Human and Plant Pathogens A great deal of information is available concerning the biocontrol of plant pathogens (Shoda 2000; Wilson and Wisniewski 1989). Plant and human pathogens have been 205

206

Section III. Postharvest Interventions

shown to interact (Liao and others 2003). For example, Salmonella Poona penetrated 3–4 cm below a wound site into the edible part of cantaloupe when inoculated alone, but the penetration was enhanced by the presence of two plant pathogenic fungi, particularly Cladosporium cladosporoides (Richards and Beuchat 2005). The growth of Escherichia coli O157:H7 in apple wounds was prevented by the application of Pseudomonas syringae, which is used in products to control postharvest decay of apples and pears (Janisiewicz and others 1999). E. coli O157:H7 grew when incubated at room temperature on punctured apples coinoculated with the plant pathogen Glomerella cingulata but at a rate similar to that obtained in its absence, and the presence of Penicillium expansum resulted in a reduction in pathogen numbers (Riordan and others 2000). Similar findings were made for L. monocytogenes (Conway and others 2000). It has been observed that fruits and vegetables affected by bacterial soft rot are more frequently contaminated by Salmonella (Wells and Butterfield 1997), although a similar relationship was not shown clearly for produce affected by fungal rot or physical injury (Wells and Butterfield 1999). These examples show that plant pathogens, or their associated biocontrol agents, can affect the establishment and/or growth of human pathogens on produce. Research on the biocontrol of plant pathogens can offer options for the control of human pathogens, as illustrated later in this chapter. Location of Pathogens in Produce Pathogens may be introduced to the external surfaces of produce via the application of improperly composted manure or contaminated water. Furthermore they may be physically introduced via a piercing wound (Burnett and others 2000) or from cut edges (Takeuchi and Frank 2000). It has been proposed that pathogens may become internalized into produce via the root system during cultivation, taken up during processing (US FDA 1999) or by a number of other routes (Cooley and others 2003). The possible presence of human pathogens in internal tissues means that they may not be amenable to treatment by chemical sanitizers (Takeuchi and Frank 2000). Under these circumstances biocontrol may prove advantageous over postprocessing chemical treatments by preventing initial colonization. For example, inhibition of plant colonization by E. coli O157:H7 and Salmonella has been demonstrated by an Enterobacter asburiae isolate from the phyllosphere (Cooley and others 2003). In addition, reduction of E. coli O157:H7 and L. monocytogenes on lettuce has been shown after coinfection with Enterobacter cloacae (Jablasone and others 2005). The formation of biofilms may be of significance for both the survival of human pathogens on produce and the ability of biocontrol organisms to control them. Naturally occurring biofilms may provide human pathogens with safe harbor and protect them from the action of sanitizers (Fett 2000). This has been shown in a Salmonella biofilm formed on parsley (Lapidot and others 2006), although other unknown factors were also involved. Salmonella can form biofilms on tomatoes after 4 days storage at 20 °C and 95% relative humidity (Iturriaga and others 2007) and autoinducer-2–like activity, i.e., quorum sensing (QS), is suggested as having a role in biofilm formation by E. coli O157:H7 on the same fruit (Lu and others 2005).

Biological Control of Human Pathogens

207

Biofilm formation may be prevented by QS disruption or by the use of phages able to degrade biofilm polymers, both of which are discussed below.

Biocontrol Technologies The following section is divided into two main subsections: the use of organisms and the use of products derived from them. Coverage of organisms includes bacteriophages, protective bacteria cultures (with the inclusion of bacteriocins), and yeast and molds; derived products include essential oils and other plant-derived chemicals, quorum-sensing molecules, siderophores, enzymes, and polylysine. Organisms Bacteriophages (Phages) A substantial amount of research has been generated on the use of virulent phages as biocontrol agents in foods (Greer 2005; Hudson and others 2005). This approach offers some particular advantages over other treatments mentioned. Virulent phages occur naturally in all environments where bacteria are found, including produce (Kennedy and others 1986). All phages require a bacterial host for replication, and virulent phages kill host cells by cell lysis. Host-phage interaction is mediated by the specificity of the phage, and the interaction can thus be specific at genus, species, or even strain level. Lytic infection results in an increase in phage numbers as they destroy their host, which is a desirable characteristic as is the lack of interference with the growth of the natural competitive microbiota. The use of a proprietary mixture of four virulent phages to control Salmonella Enteritidis on cut apple and melon surfaces stored at 5, 10, and 20 °C has been reported (Leverentz and others 2001). Salmonella survived on both fruits at 5 °C and growth occurred at 10 and 20 °C, with increases of up to 2 log10 and 5 log10 colony forming units (CFU), respectively. The addition of phages reduced Salmonella populations on melon by 3.5 log10 at storage temperatures of 5 and 10 °C; a 2.5-log10 reduction was measured at 20 °C. These reductions were greater than those attributed to the use of chemical sanitizers. Similar reductions in host numbers were not observed for apple slices treated with phages. This was possibly due to inactivation of phages at the lower pH of apple. The authors postulated that “low-pH-tolerant phage mutants” might be needed to improve the efficacy of treatment. Subsequent work has focused on the biocontrol of L. monocytogenes on fresh-cut produce (Leverentz and others 2003). This research described the use of virulent phages in combination with the bacteriocin nisin and examined whether application by spraying or pipetting had a significant effect on host inactivation. Growth of L. monocytogenes occurred most markedly on cut melon stored at 10 °C, with little growth on apple slices stored at the same temperature. Reductions of up to 4.6 log10 were observed when melons were treated with phage, and improved to up to 5.7 log10 when combined with nisin. As before, the lower pH of apple may have minimized the effectiveness of the phage treatment, and reductions in Listeria numbers were observed for only the combined treatment incorporating nisin. Spray application was deemed to be at least as effective as pipetting, and in some cases was better.

208

Section III. Postharvest Interventions

The optimization of the delivery of phages to melon surfaces has been described (Leverentz and others 2004). This research demonstrated the importance of using high-phage concentrations to reduce populations of L. monocytogenes to undetectable levels. Phage application appeared to be most effective when added up to 1 hour prior to contamination, with measurable reductions of up to 6.8 log10 after 7 days of storage at 10 °C. However, addition of phage even up to 4 hours after contamination still resulted in pathogen reductions. The application of two phages to control salmonellae in artificially contaminated broccoli and mustard seeds has been described (Pao and others 2004). The control achieved was limited because log10 reductions of only 1.4 and 1.5 were achieved for one phage on mustard seeds and a combination of both phages on broccoli seeds, respectively. These reductions are significantly lower than reported on cut melon (Leverentz and others 2004), and lower than the reductions attributed to the use of hypochlorite washing on produce in general (Sapers 2006). These reports focused on postharvest application of phages. Although we know of no such work being reported for human pathogen control during plant culture, phage applications to control plant pathogens in the field are known (e.g., Balogh and others 2003), and the ecology discussed (Gill and Abedon 2003). Problems occurring with field application include limitations on the ability of phages to survive in the environment. In particular, the rapid inactivation of some phages by UV light is problematic (Iriarte and others 2007), making the time and means of delivery (i.e., protective formulations) important. An alternative to the use of intact viable phages is to use cell-wall degrading enzymes (endolysins) encoded by phages. This potential has been demonstrated for a plant pathogen Erwinia amylovora (Kim and others 2004). In this work, pear slices were protected against attack by the pathogen when treated with the enzyme, except when the inoculum was the highest (4.7log10/slice) where some signs of infection were evident. Similar activity has been shown for the endolysin from a Xanthomonas oryzae phage infecting Xanthomonas spp. that causes plant disease (Lee and others 2006). Most of the work of this nature in the medical field has focused on Gram-positive organisms because they lack an outer membrane, but the examples cited above and other reports (Paradis-Bleau and others 2007) indicate that Gram-negative pathogens may also be controlled by this means. There is potential for phage treatments to inactivate bacteria present in biofilms. Phages with naturally occurring polymer-degrading enzymes have been described (Hughes and others 1998), and a paper has recently described the engineering of a phage to express a potent biofilm-degrading enzyme to capture the cell-killing ability of the phage and the activity of the enzyme in a modular manner (Lu and Collins 2007). The potential uptake of phage-based biocontrol applications by the food industry has been enhanced in the last 2–3 years by regulatory approvals for the technology. Listex™, a product for the control of Listeria, has been accepted as generally recognized as safe (GRAS) by the FDA and USDA for use on all foods “susceptible to Listeria” (Anonymous 2007). Previously the product had been confirmed as “organic” in Europe. Another phage product, LMP 102, has also received FDA approval for use on ready-to-eat food to control L. monocytogenes (Anonymous 2006).

Biological Control of Human Pathogens

209

Bacterial Protective Cultures and Related Antimicrobial Metabolites The use of protective (or antagonistic) cultures is a relatively well-established and consumer-accepted biocontrol approach, particularly for fermented meats and dairy products (Breidt and Fleming 1997). For various reasons including a previous lack of association of produce with foodborne illness and the nonfermented nature of these foods, specific applications related to fresh fruits and vegetables have, until recently, been rarely documented. Applied research activity is on the increase, and reviews addressing the use of protective cultures (Kostrzynska and Bachand 2006; Leverentz and others 2002) and bacteriocins (Settanni and Corsetti 2008) in fresh produce have arisen as a consequence. The concept of protective cultures is derived from the acknowledged contribution the native microflora makes to intrinsic factors governing the growth of microorganisms, including pathogens, in that food. The ability of certain microbial species, particularly lactic acid bacteria (LAB), to behave antagonistically toward closely related organisms, including pathogens such as L. monocytogenes, is well established and extensively reviewed, as are the mechanisms whereby antagonism occurs, e.g., competition for space and nutrients, acidification of the environment, and production of antimicrobial metabolites (see Leverentz and others 2002). Of the antimicrobial metabolites produced by bacteria, bacteriocins have been particularly well studied. Nisin is the best known example of a commercially applied biocontrol system in low pH foods susceptible to contamination by L. monocytogenes and other Gram-positive bacteria such as Bacillus and Clostridium spp. Nisin has an unusually broad range of activity, hence its preferred status for food applications, but it is ineffective against Gram-negative bacteria. Antagonism can be achieved either by addition of GRAS protective cultures or directly via commercial bacteriocin preparations such as Nisaplin®. The former option is preferable because it offers a number of advantages including potential in situ production of antimicrobial metabolites (assuming growth conditions are met) and lower associated costs, and it may also alleviate labeling requirements. See Gálvez and others (2007) for a more complete review. However, optimizing either approach requires understanding of a number of complex aspects, including 1) the food matrix and related processing and storage conditions; 2) the composition, diversity, and quantity of the competing microflora; and 3) the susceptibility of the target species. Ideally, protective cultures should survive under typical storage conditions, grow at abuse temperatures without negatively impacting organoleptic characteristics and shelf life (Kostrzynska and Bachand 2006), and actively reduce populations of a broad range of pathogens (although the limitation of growth may be sufficient in some cases). Efficacious strains indigenous to the food in question are therefore of most interest. To this end, a number of groups have isolated and characterized antagonistic bacteria (predominantly lactococci, lactobacilli, and pediococci) from the microflora of various produce items and reported on their ability to inhibit produce-related pathogens (Bennik and others 1997; Cai and others 1997; Franz and others 1998; Kelly and others 1998; Liao and Fett 2001; Schuenzel and Harrison 2002; Wilderdyke and others 2004; Yildirim and Johnson 1998). It is evident from these reports that there is significant disparity in the spectrum of activity of antagonists against pathogens tested in vitro, particularly Gram-negative

210

Section III. Postharvest Interventions

pathogens that are arguably more likely to be hazards in fresh produce. For example, bacteriocinogenic LAB tested against Gram-negative pathogens showed no antagonistic activity (Kelly and others 1998; Yildirim and Johnson 1998), although Wilderdyke and others (2004) reported significant reductions in populations of Salmonella and E. coli O157:H7 by two LAB cultures (Lactococcus lactis subsp. lactis and Pediococcus acidilactici), primarily through acid production. Non-LAB such as pseudomonads and aeromonads have also been reported to inhibit Gram-negative bacteria (Schuenzel and Harrison 2002) but the genus Pseudomonas includes a number of spoilage and pathogenic species, and the status of Aeromonas hydrophila as a foodborne pathogen is becoming more widely accepted. These considerations highlight the need to assess carefully factors such as mechanism(s) of inhibition and safety and suitability when choosing protective cultures for food applications. While in vitro experiments are a useful starting point, it is often difficult to mimic accurately conditions relevant to the food under typical storage conditions. Consequently it is hard to predict how in vitro results will translate to in vivo conditions. For example, in research by Bennik and others (1999), bacteriocinogenic Enterococcus mundtii demonstrated in vitro growth inhibition or reduction of L. monocytogenes on sprout-derived vegetable agar stored at 8 °C over a 5-day period but could not prevent growth on fresh mung bean sprouts held under similar conditions, a situation attributed to proteolytic degradation. However the purified bacteriocin (mundticin) was shown to have some potential either as a washing solution or in a coating, although bacteriocin inactivation or resistance development was postulated to occur during extended storage. Many of the characterization papers cited previously did not consider activity in vivo, and have therefore been excluded from the summary provided in Table 11.1 describing reported applications of bacterial antagonists in fresh produce. These publications have focused exclusively on salad vegetables, melons, and sprouted seeds. Most studies have investigated activity against L. monocytogenes, which, although a debatable cause of significant fresh produce-associated disease, can grow on these foods and is susceptible to the action of bacteriocins. Reported reductions in L. monocytogenes levels using protective cultures only have been mixed—from no reduction up to 4 log10 CFU—reflecting the experimental variability in terms of cultures examined, storage conditions studied, and produce items employed. To improve efficacy, the Agriculture Research Service in the U.S. has been investigating the use of a Gluconobacter asaii protective culture in parallel with a mixture of six phages. This combination has been reported to reduce L. monocytogenes by up to 5 log10 CFU on melon, more than the individual treatments alone (Reynolds 2007). The approach offers both immediate and longer-term control of one pathogen via the combined action of phages and antagonistic bacteria, respectively, but could be modified to target other bacterial pathogens via the application of suitable phages. Expanding the protective culture approach to Gram-negative pathogens by the application of Pseudomonas cultures has also been reported in vivo by several groups (Liao 2008; Liao and Fett 2001; Matos and Garland 2005; Fett 2006). A summary of bacteriocin-specific applications to produce food safety has been compiled in Table 11.2. As before, the majority of studies have evaluated antagonistic activity against L. monocytogenes in association with salad vegetables, melons, and

Table 11.1. Reported applications of bacterial antagonists to improve the safety of fresh produce Antagonists Gluconobacter asaii Gluconobacter asaii Native microflora (baby carrots)

Pseudomonas fluorescens 2–79

Pseudomonas fluorescens 2–79 Pseudomonas fluorescens 2–79 Native microflora Lactobacillus casei LC34GF

211

Pseudomonas fluorescens A3, Yeast D1 Pediococcus parvulus, Enterococcus mundtii Lactobacillus lactis Enterococcus faecium Lactobacillus casei IMPC LC34 Native microflora Lactobacillus casei, Lactobacillus plantarum, Pediococcus spp. Lactobacillus casei, Pediococcus pentosaceus, Pediococcus spp.

Other Hurdles Bacteriophages

Target Organisms L. monocytogenes L. monocytogenes, Salmonella Poona L. monocytogenes, Y. enterocolitica, Salmonella enterica, E. coli O157:H7 S. Anatum, S. Infantis, S. Muenchen, S. Newport, S. Stanley S. Anatum, S. Infantis, S. Newport, S. Stanley S. Newport, S. Anatum, S. Stanley, S. Infantis L. monocytogenes Staph. aureus, L. monocytogenes, A. hydrophila, E. coli S. Chester, E. coli, L. monocytogenes L. monocytogenes

Tested Food Honeydew melon Apple plugs

Reference Reynolds 2007 Leverentz and others 2006

Baby carrot homogenate, bell peppers

Liao 2007

Alfalfa seed

Fett 2006

Alfalfa seed

Matos and Garland 2005

Sprouting alfalfa seeds

Liao 2008

Cantaloupe melon Scarola salad leaves

Ukuku and others 2004 Scolari and Vescovo 2004

Green peppers

Liao and Fett 2001

Mungbean sprouts

Bennik and others 1999

L. monocytogenes

Fresh-cut Caesar salad

Cai and others 1997

A. hydrophila L. monocytogenes Staph. aureus, L. monocytogenes, A. hydrophila, S. Typhimurium

Mixed salad vegetables Endive leaves Vegetable salads, vegetable juice

Torriani and others 1997 Carlin and others 1996 Vescovo and others 1996

Coliforms, faecal coliforms, enterococci

Vegetable salad

Vescovo and others 1995

212 Table 11.2. Reported applications of bacterial metabolites to improve the safety of fresh produce Metabolites LAB culture supernatants LAB

Other Hurdles

Enterocin AS-48

Various chemical preservatives

Nisin

Hydrogen peroxide, sodium lactate, citric acid Sodium lactate, citric acid, phytic acid, EDTA, potassium sorbate EDTA, sodium lactate, potassium sorbate

Nisin and Pediocin

Nisin Colicin Hu194 Mundticin Culture permeate (L. casei)

Target Organisms L. monocytogenes L. monocytogenes, S. Typhimurium, E. coli L. monocytogenes L. monocytogenes, E. coli O157:H7

Tested Food Lettuce Wounded apples, cut lettuce

Reference Allende and others 2007 Trias and others 2008

Alfalfa, soybean sprouts, green asparagus Cantaloupe, honeydew melons

Molinos and others 2005 Ukuku and others 2005

L. monocytogenes

Cabbage, broccoli, mung bean sprouts

Bari and others 2005

S. Poona, S. Newport, S. Stanley, S. Anatum, S. Infantis E. coli O157:H7 L. monocytogenes A. hydrophila

Cantaloupe melon

Ukuku and Fett 2004

Alfalfa seeds Mungbean sprouts Mixed salad vegetables

Nandiwada and others 2004 Bennik and others 1999 Torriani and others 1997

Biological Control of Human Pathogens

213

sprouted seeds. Results published by Molinos and others (2005) in particular demonstrate the variability in achieved pathogen reductions in relation to produce type, storage temperature, and concentration of bacteriocin employed. The application of defined antimicrobials does lend itself more easily to the option of combined strategies, either to improve efficacy or extend the biocontrol approach to Gram-negative pathogens. Various chelating agents and other antimicrobials have been successfully employed in combination with nisin to enhance activity against Salmonella spp. (Ukuku and Fett 2004), E. coli O157:H7, and L. monocytogenes (Ukuku and others 2005). Alternatively, using colicins produced by E. coli has been shown to be effective in treating alfalfa seeds contaminated with E. coli O157:H7, with reductions of 4 to >6 log10 CFU observed within 24 to 48 hours of treatment (Nandiwada and others 2004). As previously mentioned, the development of resistance to bacteriocins during extended exposure is a potential problem. However, reported means of dealing with this include combining bacteriocins (Bari and others 2005) or alternatively applying bacteriocins in combination with bacteriocin-producing cultures (reviewed by Settanni and Corsetti 2008). The latter approach has also been shown to enhance overall activity and reduce the amount of nisin required. The ideal protective culture or bacteriocin must be able to target a range of pathogens at concentrations that do not influence the organoleptic properties or shelf life of the product, without susceptibility to phage attack (in the case of protective cultures) or the development of resistance (to bacteriocins). Given that protective cultures and antimicrobial metabolites are not in their own right a complete biocontrol approach, additional hurdles must be considered. The application of multiple bacteriocinogenic strains (both LAB and non-LAB) in combination with defined bacteriocins and other antimicrobial agents appears to offer one means of achieving biocontrol targets. Yeast and Fungi Isolates of yeast and fungi have been extensively evaluated for the biocontrol of postharvest diseases of produce. The yeast Candida oleophilia Montrocher, isolated from tomato peel, has also been registered with the EPA under the trade name Aspire since 1995 for the control of fungal citrus postharvest diseases (Richards and others 2004). Although there are relatively few reports of fungi or yeast being used for the biocontrol of human pathogens, they have many features that make them good candidates for this purpose. Yeasts are frequent colonizers of vegetables and fruits, with up to 106 cells per apple reported (Deak and Beuchat 1997), have simple nutritional requirements, grow easily in fermenters, and survive in a wide range of environments (Bleve and others 2006). Yeast and fungal biocontrol of plant pathogens has been found to be complex and involve multiple mechanisms, including competition with the pathogen for space and limiting nutrients; induction of host plant responses; and production of killer toxins, lytic enzymes, or iron-chelating siderophores (Calvente and others 1999; Chan and Tian 2005; Ippolito and others 2000; Lowes and others 2000). Some of these mechanisms have also been shown to be involved in the control of human pathogens. Leverentz and others (2006) tested yeast biocontrol of L. monocytogenes and Salmonella Poona on fresh-cut apples. Metschnikowia pulcherrima, Candida, and

214

Section III. Postharvest Interventions

Discospaerina fagi showed activity against L. monocytogenes, and D. fagi and M. pulcherrima activity against S. Poona. For L. monocytogenes at 10 °C, a 2.1–2.8-log10 reduction after 5 days was observed compared to the control, and at 25 °C up to a 6-log10 reduction occurred in 7 days. No antagonists were effective against Salmonella at 10 °C, but D. fagi (4.8-log10 reduction after 7 days incubation) and M. pulcherrima (2.4 log10) reduced pathogen numbers at 25 °C. Preliminary evidence also suggested that combining phage cocktails, which have a quick knockdown, with this slower effect might be beneficial for controlling pathogens over the long term, reinforcing similar observations made in the previous section. Application of mixed cultures of bacteria and yeasts, or yeasts and molds, may also present a useful strategy. During a survey for antagonistic microflora on produce, yeast strain D1 was isolated from prepeeled baby carrots and inhibited L. monocytogenes, E. coli, S. Chester, and Erwinia carotovora in vitro (Liao 2007). However, the combination of P. fluorescens A3 and yeast D1 was found to be a more effective treatment, reducing S. Chester by 1 log10 and L. monocytogenes by 2 log10 compared to controls when tested on green pepper discs over 3 days. Richards and others (2004) found that Candida oleophila and Rhodotorula glutinis at 4 °C (<1-log10 reduction), and C. albidus, Debaryomyces hansenii and Pichia guilliermondii at 20 °C (1.2–2.7-log10 reduction) could be effective antagonists to S. Poona in cantaloupe wounds when they were coinfected by the plant pathogen Geotrichum candidum. Some species of plant pathogenic molds produce enzymes that release breakdown products into the surrounding plant tissue causing localized changes in pH. This can significantly change the suitability of the niche for enteric pathogens and therefore may be useful for biocontrol purposes. It is known that some molds increase the pH of fruits and promote the growth of pathogens (Draughon and others 1988; Riordan and others 2000). However, on fresh-cut apple slices Penicillium expansum was demonstrated to lower the pH from 4.7 to 3.7 and cause a significant decline, up to 4 log10, in numbers of coinoculated Listeria (Conway and others 2000). Another study revealed that yeasts could negate increases in the pH of cantaloupe fruit wounds but that this was insufficient to control S. Poona when coinoculated (Richards and others 2004). There have been some indications (Izgu and Altinbay 1997) that yeast killer toxins, generally only thought to target yeasts and some fungi, might also inhibit Grampositive bacteria including Listeria but these results remain largely unsubstantiated. More recently, Goerges and others (2006) screened 304 yeasts and found 11 that could significantly inhibit two or more L. monocytogenes isolates in vitro. Ten of these yeasts were Candida intermedia and the other was Yarrowia lipolytica. A selection of 14 yeasts produced between 1- and 4-log10 reduction compared to the control in a 24-hour cocultivation assay, the strongest inhibition being caused by a C. intermedia isolate. An assay for killer toxins detected only very small clearing zones, and their presence did not correlate well with results from other assays. So, although it remains possible that killer toxins may be influential in yeast-pathogen interactions in produce, substantially more work needs to be done before any conclusions can be drawn. In grapes and other produce, the production of ochratoxins by some Aspergillus and Penicillium species is of particular concern because these compounds are nephrotoxic and carcinogenic to animals. Bleve and others (2006) collected epiphytic yeast from the berry surface of Vitis vinifera cv. Negroamaro and tested 144 unique isolates

Biological Control of Human Pathogens

215

against ochratoxin producing Asper. niger and Asper. carbonarius. Six strong inhibitors of Aspergillus spp. were identified: Metschnikowia pulcherrima, Issatchenkia orientalis (two isolates), I. terricola, Kluyveromyces thermotolerans, and Candida incommunis. These isolates were applied to wounded grape berry surfaces with the ochratoxigenic molds. When inoculated at 109 CFU/wound, all yeasts except K. thermotolerans reduced conidiophore production (a measure of growth). I. orientalis demonstrated the best inhibition, possibly due to the production of a secondary metabolite with antifungal properties. Another biocontrol yeast, Rhodotorula glutinis (Castoria and others 2005), has the ability to metabolize another mycotoxin, Patulin. Biologically Derived Chemicals Essential Oils and Plant Extracts Plant extracts have several potential properties, including biodegradability, safety, cost competitiveness, and efficacy that make them attractive candidates for use in controlling microorganisms on produce. Essential oils (EO) are very complex organic extracts usually obtained by steam distillation and are composed of many individual components, of which two or three are usually responsible for the key biological properties of the oil (Bakkali and others 2008). The most frequently described group of inhibitory compounds are phenolic-based, with aromatic and isothiocyanate-based components also isolated. Synergy between individual components from the same EO is possible because fractions are sometimes less efficacious than whole extracts. The mechanisms of action of EO are not fully understood, but it is known that phenolic components appear to integrate into the cytoplasmic membrane of the bacterium, causing increased proton and potassium permeability (Perez-Conesa and others 2006). The EO component carvacrol has also been shown to induce heat-shock proteins and prevent the development of flagella in E. coli O157:H7 (Burt and others 2007). EO have been reported to be active in vitro against a wide range of foodborne pathogens including E. coli O157:H7, Salmonella, Shigella, Campylobacter, L. monocytogenes, Staph. aureus, B. cereus, and Vibrio spp. The simplest application method for EO on produce is their incorporation into aqueous washes or dips. The washing of shredded lettuce in 0.1% thyme oil was more effective than aqueous chlorine dioxide and ozonated water for decontamination of E. coli O157:H7, and combining these treatments in sequential washes was able to deliver a 3.5-log10 reduction on lettuce and baby carrots (Singh and others 2002). The same sequential wash treatment gave a 4-log10 reduction in E. coli O157:H7 on alfalfa sprout seeds, but complete eradication was not achieved because during sprout germination pathogen numbers increased to high levels (Singh and others 2003). Sanitation of separated lettuce leaves with 0.05% solutions of carvacrol, thymol, and thyme essential oil has also achieved a 2-log10 reduction in Shigella spp. (Bagamboula and others 2004). Dip treatment of fresh-cut fruit with EO at refrigeration temperatures has shown promise for pathogen reduction, with kiwifruit and melon pieces treated with 1 mM carvacrol or cinnamic acid having levels of endogenous flora up to 4 log10 lower than controls (Roller and Seedhar 2002). A useful property of some EO is their bioactivity in the vapor phase, because aqueous decontamination of certain types of fresh produce, including fruits and sprouts, can be difficult. Exposure of fruit to EO vapor can also improve some

216

Section III. Postharvest Interventions

quality-related attributes, but care is needed in their application because continuous fumigation can lead to taint, with thin-skinned products more prone than those that are thicker-skinned (Tzortzakis 2007). EO vapors have shown potential for the decontamination of cabbage leaves. Citral and linalool vapors produced 2- to 6-log10 reductions of L. monocytogenes, Staph. aureus, B. cereus, and Arcobacter spp. after 8–24 h, and bergamot vapors had a similar effect but did not affect Staph. aureus (Fisher and Phillips 2006; Fisher and others 2007). Another EO volatile compound allyl isothiocyanate (AIT), derived from cruciferous plants, has been tested for the elimination of E. coli O157:H7 on wet and dry alfalfa seeds with mixed results. Greater than a 2-log10 reduction (inoculum 2.7 log10) was achieved on wet seeds exposed to AIT for 24 h at 37 or 47 °C, although loss of seed viability was a problem. On dry seeds the effects were not observed (Park and others 2000). An effect against Salmonella was, however, noted for treatment of alfalfa seeds by thymol at 60 °C for 7 hours (Weissinger and others 2001). Essential oil vapors also have the potential to be used in combination with modified atmosphere packaging (MAP) to enhance both product safety and quality. For example, eugenol, thymol, and carvacrol vapors have been tested separately and in combination for antimicrobial properties on MAP-packed table grapes. Around a 2-log10 reduction in both yeasts and molds and aerobic mesophiles was achieved. The combination of multiple EO allowed reduced quantities to be used, and odors associated with the EOs were found to dissipate rapidly upon opening of the seal (Guillen and others 2007; Valero and others 2006). Similar results were achieved in MAP sweet cherries (Serrano and others 2005). Other plant materials with potential for biocontrol use in produce include extracts from berries, seeds, and tea. The skins, juice, and seed extracts of grapes have been found to be inhibitory to L. monocytogenes, E. coli, Salmonella Poona, and B. cereus in vitro (Rhodes and others 2006; Serra and others 2008). Both Gram-negative and Gram-positive foodborne pathogens were inhibited by addition of Nordic berry extracts to growth media, with cloudberry and raspberry the best inhibitors (PuupponenPimia and others 2005). Whole tea infusions and flavonoid extracts were inhibitory when applied to B. cereus cultures on solidified media (Friedman and others 2006). Grapefruit seed extract has been tested in combination with nisin, citric acid, sodium lactate, and potassium sorbate on fresh-cut cucumber and lettuce. Either alone or in combination with nisin and citric acid, grapefruit seed extract caused significant inhibition (>2 log10) of Salmonella and Listeria on the vegetables (inoculum 4 log10) and prolonged the sensory quality of the product (Xu and others 2007). A completely different approach to elimination of pathogens on produce may be to stimulate the innate pathogen defense mechanisms of the plant tissues by adding extracted plant signaling compounds. Methyl jasmonate is known to affect plant growth and development and is involved in responses to pathogen attack and wounding. When applied as a vapor to fresh-cut celery at 10 °C, a 3-log10 reduction in the native flora was achieved and the shelf life extended. Similar effects were observed on green pepper strips when applied as a dip (Buta and Moline 1998). Methyl jasmonate vapors and dips were also effective in reducing numbers of endogenous flora on fresh-cut pineapple chunks by 3 log10 when in a sealed container (Martinez-Ferrer and Harper 2005). Other signaling compounds of interest include the plant defense

Biological Control of Human Pathogens

217

compounds hexanal and 2-(E)-hexenal, which when added to apple slices in MAP demonstrated activity against L. monocytogenes, E. coli O157:H7, and Salmonella Enteritidis (Lanciotti and others 2003). In addition the wounding response compounds (E,Z)-2,6-nonadienal and (E)-2-nonenal from cucumber were inhibitory to B. cereus, E. coli O157:H7, L. monocytogenes, and Salmonella Typhimurium when added to growth media (Cho and others 2004). The stress-induced compound 6,7-dimethoxycoumarin, which was isolated from Valencia oranges (Citrus sinesis cv. Valencia), was shown to control the production of aflatoxins (Mohanlall and Odhav 2006) and it was suggested that its use could be extended to the postharvest control of mycotoxins in stored grains. Quorum Sensing Signaling Molecules Bacteria present in an ecosystem communicate with one another using chemical signals (autoinducers) to regulate gene expression. This process, known as quorum sensing (QS), has been studied in depth for plant pathogens (von Bodman and others 2003). QS is used by bacteria to regulate a number of cellular responses, including sporulation, biofilm formation, bacteriocin production, and motility. QS signals have been shown to be involved with the microbial spoilage of bean sprouts (Rasch and others 2005). Autoinducers have also been detected on tomato, cantaloupe, and carrots (Lu and others 2004), and on 11 of 12 produce items tested in another study (Lu and others 2005). Similarly, Yersinia enterocolitica produced signals in vegetable simulating agar (Medina-Martínez and others 2007). QS in foodrelated bacteria has recently been reviewed (Gobbetti and others 2007). The possibility of interfering with this communication for food safety purposes has been proposed (Smith and others 2004) but is still in the early stages of development. For example, signal antagonists might be used to interrupt QS signals to interfere with biofilm formation and virulence. Such QS inhibitors (possibly signal-degrading enzymes or mimics) have been identified in a number of produce extracts (Rasmussen and others 2005) indicating that this mechanism forms part of the plant’s defense against bacterial attack. Biocontrol agents might also be used as defense barriers in the form of biofilms. Biofilm formation and production of surfactin by B. subtilis have been shown to be effective in preventing the invasion of Arabidopsis by the plant pathogen P. syringae (Bais and others 2004). Such observations suggest a possible approach to prevent the colonization of edible plants by human pathogens. Siderophores Siderophores are low–molecular-weight, ferric-chelating molecules that may enhance the effectiveness of biocontrol agents on produce through competition for iron. Many antagonistic bacteria, such as pseudomonads and aeromoads, that have the potential to produce siderophores have been isolated from produce (Schuenzel and Harrison 2002). Calvente and others (1999) isolated two strains of the yeast Rhodotorula glutinis from apples that produced the siderophore rhodotorulic acid. Biocontrol of Penicillium expansum on wounded apples was found to be more effective with a combination of the yeast and siderophore than the yeast alone (6% and 34% disease prevalence, respectively). Sansone and others (2005) also evaluated

218

Section III. Postharvest Interventions

Rhodotorula glutinis and rhodotorulic acid for control of Botrytis cinerea on apple wounds and found R. glutinis reduced disease severity by 54% alone, and by 75% in combination with rhodotorulic acid. In both instances, it is thought that the siderophore inhibited conidial germination because large amounts of iron are required for this to occur. The combined use of siderophores and yeasts on fresh-cut produce may facilitate a more rapid colonization of available niches and reduce the probability of pathogen colonization at these sites. Lysozyme Lysozyme is an enzyme that degrades bacterial cell walls and leads to cell lysis. These enzymes are produced by a wide range of prokaryotes and eukaryotes and form a passive defense against bacterial colonization. Some of these enzymes also have antifungal activity (Wang and others 2005). Activity is optimal against Gram-positive bacteria, but under high barometric pressure Gram-negative bacteria can also become sensitive to lysozymes (Nakimbugwe and others 2006b). This has been demonstrated with phage lambda lysozyme in banana juice (Nakimbugwe and others 2006a). The use of lysozyme, possibly in enzyme cocktails, has been suggested for pathogen control. Polylysine Culture filtrates of a Streptomyces sp. contain the antimicrobial compound ε-Polylysine, a linear polymer of 25–30 L-lysine residues. A commercially available preparation of this compound reduced the numbers of E. coli O157:H7, Salmonella Typhimurium and L. monocytogenes in a blended mixture of cauliflower and broccoli, although the vegetables had been autoclaved prior to use (Geornaras and others 2007). In all three cases pathogens were able to grow on the substrate, but in the presence of ε-Polylysine their numbers declined to become nondetectable after 2 to 4 days of incubation at 12 °C. This compound is considered GRAS for use in rice.

Conclusions The information presented here illustrates the wide diversity of potential biocontrol approaches that are available. They represent a toolbox of controls that can be applied across the farm-to-fork spectrum. Examples have been provided whereby these controls have been used in combination, an approach which would seem likely to present the best possibility for success. The application of multiple controls as hurdles means that each individual control need not inflict a huge reduction in pathogen numbers, comparable to the 12D “bot cook” for low-acid canned foods, but the combined effect is satisfactory control of the pathogen. In addition the use of multiple controls using different mechanisms reduces the likelihood of resistant pathogens emerging. None of these controls will be of use unless they are approved by the regulators. However, when provided with adequate information, regulatory approval has been readily forthcoming, at least in the area of phage biocontrol. The advantage of these controls is that they are by definition derived from nature, which should be appealing to the consumer, given the dislike for synthetic “chemical” controls, and should make the prospect of obtaining regulatory approval more likely.

Biological Control of Human Pathogens

219

An important factor in the successful application of biocontrol is an excellent understanding of the interactions of the plant, plant pathogens, human pathogens, and commensal organisms. Progress in the field can be optimized through comprehensive studies accounting for these areas. Work to date suggests that investment in such research would yield results of benefit to both the industry and public health.

References Allende A, Martínez B, Selma V, Gil MI, Suárez JE, Rodríguez A. 2007. Growth and bacteriocin production by lactic acid bacteria in vegetable broth and their effectiveness at reducing Listeria monocytogenes in vitro and in fresh-cut lettuce. Food Microbiol 24(7–8):759–766. Anonymous. 2006. FDA approves Intralytix’s food safety product (LMP 102). www.intralytix.com/Intral_ News_PR081906.htm. Accessed on 12 December 2007. ———. 2007. FDA extends GRAS approval LISTEX(™) to all food products. www.ecplaza.net/news/news. do?cmd=print&no=1960. Accessed on 19 November 2007. Babic I, Watada AE, Buta JG. 1997. Growth of Listeria monocytogenes restricted by native microorganisms and other properties of fresh-cut spinach. J Food Prot 60(8):912–917. Bagamboula CF, Uyttendaele M, Debevere J. 2004. Inhibitory effect of thyme and basil essential oils, carvacrol, thymol, estragol, linalool and p-cymene towards Shigella sonnei and S. flexneri. Food Microbiol 21(1):33–42. Bais HP, Fall R, Vivanco JM. 2004. Biocontrol of Bacillus subtilis against infection of Arabidopsis roots by Pseudomonas syringae is facilitated by biofilm formation and surfactin production. Plant Physiol 134(1):307–319. Bakkali F, Averbeck S, Averbeck D, Idaomar M. 2008. Biological effects of essential oils—A review. Food Chem Toxicol 46(2):446–475. Balogh B, Jones JB, Momol MT, Obradovic A, King P, Jackson LE. 2003. Improved efficacy of newly formulated bacteriophages for management of bacterial spot on tomato. Plant Dis 87(8):949–954. Bari ML, Ukuku DO, Kawasaki T, Inatsu Y, Isshiki K, Kawamoto S. 2005. Combined efficacy of nisin and pediocin with sodium lactacte, citric acid, phytic acid, and potassium sorbate and EDTA in reducing the Listeria monocytogenes population of inoculated fresh-cut produce. J Food Prot 68(7):1381–1387. Bennik MHJ, Smid EJ, Gorris LGM. 1997. Vegetable-associated Pediococcus parvulus produces pediocin PA-1. Appl Environ Microbiol 63(5):2074–2076. Bennik MHJ, van Overbeek W, Smid EJ, Gorris LGM. 1999. Biopreservation in modified atmosphere stored mungbean sprouts: the use of vegetable-associated bacteriocinogenic lactic acid bacteria to control the growth of Listeria monocytogenes. Lett Appl Microbiol 28(3):226–232. Beuchat LR. 2002. Ecological factors influencing survival and growth of human pathogens on raw fruits and vegetables. Microb Infect 4(4):413–423. Bleve G, Grieco F, Cozzi G, Logrieco A, Visconti A. 2006. Isolation of epiphytic yeasts with potential for biocontrol of Aspergillus carbonarius and A. niger on grape. Int J Food Microbiol 108(2):204–209. Breidt F, Fleming HP. 1997. Using lactic acid bacteria to improve the safety of minimally processed fruits and vegetables. Food Technol 51(9):44–49. Burnett SL, Chen J, Beuchat LR. 2000. Attachment of Escherichia coli O157:H7 to the surfaces and internal structures of apples as detected by confocal scanning laser microscopy. J Food Prot 66(11):4679– 4687. Burt SA, van der Zee R, Koets AP, de Graaff AM, van Knapen F, Gaastra W, Haagsman HP, Veldhuizen EJA. 2007. Carvacrol induces heat shock protein 60 and inhibits synthesis of flagellin in Escherichia coli O157:H7. Appl Environ Microbiol 73(14):4484–4490. Buta JG, Moline HE. 1998. Methyl jasmonate extends shelf life and reduces microbial contamination of fresh-cut celery and peppers. J Agric Food Chem 46(4):1253–1256. Cai Y, Ng L-K, Farber JM. 1997. Isolation and characterization of nisin-producing Lactococcus lactis subsp. lactis from bean-sprouts. J Appl Microbiol 83(4):499–507. Calvente V, Benuzzi D, de Tosetti MIS. 1999. Antagonistic action of siderophores from Rhodotorula glutinis upon the postharvest pathogen Penicillium expansum. Int Biodet Biodeg 43(4):167–172.

220

Section III. Postharvest Interventions

Carlin F, Nguyen-the C, Morris CE. 1996. Influence of background microflora on Listeria monocytogenes on minimally processed fresh broad-leaved endive (Cichorium endivia var. latifolia). J Food Prot 59(7):698–703. Castoria R, Morena V, Caputo L, Panfili G, De Curtis F, De Cicco V. 2005. Effect of the biocontrol yeast Rhodotorula glutinis strain LS11 on patulin accumulation in stored apples. Phytopathol 95(11):1271–1278. Chan Z, Tian S. 2005. Interaction of antagonistic yeasts against postharvest pathogens of apple fruit and possible mode of action. Postharvest Biol Technol 36(2):215–223. Cho MJ, Buescher RW, Johnson M, Janes M. 2004. Inactivation of pathogenic bacteria by cucumber volatiles (E,Z)-2,6-nonadienal and (E)-2-nonenal. J Food Prot 67(5):1014–1016. Conway WS, Leverentz B, Saftner RA, Janisiewicz WJ, Sams CE, Leblanc E. 2000. Survival and growth of Listeria monocytogenes on fresh-cut apple slices and its interaction with Glomerella cingulata and Penicillium expansum. Plant Dis 84(2):177–181. Cooley MB, Miller WG, Mandrell RE. 2003. Colonization of Arabidopsis thalania with Salmonella enterica and enterohaemorrhagic Escherichia coli O157:H7 and competition by Enterobacter asburiae. Appl Environ Microbiol 69(8):4915–4926. Deak T, Beuchat LR. 1997. Handbook of Food Spoilage Yeasts. New York: CRC Press. 224 p. Draughon FA, Chen S, Mundt JO. 1988. Metabiotic association of Fusarium, Alternaria, and Rhizoctonia with Clostridium botulinum in fresh tomatoes. J Food Sci 53(1):120–123. Fett WF. 2000. Naturally occurring biofilms in alfalfa and other types of sprouts. J Food Prot 63(5):625–632. ———. 2006. Inhibition of Salmonella enterica by plant-associated pseudomonads in vitro and on sprouting alfalfa seeds. J Food Prot 69(4):719–728. Fisher K, Phillips CA. 2006. The effect of lemon, orange and bergamot essential oils and their components on the survival of Campylobacter jejuni, Escherichia coli O157, Listeria monocytogenes, Bacillus cereus and Staphylococcus aureus in vitro and in food systems. J Appl Microbiol 101(6):1232–1240. Fisher K, Rowe C, Phillips CA. 2007. The survival of three strains of Arcobacter butzleri in the presence of lemon, orange and bergamot essential oils and their components in vitro and on food. Lett Appl Microbiol 44(5):495–499. Food Standards Agency. 2007. Consumer attitudes to food standards. http://www.food.gov.uk/multimedia/ pdfs/cas07uk.pdf. Accessed on 15 October 2008. Franz CMAP, Du Toit M, Olasupo NA, Schillinger U, Holzapfel WH. 1998. Plantaricin D, a bacteriocin produced by Lactobacillus plantarum BFE 905 from ready-to-eat salad. Lett Appl Microbiol 26(3):231–235. Friedman M, Henika PR, Levin CE, Mandrell RE, Kozukue N. 2006. Antimicrobial activities of tea catechins and theaflavins and tea extracts against Bacillus cereus. J Food Prot 69(2):354–361. Gálvez A, Abriouel H, Lucas López R, Ben Omar N. 2007. Bacteriocin-based strategies for food biopreservation. Int J Food Microbiol 120(1–2):51–70. Geornaras I, Yoon Y, Belk KE, Smith GC, Sofos JN. 2007. Antimicrobial activity of ε-polylysine against Escherichia coli O157:H7, Salmonella Typhimurium, and Listeria monocytogenes in various food extracts. J Food Sci 72(8):M330–M334. Gill J, Abedon ST. 2003. Bacteriophage ecology and plants. www.apsnet.org/online/feature/phages/. Accessed on 20 November 2007. Gobbetti M, De Angelis M, Di Cagno R, Minervini F, Limitone A. 2007. Cell-cell communication in food related bacteria. Int J Food Microbiol 120(1–2):34–45. Goerges S, Aigner U, Silakowski B, Scherer S. 2006. Inhibition of Listeria monocytogenes by food-borne yeasts. Appl Environ Microbiol 72(1):313–318. Gould GW. 1996. Industry perspectives on the use of natural antimicrobials and inhibitors for food applications. J Food Prot (Supp) 82–86. Greer GG. 2005. Bacteriophage control of foodborne bacteria. J Food Prot 68(5):1102–1111. Guillen F, Zapata PJ, Martinez-Romero D, Castillo S, Serrano M, Valero D. 2007. Improvement of the overall quality of table grapes stored under modified atmosphere packaging in combination with natural antimicrobial compounds. J Food Sci 72(3):S185–S190. Hudson JA, Billington C, Carey-Smith G, Greening G. 2005. Bacteriophages as biocontrol agents in food. J. Food Prot 68(2):426–437.

Biological Control of Human Pathogens

221

Hughes KA, Sutherland IW, Jones MV. 1998. Biofilm susceptibility to bacteriophage attack: the role of phage-borne polysaccharide depolymerase. Microbiology 144(11):3039–3047. Ippolito A, El Ghaouth A, Wilson CL, Wisniewski M. 2000. Control of postharvest decay of apple fruit by Aureobasidium pullulans and induction of defense responses. Postharvest Biol Technol 19(3):265–272. Iriarte FB, Balogh B, Momol MT, Smith LM, Wilson M, Jones JB. 2007. Factors affecting survival of bacteriophage on tomato leaf surfaces. Appl Environ Microbiol 73(6):1704–1711. Iturriaga MH, Tamplin ML, Escartín EF. 2007. Colonization of tomatoes by Salmonella Montevideo is affected by relative humidity and storage temperature. J Food Prot 70(1):30–34. Izgu F, Altinbay D. 1997. Killer toxins of certain yeast strains have potential growth inhibitory activity on Gram-positive pathogenic bacteria. Microbios 89(358):15–22. Jablasone J, Warriner K, Griffiths M. 2005. Interactions of Escherichia coli O157:H7, Salmonella and Listeria monocytogenes plants cultivated in a gnotobiotic system. Int J Food Microbiol 99(1):7–18. Janisiewicz WJ, Conway WS, Leverentz B. 1999. Biological control of postharvest decays of apple can prevent growth of Escherichia coli O157:H7 in apple wounds. J Food Prot 62(12):1372–1375. Kelly WJ, Davey GP, Ward LJH. 1998. Characterization of lactococci isolated from minimally processed fruits and vegetables. Int J Food Microbiol 45(2):85–92. Kennedy JE, Wei CI, Oblinger JL. 1986. Distribution of coliphages in various foods. J Food Prot 49(12):944–951. Kim W-S, Salm H, Geider K. 2004. Expression of the bacteriophage ϕEa1h lysozyme in Escherichia coli and its activity in growth inhibition of Erwinia amylovora Microbiology 150(8):2707–2714. Kostrzynska M, Bachand A. 2006. Use of microbial antagonism to reduce pathogen levels on produce and meat products: a review. Can J Microbiol 52(11):1017–1026. Lanciotti R, Belletti N, Patrignani F, Gianotti A, Gardini F, Guerzoni ME. 2003. Application of hexanal, (E)-2-hexenal, and hexyl acetate to improve the safety of fresh-sliced apples. J Agric Food Chem 51(10):2958–2963. Lapidot A, Romling U, Yaron S. 2006. Biofilm formation and the survival of Salmonella Typhimurium on parsley. Int J Food Microbiol 109(3):229–233. Lee C-N, Lin J-W, Chow T-Y, Tseng Y-H, Weng S-F. 2006. A novel lysozyme from Xanthomonas oryzae phage φXo411 active against Xanthomonas and Stenotrophomonas. Prot Expr Purif 50(2):279–237. Leverentz B, Conway WS, Alavidze Z, Janisiewicz WJ, Fuchs Y, Camp MJ, Chighladze E, Sulakvelidze A. 2001. Examination of bacteriophage as a biocontrol method for Salmonella on fresh-cut fruit: a model study. J Food Prot 64(8):1116–1121. Leverentz B, Conway WS, Camp MJ, Janisiewicz WJ, Abuladze T, Yang M, Saftner R, Sulakvelidze A. 2003. Biocontrol of Listeria monocytogenes on fresh-cut produce by treatment with lytic bacteriophages and a bacteriocin. Appl Environ Microbiol 69(8):4519–4526. Leverentz B, Conway WS, Janisiewicz W, Abadias M, Kurtzman CP, Camp MJ. 2006. Biocontrol of the food-borne pathogens Listeria monocytogenes and Salmonella enterica serovar Poona on fresh-cut apples with naturally occurring bacterial and yeast antagonists. Appl Environ Microbiol 72(2):1135–1140. Leverentz B, Conway WS, Janisiewicz WJ, Camp MJ. 2004. Optimizing concentration and timing of a phage spray application to reduce Listeria monocytogenes on honeydew melon tissue. J Food Prot 67(8):1682–1686. Leverentz B, Janisiewicz WJ, Conway WS. 2002. Biological control of minimally processed fruits and vegetables. In: JS Novak, JG Saper, VK Juneja (ed). Microbial Safety of Minimally Processed Foods. New York: CRC Press. pp. 319–332. Liao C-H. 2007. Inhibition of foodborne pathogens by native microflora recovered from fresh peeled baby carrot and propagated in cultures. J Food Sci 72(4):M134–M139. ———. 2008. Growth of Salmonella on sprouting alfalfa seeds as affected by the inoculum size, native microbial load and Pseudomonas fluorescens 2–79. Lett Appl Microbiol 46(1):232–236. Liao C-H, Fett WF. 2001. Analysis of native microflora and selection of strains antagonistic to human pathogens on fresh produce. J Food Prot 64(8):1110–1115. Liao C-H, McEvoy JL, Smith JL. 2003. Control of bacterial soft rot and foodborne human pathogens on fresh fruits and vegetables. In: H-C Huang, SN Acharya (ed). Advances in Plant Disease Management. Kerala, India: Research Signpost. pp. 165–193.

222

Section III. Postharvest Interventions

Lowes KF, Shearman CA, Payne J, MacKenzie D, Archer DB, Merry RJ, Gasson MJ. 2000. Prevention of yeast spoilage in feed and food by the yeast mycocin HMK. Appl Environ Microbiol 66(3):1066–1076. Lu L, Hume ME, Pillai SD. 2004. Autoinducer-2–like activity associated with foods and its interaction with food additives. J Food Prot 67(7):1457–1462. ———. 2005. Autoinducer-2–like activity on vegetable produce and its potential involvement in biofilm formation on tomatoes. Foodborne Path Dis 2(3):242–249. Lu TK, Collins JJ. 2007. Dispersing biofilms with engineered enzymatic bacteriophage. Proc Nat Acad Sci 104(27):11197–11202. Martinez-Ferrer M, Harper C. 2005. Reduction in microbial growth and improvement of storage quality in fresh-cut pineapple after methyl jasmonate treatment. J Food Qual 28(1):3–12. Matos A, Garland JL. 2005. Effects of community versus single strain inoculants on the biocontrol of Salmonella and microbial community dynamics in alfalfa sprouts. J Food Prot 68(1):40–48. Medina-Martínez MS, Uyttendaele M, Meireman S, Debevere J. 2007. Relevance of N-acyl-L-homoserine lactone production by Yersinia enterocolitica in fresh foods. J Appl Microbiol 102(4):1150–1158. Mohanlall V, Odhav B. 2006. Biocontrol of aflatoxins B1, B2, G1, G2 and fumonisin B1 with 6,7dimethoxycoumarin, a phytoalexin from Citrus sinensis. J Food Prot 69(9):2224–2229. Molinos AC, Abriouel H, Ben Omar N, Valdivia E, Lucas López R, Maqueda M, Martínez Cañamero M, Gálvez A. 2005. Effect of immersion solutions containing Enterocin AS-48 on Listeria monocytogenes in vegetable foods. Appl Environ Microbiol 71(12):7781–7787. Nakimbugwe D, Masschalck B, Anim G, Mochiels CW. 2006a. Inactivation of Gram-negative bacteria in milk and banana juice by hen egg white and lambda lysozyme under high hydrostatic pressure. Int J Food Microbiol 112(1):19–25. Nakimbugwe D, Masschalck B, Atanassova M, Zewdie-Bosüner A, Michiels CW. 2006b. Comparison of bactericidal activity of six lysozymes at atmospheric pressure and under high hydrostatic pressure. Int J Food Microbiol 108(3):355–363. Nandiwada LS, Schamberger GP, Schafer HW, Diez-Gonzalez F. 2004. Characterization of an E2-type colicin and its application to treat alfalfa seeds to reduce Escherichia coli O157:H7. Int J Food Microbiol 93(3):267–279. Pao S, Randolph SP, Westbrook EW, Shen H. 2004. Use of bacteriophages to control Salmonella in experimentally contaminated sprout seeds. J Food Sci 69(5):M127–M130. Paradis-Bleau C, Cloutier I, Lemieux L, Sanschagrin F, Laroche J, Auger M, Garnier A, Levesque RC. 2007. Peptidoglycan lytic activity of the Pseudomonas aeruginosa phage φKZ gp144 lytic transglycosylase. FEMS Microbiol Lett 266(2):201–209. Park CM, Taormina PJ, Beuchat LR. 2000. Efficacy of allyl isothiocyanate in killing enterohemorrhagic Escherichia coli O157:H7 on alfalfa seeds. Int J Food Microbiol 56(1):13–20. Perez-Conesa D, McLandsborough L, Weiss J. 2006. Inhibition and inactivation of Listeria monocytogenes and Escherichia coli O157:H7 colony biofilms by micellar-encapsulated eugenol and carvacrol. J Food Prot 69(12):2947–2954. Puupponen-Pimia R, Nohynek L, Hartmann-Schmidlin S, Kahkonen M, Heinonen M, Maatta-Riihinen K, Oksman-Caldentey KM. 2005. Berry phenolics selectively inhibit the growth of intestinal pathogens. J Appl Microbiol 98(4):991–1000. Rasch M, Andersen JB, Nielsen KF, Flodgaard LR, Christensen H, Givskov M, Gram L. 2005. Involvement of bacterial quorum-sensing signals in spoilage of bean sprouts. Appl Environ Microbiol 71(6):3321–3330. Rasmussen TB, Bjarnsholt T, Skindersoe ME, Hentzer M, Kristoffersen P, Köte M, Nielsen J, Eberl L, Givskov M. 2005. Screening for quorum-sensing inhibitors (QSI) by use of a novel genetic system, the QSI selector. J Bacteriol 187(5):1799–1814. Reynolds G. 2007. Bacteria effective against fresh cut fruit contamination. http://www.foodnavigator-usa. com/news/printNewsBis.asp?id=75142. Accessed on 22 March 2007. Rhodes PL, Mitchell JW, Wilson MW, Melton LD. 2006. Antilisterial activity of grape juice and grape extracts derived from Vitis vinifera variety Ribier. Int J Food Microbiol 107(3):281–286. Richards GM, Beuchat LR. 2005. Infection of cantaloupe rind with Cladosporium cladosporoides and Penicillium expansum, and associated migration of Salmonella poona into edible tissues. Int J Food Microbiol 103(1):1–10.

Biological Control of Human Pathogens

223

Richards GM, Buck JW, Beuchat LR. 2004. Survey of yeasts for antagonistic activity against Salmonella Poona in cantaloupe juice and wounds in rinds coinfected with phytopathogenic molds. J Food Prot 67(10):2132–2142. Riordan DCR, Sapers GM, Annous BA. 2000. The survival of Escherichia coli O157:H7 in the presence of Penicillium expansum and Glomerula cingulata in wounds on apple surfaces. J Food Prot 63(12):1637–1642. Roller S, Seedhar P. 2002. Carvacrol and cinnamic acid inhibit microbial growth in fresh-cut melon and kiwifruit at 4 ° and 8 °C. Lett Appl Microbiol 35(5):390–394. Sansone G, Rezza I, Calvente V, Benuzzi D, Tosetti MI. 2005. Control of Botrytis cinerea strains resistant to iprodione in apple with rhodotorulic acid and yeasts. Postharvest Biol Technol 35(3):245–251. Sapers GM. 2006. Washing and sanitizing treatments for fruits and vegetables. In: GM Sapers, JR Gorny, and AE Yousef (ed). Microbiology of Fruits and Vegetables. New York: CRC Press. pp. 375–400. Schuenzel KM, Harrison MA. 2002. Microbial antagonists of foodborne pathogens on fresh, minimally processed vegetables. J Food Prot 65(12):1909–1915. Scolari G, Vescovo M. 2004. Microbial antagonism of Lactobacillus casei added to fresh vegetables. Ital J Food Sci 16(4):465–475. Serra AT, Matias AA, Nunes AVM, Leitão MC, Brito D, Bronze R, Silva S, Pires A, Crespo MT, Romão MVS, Duarte CM. 2008. In vitro evaluation of olive- and grape-based natural extracts as potential preservatives for food. Innov Food Sci Emerg Technol doi:10.1016/j.ifset.2007.011. Serrano M, Martinez-Romero D, Castillo S, Guillen F, Valero D. 2005. The use of natural antifungal compounds improves the beneficial effect of MAP in sweet cherry storage. Innov Food Sci Emerg Technol 6(1):115–123. Settanni L, Corsetti A. 2008. Application of bacteriocins in vegetable food biopreservation. Int J Food Microbiol 121(2):123–138. Shoda M. 2000. Bacterial control of plant diseases. J Biosci Bioeng 89(6):515–521. Singh N, Singh RK, Bhunia AK. 2003. Sequential disinfection of Escherichia coli O157:H7 inoculated alfalfa seeds before and during sprouting using aqueous chlorine dioxide, ozonated water, and thyme essential oil. Lebensm Wiss Technol 36(2):235–243. Singh N, Singh RK, Bhunia AK, Stroshine RL. 2002. Efficacy of chlorine dioxide, ozone, and thyme essential oil or a sequential washing and killing Escherichia coli O157:H7 on lettuce and baby carrots. Lebensm Wiss Technol 35(8):720–729. Smith JL, Fratamico PM, Novak FS. 2004. Quorum sensing: A primer for food microbiologists. J Food Prot 67(5):1053–1070. Takeuchi K, Frank JF. 2000. Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and effect of chlorine treatment on cell viability. J Food Prot 63(4):434–440. Torriani S, Orsi C, Vescovo M. 1997. Potential of Lactobacillus casei, culture permeate, and lactic acid to control microorganisms in ready-to-use vegetables. J Food Prot 60(12):1564–1567. Trias R, Bañeras L, Badosa E, Montesinos E. 2008. Bioprotection of Golden Delicious apples and Iceberg lettuce against foodborne bacterial pathogens by lactic acid bacteria. Int J Food Microbiol (1–2):50–60. Tzortzakis NG. 2007. Maintaining postharvest quality of fresh produce with volatile compounds. Innov Food Sci Emerg Technol 8(1):111–116. Ukuku DO, Bari ML, Kawamoto S, Isshiki K. 2005. Use of hydrogen peroxide in combination with nisin, sodium lactate and citric acid for reducing transfer of bacterial pathogens from whole melon surfaces to fresh-cut pieces. Int J Food Microbiol 104(2):225–233. Ukuku DO, Fett WF. 2004. Effect of nisin in combination with EDTA, sodium lactate and potassium sorbate for reducing Salmonella in whole and fresh-cut cantaloupe. J Food Prot 67(10):2143–2150. Ukuku DO, Fett WF, Sapers GM. 2004. Inhibition of Listeria monocytogenes by native microflora of whole cantaloupe. J Food Safety 24(2):129–146. US FDA C. 1999. Potential for infiltration, survival and growth of human pathogens within fruits and vegetables. www.cfsan.fda.gov/∼comm/juicback/html. Accessed on 13 November 2007. Valero D, Valverde JM, Martinez-Romero D, Guillen F, Castillo S, Serrano M. 2006. The combination of modified atmosphere packaging with eugenol or thymol to maintain quality, safety and functional properties of table grapes. Postharvest Biol Technol 41(3):317–327.

224

Section III. Postharvest Interventions

Vescovo M, Orsi C, Scolari G, Torriani S. 1995. Inhibitory effect of selected lactic acid bacteria on microflora associated with ready-to-eat vegetables. Lett Appl Microbiol 21(2):121–125. Vescovo M, Torriani S, Orsi C, Macchiarolo F, Scolari G. 1996. Application of antimicrobial-producing lactic acid bacteria to control pathogens in ready-to-use vegetables. J Appl Microbiol 81(2):113–119. von Bodman SB, Bauer WD, Coplin DL. 2003. Quorum sensing in plant-pathogenic bacteria. Ann Rev Phytopathol 41:455–482. Wang S, Ng TB, Chen T, Lin D, Wu J, Rao P, Ye X. 2005. First report of a novel plant lysozyme with both antifungal and antibacterial activities. Biochem Biophys Res Comm 327(3):820–827. Weissinger WR, McWatters KH, Beuchat LR. 2001. Evaluation of volatile chemical treatments for lethality to Salmonella on alfalfa seeds and sprouts. J Food Prot 64(4):442–450. Wells JM, Butterfield JE. 1997. Salmonella contamination associated with bacterial soft rot of fresh fruits and vegetables in the marketplace. Plant Dis 81(8):867–872. ———. 1999. Incidence of Salmonella on fresh fruits and vegetables affected by fungal rots or physical injury. Plant Dis 83(8):722–726. Wilderdyke MR, Smith DA, Brashears MM. 2004. Isolation, identification, and selection of lactic acid bacteria from alfalfa sprouts for competitive inhibition of foodborne pathogens. J Food Prot 67(5):947–951. Wilson CL, Wisniewski ME. 1989. Biological control of postharvest diseases of fruits and vegetables: an emerging technology. Annu Rev Phytopathol 27:425–441. Xu W, Qu W, Huang K, Guo F, Yang J, Zhao H, Luo Y. 2007. Antibacterial effect of grapefruit seed extract on food-borne pathogens and its application in the preservation of minimally processed vegetables. Postharvest Biol Technol 45(1):126–133. Yildirim Z, Johnson MG. 1998. Detection and characterization of a bacteriocin produced by Lactococcus lactis subsp. cremoris R isolated from radish. Lett Appl Microbiol 26(4):297–304.

12

Extension of Shelf Life and Control of Human Pathogens in Produce by Antimicrobial Edible Films and Coatings Tara H. McHugh, Roberto J. Avena-Bustillos, and Wen-Xian Du

Biopolymers Used for Edible Films and Coatings Components of edible films and coatings can be divided into three categories: hydrocolloids, lipids, and composites. Hydrocolloids include proteins and polysaccharides, such as starch, alginate, cellulose derivatives, chitosan, and agar. Lipids include waxes, acylglycerols, and fatty acids (Min and Krochta 2005). Composites contain combinations of both hydrocolloid components and lipids. The choice of formulation for edible film or coating is largely dependent on its desired function—such as biodegradability, edibility, aesthetic appearance, and good barrier properties against oxygen—which varies based on the composition of the film (Cha and Chinnan 2004). In addition, edible films and coatings can serve as supports containing antimicrobial, nutritional, and antioxidant substances (Gennadios and others 1997). Depending on their composition, the functionality of edible film and coating materials may vary because each component confers different properties on the composite matrix. Films made of hydrocolloids (polysaccharides or proteins) usually have strong mechanical and gas barrier properties, but also have poor water vapor barrier properties and high permeability to moisture. In contrast, films composed of lipids exhibit good water vapor barrier properties, but they tend to show poor mechanical strength and high oxygen permeability. Combining these components into one matrix allows them to physically and/or chemically interact and may result in films with improved properties (Diab and others 2001). For example, fruit-based edible films can be made with excellent oxygen barrier properties, but not very good moisture barrier properties. Combining fruit purées with various gelling agents (such as alginate) improves the water barrier and tensile properties of the resultant fruit-based films (Mancini and McHugh 2000). Polysaccharides are commonly used for edible films because their film-forming properties are derived from cellulose, starch, alginate, and their mixtures. A plasticizer is normally added to increase the flexibility of the film, and occasionally it is used only to facilitate the polymer processing. The most commonly used plasticizers in starch-based films are polyols, such as sorbitol and glycerol. They are frequently added into edible films to relax the intermolecular forces and increase the mobility of the polymeric chains to improve flexibility (Durango and others 2006). Glycerol is a low– molecular-weight nonvolatile substance that is often used to modify the mechanical 225

226

Section III. Postharvest Interventions

properties of hydrophilic films. The addition of glycerol into films reduces internal hydrogen bonding between polymer chains while increasing molecular volume, resulting in an improvement in film flexibility (Mali and others 2006). The development of films from water-soluble polysaccharides has led to promising new types of materials for the preservation of fruits and vegetables, because these biopolymers show selective permeability to O2 and CO2. These films reduce O2 levels and increase CO2 levels in the internal atmospheres of coated fruits and vegetables and reduce respiration rates, thereby extending the shelf life of fresh produce in a manner similar to modified/controlled atmosphere storage (Diab and others 2001).

Edible Coatings for Fresh Fruits and Vegetables Edible coatings are continuous biopolymeric matrices formed as films and directly applied on the exterior surface of fresh fruits and vegetables. Edible wax coatings have been used in fresh produce since the 1930s in the United States to reduce moisture loss and improve glossiness (Park 1999). Edible coatings are prepared as solutions and emulsions from proteins, polysaccharides, and lipids and are applied on produce surfaces by different mechanical procedures, such as dipping, spraying, and brushing (Avena-Bustillos and others 1994, 1997), or by electrostatic deposition (Amefia and others 2006). The chemical and physical characteristics of the edible coating solution and the coating thickness, homogeneity, and adhesiveness depend on the surface structure and morphology of the fruits and vegetables (Miller and Krochta 1997). Produce skin pores, trichomes, and natural waxes all affect the oxygen, carbon dioxide, and water permeability properties of the coatings and influence their capability to maintain freshlike quality of produce (Avena-Bustillos and others 1994; Park 1999). Coating matrices also can incorporate active antimicrobial agents to provide produce with microbial produce stability and protect against foodborne outbreaks.

Edible Films for Fresh Fruits and Vegetables Edible films are thin films prepared from edible material that act as a barrier to control moisture, oxygen, carbon dioxide, flavor, and aroma exchange between food components or with the atmosphere surrounding the food and also to protect the product, extend its shelf life, and improve its quality (Suyatma and others 2005). For edible films to be used in foods, there are several requirements to be considered, such as appropriate gas and water barrier properties; good mechanical strength and adhesion; reasonable microbial, biochemical, and physicochemical stability; effective carrier capability for antioxidant, flavor, color, nutritional, or antimicrobial additives; safety for human consumption (free of pathogenic microorganisms and hazardous compounds); acceptable sensorial characteristics; low cost of raw materials; and simple technology for production (Debeaufort and others 1998). Generally, an edible film is defined as a preformed thin layer or solid sheet of edible material placed on or between food components (Krochta and De Mulder-Johnston 1997). Edible films can be used as wraps or pouches for food. Wrapped films were shown to be advantageous over traditional coatings for retarding moisture and color losses in fresh-cut apples during storage (McHugh and Senesi 2000). Edible films can also enhance or improve food’s appearance and nutritional value.

Antimicrobial Edible Films and Coatings

227

The applications of edible films to fresh fruits and vegetables have received increasing interest because these films can serve as carriers for various antimicrobial compounds that can reduce the risk of pathogen growth. Preservatives, acidulants, antioxidants, and antibiotic compounds can be added to edible films to reduce surface microbial populations on foods and enhance oxygen-barrier properties. Edible films can also enhance food nutritional value and improve the appearance of foods. A greater emphasis on safety features associated with the addition of antimicrobial agents is the next area for development in edible films technology (Cha and Chinnan 2004).

Fruit and Vegetable-Based Edible Films McHugh and others (1996) developed the first edible films made from fruit purées and characterized their water vapor and oxygen permeability properties. Fruit-based edible films were excellent oxygen barriers, particularly at low to moderate relative humidities. McHugh and Senesi (2000) coated apple pieces by dipping into solutions and then drying or wrapping in preformed apple-based edible films. Increasing the lipid concentration of the films significantly improved its moisture barrier properties. Water vapor permeability values were reduced from 325 to 69 g-mm/kPa-d-m2 through the addition of lipids. Apple-based wraps significantly reduced moisture loss and browning in freshcut apples, retaining color for 12 days at 5 °C. Wraps were significantly more effective than coatings of the same composition (McHugh and others 1997). In addition to providing antimicrobial properties, fruit- and vegetable-based edible films can benefit consumers in other ways. For example, although the USDA Food Guide Pyramid recommends that mature adults consume 2–4 servings of fruit per day, less than half of Americans meet these dietary recommendations. Because consumers demand convenience and variety, there is a need to provide access to fruit products in new, innovative forms. Incorporation of fruit purées into edible barrier films can help meet these needs. Fruit films, due to their low moisture levels, are concentrated sources of natural nutrients and can impart appealing colors and flavors. Apple and tomato purées have been used to prepare model edible films in recent studies to incorporate antimicrobial plant essential oils (Rojas-Graü and others 2006, 2007a; Olsen and others 2008). Undoubtedly, the results can be extrapolated to other fruit- and vegetable-based films. Based on the interest in the use of fruit and vegetable films, a commercial partner, Origami Foods, has begun to commercialize fruit- and vegetable-based edible films. A potential application of fruit- and vegetable-based edible films is the controlled release of volatile active antimicrobial compounds from the natural essential oils of plants. Because plant essential oils and some fruit and vegetable products are commonly found in combination food products such as pizza, which contains tomato, basil, and oregano, it is anticipated that the flavors of plant essential oils and other antimicrobial phytochemicals added to the fruit and vegetable films will be readily acceptable to consumers (Rojas-Graü and others 2007b). Edible films can then be incorporated into conventional packaging systems (Koide and Shi 2007) for fresh and fresh-cut fruits and vegetables with a dual purpose as edible and antimicrobial components.

228

Section III. Postharvest Interventions

Edible Film Casting Methods Despite the growth in research on edible films, the extent of commercialization has not progressed as significantly. Manufacturing processing methods and the resultant mechanical and water barrier properties of edible films must be improved for practical use (Arvanitoyannis and Gorris 1999). Edible films are commonly produced via a solution casting process where the films are dried from 2 to 12 h. Shorter drying times allow the formation of films with no significant microbial contamination. Knowledge of critical control points is necessary to reduce the risk of microbial growth. The quality of the starting materials, as well as the use of heat and good sanitation during casting and drying, is necessary to ensure safety (McHugh and Olsen 2001). Most of the edible films made have been cast using inefficient technologies and there is a need to develop more efficient methodologies for the mass production of edible films. Recently, we reported significant differences in physical and antimicrobial properties of apple- and tomato-based edible films made by continuous casting under infrared heating in a pilot plant lab coater and by a batch drying process done overnight under ambient air (Du and others, 2008a,b). The continuous method for film casting was more suitable for large-scale production of fruit- and vegetable-based edible films than the batch method. The tendency of volatile active antimicrobial compounds to evaporate during casting at high temperatures can be compensated by manipulating the formulation to achieve an appropriate final concentration of the antimicrobial compound in the dried films (Du and others, 2008a,b).

Antimicrobial Plant Essential Oils in Edible Films Naturally derived biological compounds and other natural products may have application in controlling pathogens in produce. They have varied antimicrobial and antioxidant properties that can break down cellular membranes and disrupt biosynthetic pathways of microorganisms (Benaventi-García and others 1998; Bowles and Juneja 1998; Bowles and others 1995). The use of edible films as antimicrobial carriers represents an interesting approach for the external incorporation of plant essential oils and other phytochemicals onto food system surfaces. The agents can then diffuse into the food to control target microorganisms. The antimicrobial activity of plant essential oils is confined to a number of small terpenoid and phenolic compounds, which are known to exhibit antibacterial or antifungal activity. Recent studies have shown that essential oils of oregano (Origanum vulgare), thyme (Thymus vulgaris), cinnamon (Cinnamon casia), lemongrass (Cymbopogon citratus), and clove (Eugenia caryphyllata) are among the most active antimicrobials against strains of Escherichia coli (Dorman and Deans 2000; Friedman and others 2002; Hammer and others 1999; Smith-Palmer and others 1998). Although the effectiveness of all these compounds has been widely reported, carvacrol (a major component of the essential oils of oregano and thyme) appears to have received the most attention from investigators. Carvacrol is generally regarded as safe (GRAS) and used as a flavoring agent in baked goods, sweets, ice cream, beverages, and chewing gum (Fenaroli 1995). However, widespread application of plant essential oils in food systems has been limited by the incompatibility of their strong flavors with some foods. Plant essential oils and their components are compatible with the sensory characteristics of fruits and vegetables and have been shown to prevent bacterial growth.

Antimicrobial Edible Films and Coatings

229

Among the complex constituents of citrus essential oils, the terpene citral is known to have strong antifungal properties (Rodov and others 1995). In addition, cinnamon oil and its active compound (cinnamaldehyde) also have been tested for their inhibitory activity against E. coli (Friedman and others 2004a,b; Helander and others 1998). Phenolic compounds are found in numerous plant species (Walsh 2003). These compounds appear to be involved in the defense of plants against invading pathogens, including bacteria, fungi, and viruses. Phenolic compounds present in teas (Friedman and others 2005, 2006), pigmented rice brans (Nam and others 2006), and most fruits and vegetables (Shahidi and Naczk 2004) are also reported to exhibit antimicrobial effects (Friedman and others 2003, 2005). Some of these have been incorporated in edible films (Cagri and others 2004). Studies on the antibacterial activity of oregano, lemongrass, and cinnamon plant essential oils and their major components carvacrol, citral, and cinnamaldehyde in apple purée film-forming solutions against the foodborne pathogen E. coli O157 : H7 and Salmonella enterica show that oregano oil as well as its major component carvacrol killed E. coli O157 : H7 and S. enterica practically on contact (3 min). The order of antimicrobial activities was as follows: carvacrol > oregano > citral > cinnamaldehyde > lemongrass > cinnamon oil (Friedman and others 2004). The evaluation of the physicochemical properties of films made from apple slurries revealed no adverse effect of the additives on water vapor permeability properties (Rojas-Graü and others 2006, 2007a). The antimicrobial films showed good oxygen barrier properties and their tensile strength did not differ significantly from control films without added antimicrobials.

Physical Properties of Edible Films Containing Plant Essential Oils The ideal characteristics of an edible film would be low water vapor permeability and high mechanical strength. The physicochemical properties of edible films (e.g., color, tensile strength, water vapor, and oxygen permeability) relate to the ability of the coating to enhance the mechanical integrity of foods, inhibit moisture loss and oxidative rancidity, and improve final-product appearance (Debeaufort and others 1998). A complete analysis of both antimicrobial and physicochemical properties is important for predicting the behavior of antimicrobial edible films in the food system (Cagri and others 2001; McHugh and Krochta 1994b). McHugh and others (1996) demonstrated that apple-based edible films were not very good moisture barriers and that the addition of lipids could potentially improve the water barrier properties of fruit-based films. Rojas-Graü and others (2006) found that water vapor permeability decreased when the proportion of the hydrophobic compounds increased in apple-based edible films, this effect being more prominent when oregano oil was used in the composition of the films. Adding carvacrol addition to apple purée edible films resulted in significant decrease in film water vapor permeability. Water vapor transfer generally occurs through the hydrophilic portion of the film; thus, water vapor permeability depends on the hydrophilic-hydrophobic ratio of the film components (Hernández 1994). Water vapor permeability increases with polarity, unsaturation, and degree of branching of the lipid, but it also depends on the water absorption properties of the polar part of the film (Gontard and others 1994).

230

Section III. Postharvest Interventions

The chemical nature of the essential oils also plays an important role in the barrier properties of edible films. Differences observed in these properties can be explained by the hydrophobicity of the plant essential oils. Carvacrol, a phenolic compound containing an alcohol group in its chemical structure, seems to be a good barrier compared to aldehyde compounds (e.g., cinnamaldehyde, citral) because the hydroxyl group has less affinity for water than for the carbonyl groups. Carvacrol then offers the possibility not only to enhance antimicrobial efficiency but also to improve water barrier properties of edible films. McHugh and Senesi (2000) suggested that lipids with lower melting points, such as vegetable oil, oleic acid, and myristyl alcohol, exhibit superior barrier properties presumably due to their smooth structure and lack of channels between crystalline platelets through which water could migrate easily. The incorporation of emulsion droplets in the film increases the distance traveled by water molecules that diffuse through the film, thereby decreasing water vapor permeability (McHugh and Krochta 1994c). McHugh and others (1996) demonstrated that apple-based edible films are excellent oxygen barriers, particularly at low-to-moderate relative humidities. An apple purée edible film was a good oxygen barrier, exhibiting low oxygen permeability values of 22.6 ± 1.3 cm3μm/m2-d-kPa. The oxygen permeability values of this film increased as higher amounts of plant essential oils were incorporated. McHugh and Krochta (1994a) indicated that films containing lipids exhibit relatively poor oxygen barrier properties. The oil chemical nature plays a major role in the barrier properties of edible films. Lower oxygen permeability was observed in films that contained oregano, lemongrass, and cinnamon oils than in those that contained its antibacterial compounds carvacrol, citral, and cinnamaldehyde, respectively (Rojas-Graü and others 2006, 2007a). Tensile strength is one of the most common indicators of the mechanical property of an edible film. It expresses the maximum stress developed in a film specimen during tensile testing (Gennadios and others 1994). The incorporation of plant essential oils in apple-based edible films caused a significant increase in tensile strength, % elongation, and elastic modulus of the film. These differences could be related to differences in their polarities. These results are in agreement with those obtained by Pranoto and others (2005), who studied the physical and antibacterial properties of alginate edible film with garlic oil. Elongation at break is a measure of the film stretchability prior to breakage (Krochta and De Mulder-Johnston 1997). Zivanovic and others (2005) studied the antimicrobial and physicochemical properties of polysaccharide (chitosan) films enriched with essential oils. They observed a decrease in tensile strength and an increase in elongation percentage when the essential oils were introduced into the films. This behavior also was observed by Bégin and Van Calsteren (1999).

Evaluation of Antimicrobial Activity of Volatile Components The growth of microorganisms on the surface of a food is a key factor affecting the safety and/or spoilage of food products (Padgett and others 1998). The direct addition of an antimicrobial additive into foods might reduce its effectiveness, due to the presence of substances that interact with it, to reduce its antimicrobial effect (Durango and others 2006). The use of antimicrobial films could be more efficient than adding

Antimicrobial Edible Films and Coatings

231

antimicrobials directly to the food. The antimicrobials migrate selectively and gradually from the film surface toward the surface of the food, and therefore maintain a high concentration of antimicrobial at the food surface for extended exposure (Ouattara and others 2000). Antimicrobial substances incorporated into edible films can control microbial contamination of fruits and vegetables by reducing the growth rate of target microorganisms, or by inactivating microorganisms by direct contact. Most of the existing methods for testing the antimicrobial activities of substances require direct contact between the active agent and the microorganism (i.e., food), and thus are not relevant to many commercial products in which there is little or no direct contact between the food and the packaging material (Rodríguez and others 2007). Vapor phase tests, which are not direct contact assays, can be used to assess the protection provided by the antimicrobial volatile materials under no direct contact conditions. One advantage of essential oils is their bioactivity in the vapor phase, a characteristic that makes them useful as possible fumigants for stored commodity protection. The antimicrobial activity of essential oils by vapor contact was first reported by Kellner and Kober (1954). They studied the effect of 175 essential oils in the gaseous state against eight airborne bacteria and fungi using an inverted petri plate technique (Maruzzella and Sicurella 1960). A volatile compound contained in a cup or on a paper disc was exposed to the inverted agar medium inoculated with a test organism. The size of the growth inhibitory zone after incubation is used as the measure of vapor activity. This technique is convenient for qualitative analysis, but not for quantitative comparison of the vapor activity of essential oils (Inouye and others 2003). For components to evaporate and be classed as volatile it is imperative that there is a loss of weight over a time or temperature course (Fisher and Phillips 2008). The evaporation of the essential oils is effected by external factors such as temperature, humidity, concentration, and pressure (Aumo and others 2006). Volatile compounds from plants usually have a relatively high vapor pressure and are capable of interacting with an organism through the liquid and the gas phase (Fries 1973). Storage temperature also influences the antimicrobial activity of chemical preservatives. Generally, increased storage temperature can accelerate the migration of the active agents in the film/coating layers, and refrigeration slows down the migration rate (Quintavalla and Vicini 2002).

Methods to Measure the Antimicrobial Activity of Edible Films Plant essential oils are a potentially useful source of antimicrobial compounds that can be incorporated into edible films. Factors such as the composition and solubility of the oil, bacterial strain, the sources of antimicrobial samples used, and the method of growing and enumerating the surviving bacteria can influence the determination of the antimicrobial activity of a plant oil (Friedman and others 2002; Zaika 1988). Zone of inhibition assay (agar diffusion assay) is a commonly used method for the measurement of antimicrobial activity of edible films on solid medium. A recent study on the contribution of the vapors to the antimicrobial effects in direct disc diffusion method indicated that only the water-soluble components diffused across the agar while the redeposition of the vaporized components on the surface of the agar

232

Section III. Postharvest Interventions

accounted for the remainder of the inhibition. It was found that for oils containing alcohol, ketone, ester, oxide, and hydrocarbons the major inhibition came from the vapors whereas for oils containing greater volumes of aldehydes inhibition came from diffusion (Inouye and others 2006). Minimum inhibitory concentrations (MICs) of antimicrobial edible films can be assessed by the agar diffusion assay and observing the zone of inhibition, or the agar dilution method with visible growth observed, or broth dilution with visible growth, optical density, absorbance, or viable counts measured (Burt 2004). The MIC is determined as the lowest concentration at which growth is inhibited. The major problem with the method of determining the strength of antimicrobial activity of edible films is their hydrophobic nature, which makes them insoluble in water-based media (Fisher and Phillips 2008). Recently applied methods to evaluate the effect of antimicrobial edible films on the inhibition of human pathogens are shown in Table 12.1.

Table 12.1. Effects of edible antimicrobial films on the inhibition of human pathogens Base Material for Films Whey protein isolate (WPI)

Alginate-apple puree

a b

Target Pathogens Antimicrobial E. coli O157 : H7 Staph. aureus S. enteritidis L. monocytogenes E. coli O157 : H7

Pectin-apple puree

E. coli O157 : H7

Alginate

Yam starchb

E. coli S. typhimurium Staph. aureus B. cereus S. enteritidis

Alginate-apple

E. coli O157 : H7

Chitosan

L. monocytogenes

Chitosan

E. coli O157 : H7 L. monocytogenes

Antimicrobial Agent Oregano oil Garlic oil Rosemary oil

MICa

References

2 %w/v 3 >4

Seydim and Sarikus 2006

Oregano oil Carvacrol Lemongrass oil Citral Cinnamon oil Cinnamaldehyde Oregano oil Lemongrass oil Cinnamon oil Garlic oil Garlic oil Garlic oil Garlic oil Chitosan

0.1 %w/v 0.1 0.5 0.5 0.5 0.5 0.08 %w/v 0.5 0.5 >0.4 %v/v >0.4 0.2 0.1 3

Rojas-Graü and others 2007a

Citral oil Lemongrass oil Chitosan

0.5 %w/w 0.5 >1 %w/v

Anise oil Basil oil Coriander oil Oregano oil

4 %w/w 4 4 1

Rojas-Graü and others 2006 Pranoto and others 2005

Durango and others 2006 Rojas-Graü and others (2007b) Coma and others 2001 Zivanovic and others 2005

Concentration in film solution. Method for testing inhibition was growth curve; all other tests were by zone of inhibition assay.

Antimicrobial Edible Films and Coatings

233

Use of Edible Films and Coatings on Fresh Fruits and Vegetables The most important quality attributes contributing to the marketability of fresh produce include appearance, color, texture, flavor, nutritional value, and microbial safety. These quality attributes are determined by plant variety, stage of maturity or ripening, and the pre- and postharvest conditions (Lin and Zhao 2007). Fresh fruits undergo vigorous biological reactions after harvest because their respiration accelerates the natural loss of fruit tissue. Therefore, fruits tend to lose water at room temperature; change appearance, texture, and quality; and decrease in commercial value. For use on fresh fruits and vegetables, an edible film would include good barrier properties and be odorless, tasteless, and transparent. Edible polymer films may be formed as either food coatings or stand-alone film wraps and pouches. They have the potential use with food as moisture, gas, and/or aroma barriers. A list of applications of edible films on fresh fruits and vegetables is shown in Table 12.2. Table 12.3 shows some examples of successful applications of antimicrobial edible coatings on fresh fruits and vegetables. Antimicrobial edible coatings are more promising to be used on fresh-cut fruits and vegetables than on fresh produce, except when the produce is commonly consumed without peeling like the fresh produce listed in Table 12.3. Surprisingly, little research has been done on applications of antimicrobial coatings to melons or tomatoes, both of which have been reported in several major foodborne pathogen outbreaks. The potential benefits of using edible films and coatings in the fresh produce industry include providing a moisture barrier on the surface of produce to decrease moisture loss; providing a sufficient gas barrier to control gas exchange between the fresh produce and its surrounding atmosphere to slow respiration, delay deterioration, and protect the fresh produce from brown discoloration and texture softening during storage; restricting the loss of natural volatile flavor and color compounds from the fresh produce or the acquisition of foreign odors by providing gas barriers; protecting produce from physical damage caused by mechanical impact, abrasions, pressure,

Table 12.2. Application of edible films on fresh fruits and vegetables Product Strawberry

Application Wrap, pouch

Film Materials Wheat gluten-based films

Apple

Wrap

Apple-based edible films

Lettuce

Wrap

Biodegradable protein film

Green pepper

Wrap

Polylactic acid–based biodegradable film

Functions Retention of firmness, reduced weight loss, and maintained visual quality during storage Reduced moisture loss and browning in fresh-cut apples Did not show any beneficial effects on pectic substances and pigments Can be used to maintain quality and sanitary conditions in modified atmosphere packaging

References Tanada-Palmu and Grosso 2005

McHugh and Senesi 2000 Schreiner and others 2003

Koide and Shi 2007

234 Table 12.3. Application of antimicrobial edible coatings on fresh fruits and vegetables Product Strawberry Strawberry

Antimicrobial Potassium sorbate/citric acid Chitosan

Film Materials Corn and potato starch Chitosan, lactic acid, and sodium lactate

Table grape

Aloe vera

Aloe vera

Cherry

Aloe vera

Aloe vera

Apple

Malic/lactic acid

Soy protein isolate and glycerol

Quince

Ascorbic acid

Carrot

Turmeric

Lettuce

Chitosan

Semperfresh (sucrose esters of fatty acids (FA), sodium carboxymethyl cellulose, and FA monodiglycerides) Casein, polyvinyl, and propylene gylcol alcohol Chitosan, lactic acid, and sodium lactate

Cabbage leaf

Lemon, orange, and bergamot essential oils

Lemon, orange, and bergamot essential oils

Functions Inhibition of coliforms growth, extending shelf life for 14 days Controlling decay and psychrotrophic food pathogens Extended shelf life up to 31 days and reduced initial mesophilic aerobic count Extended shelf-life, improved sensory and chemical quality, and reduced initial mesophilic aerobic count Inhibited growth of pathogenic bacteria and extended shelf life Extended shelf life up to 31 days and reduced initial mesophilic aerobic count

References García and others 1998 Devlieghere and Debevere 2004 Valverde and others 2005 Martinez-Romero and others 2006

Inhibition of coliforms growth extending shelf life for 7 days Controlling decay and psychrotrophic food pathogens Reduce 5–6 log microbial loads Grampositive and Gram-negative bacteria (Campylobacter jejuni, L. monocytogenes, B. cereus, and S. aureus)

Jagannath and others 2006 Devlieghere and Debevere 2004 Fisher and Phillips 2006

Eswaranandam and others 2006 Yurdugül 2005

Antimicrobial Edible Films and Coatings

235

vibrations, and other factors; and acting as carriers for other functional ingredients, such as antimicrobial compounds, antioxidant agents, phytochemicals, colorants, and flavor ingredients for reducing microbial loads, delaying oxidation and discoloration, and improving quality and shelf life of fresh produce (Lin and Zhao 2007).

Summary The use of edible films and coatings as carriers of natural antimicrobials (such as plant essential oils) constitutes an approach for external protection of fruits and vegetables to reduce surface microbial populations and to enhance oxygen-barrier properties, potentially increasing food safety as well as shelf life of highly perishable foods such as fresh and fresh-cut fruits and vegetables. Appropriately formulated edible films and coatings can be utilized for fresh produce to meet challenges associated with stable quality, market safety, nutritional value, and economic production cost.

References Amefia AE, Abu-Ali JM, Barringer SA. 2006. Improved functionality of food additives with electrostatic coating. Innovative Food Science & Emerging Technologies 7(3):176–181. Arvanitoyannis I, Gorris LGM. 1999. Edible and biodegradable polymeric materials for food packaging or coating. In: Processing Foods: Quality Optimization and Process Assessment, edited by Oliveira FAR, Oliveira JC, pp. 357–371. CRC Press, Boca Raton, FL. Aumo J, Warna J, Salmi T, Murzin DY. 2006. Interaction of kinetics and internal diffusion in complex catalytic three-phase reactions: activity and selectivity in citral hydrogenation. Chemical Engineering Science 61(2):814. Avena-Bustillos RJ, Krochta JM, Saltveit ME. 1997. Water vapor resistance of red delicious apples and celery sticks coated with edible caseinate-acetylated monoglyceride films. Journal of Food Science, 62(2):351–354. Avena-Bustillos RJ, Krochta JM, Saltveit ME, Rojas-Villegas RJ, y Sauceda-Perez JA. 1994. Optimization of edible coating formulations on zucchini to reduce water loss. Journal of Food Engineering 21:197–214. Bégin A, Van Calsteren MR. 1999. Antimicrobial films produced from chitosan. International Journal of Biological Macromolecules 26:63–67. Benaventi-García O, Castillo J, Maríín FR, Ortunˇ o A, Del-Río JA. 1998. Uses and properties of citrus flavonoids. Journal of Agriculture and Food Chemistry 45:4505–4515. Bowles BL, Juneja VK. 1998. Inhibition of foodborne bacterial pathogens by naturally occurring food additives. Journal of Food Safety 18:101–112. Bowles BL, Sackitey SK, Williams AC. 1995. Inhibitory effects of flavor compounds on Staphlococcus aureus WRRC B124. Journal of Food Safety 15:337–347. Burt S. 2004. Essential oils: their antibacterial properties and potential applications in foods—a review. International Journal of Food Microbiology 94(3):223–253. Cagri A, Ustunol Z, Ryser ET. 2001. Antimicrobial, mechanical, and moisture barrier properties of low pH whey protein-based edible films containing p-aminobenzoic or sorbic acids. Journal of Food Science 66:865–870. ———. 2004. Antimicrobial edible films and coatings. Journal of Food Protection 67:833–48. Cha DS, Chinnan MS. 2004. Biopolymer-based antimicrobial packaging: a review. Critical Reviews in Food Science and Nutrition 44:223–237. Coma V, Sebti I, Pardon P, Deschamps A, Pichavant FH. 2001. Antimicrobial edible packaging based on cellulosic ethers, fatty acids, and nisin incorporation to inhibit Listeria innocua and Staphylococcus aureus. Journal of Food Protection 64(4):470–475. Debeaufort F, Quezada-Gallo JA, Voilley A. 1998. Edible films and coatings: tomorrow’s packagings: a review. Critical Reviews in Food Science 38:299–313.

236

Section III. Postharvest Interventions

Devlieghere F, Debevere AV. 2004. Chitosan: antimicrobial activity, interactions with food components and applicability as a coating on fruit and vegetables. Food Microbiology 21:703–714. Diab T, Biliaderis CG, Gerasopoulos D, Sfakiotakis E. 2001. Physicochemical properties and application of pullulan edible films and coatings in fruit preservation. Journal of the Science of Food and Agriculture 81:988–1000. Dorman HJ, Deans SG. 2000. Antimicrobial agents from plants: antibacterial activity of plant volatile oils. Journal of Applied Microbiology 88:308–316. Du WX, Olsen CW, Avena-Bustillos RJ, McHugh TH, Levin CE, Friedman, M. 2008a. Storage stability and antimicrobial activity against Escherichia coli O157 : H7 of carvacrol in edible apple films prepared by two different casting methods. Journal of Agriculture and Food Chemistry 56:3082–3088. ———. 2008b. Antibacterial activity against Escherichia coli O157 : H7, physical-chemical properties, and storage stability of novel carvacrol-containing edible tomato films. Journal of Food Science 73(7):M378–M383. Durango AM, Soares NFF, Benevides S, Teixera J, Carvalho M, Wobeto C, Andrade NJ. 2006. Development and evaluation of an edible antimicrobial film based on yam starch and chitosan. Packaging Technology Science 19:55–59. Eswaranandam S, Hettiarachy NS, Meullenet JF. 2006. Effect of malic and lactic acid incorporated soy protein coatings on the sensory attributes of whole apple and fresh-cut cantaloupe. Journal of Food Science 71(3):S307–S313. Fenaroli G. 1995. Fenaroli’s Handbook of Flavor Ingredients, 3rd ed, vol. 2. CRC Press Inc., Boca Raton, FL. Fisher K, Phillips C. 2006. The effect of lemon, orange and bergamot essential oils and their components on the survival of Campylobacter jejuni, Escherichia coli O157, Listeria monocytogenes, Bacillus cereus and Staphylococcus aureus in vitro and in food systems. Journal of Applied Microbiology 101(6):1232–1240. ———. 2008. Potential antimicrobial uses of essential oils in food: is citrus the answer? Trends in Food Science and Technology. 19:156–164. Friedman M, Buick R, Elliott CT. 2004a. Antibacterial activities of naturally occurring compounds against antibiotic-resistant Bacillus cereus vegetative cells and spores, Escherichia coli, and Staphylococcus aureus. Journal of Food Protection 67:1774–1778. Friedman M, Henika PR, Levin CE, Mandrell RE. 2004b. Antibacterial activities of plant essential oils and their components against Escherichia coli O157 : H7 and Salmonella enterica in apple juice. Journal of Agricultural and Food Chemistry 52:6042–6048. Friedman M, Henika PR, Levin CE, Mandrell RE, Kozukue N. 2006. Antimicrobial activities of tea catechins and theaflavins and tea extracts against Bacillus cereus. Journal of Food Protection 69:100–107. Friedman M, Henika PR, Mandrell RE. 2002. Bactericidal activities of plant essential oils and some of their isolated constituents against Campylobacter jejuni, Escherichia coli O157 : H7, Listeria monocytogenes, and Salmonella enterica. Journal of Food Protection 65:1545–1560. ———. 2003. Antibacterial activities of phenolic benzaldehydes and benzoic acids against Campylobacter jejuni, Escherichia coli O157 : H7, Listeria monocytogenes, and Salmonella enterica. Journal of Food Protection 66:1811–1821. Friedman M, Kim SY, Lee SJ, Han PG, Han JS, Lee KR, Kozukue N. 2005. Distribution of catechins, theaflavins, caffeine, and theobromine in 77 teas consumed in the United States. Journal of Food Science 70:C550–559. Fries NF. 1973. Effects of volatile organic compounds on the growth and development of fungi. Transactions of the British Mycology Society 60:1–21. García MA, Martino MN, Zaritzky NE. 1998. Plasticized starch-based coatings to improve strawberry (Fragaria x Ananassa) quality and stability. J. Agric. Food Chem. 46:3758–3767. Gennadios A, Hanna MA, Kurth LB. 1997. Application of edible coatings on meats, poultry and seafoods: a review. Lebensmittel Wissenschaft und Technologie 30(4):337–350. Gennadios A, McHugh TH, Weller CL, Krochta JM. 1994. Edible coating and films based on proteins. In: Edible Coatings and Films to Improve Food Quality, edited by Krochta JM, Baldwin EA, NisperosCarriedo, MO, pp. 201–277. Technomic Publishing Co., Lancaster, PA.

Antimicrobial Edible Films and Coatings

237

Gontard N, Duchez C, Cuq JL, Guilbert S. 1994. Edible composite films of wheat gluten and lipids: water vapour permeability and other physical properties. International Journal of Food Science and Technology 29:39–50. Hammer KA, Carson CF, Riley TV. 1999. Antimicrobial activity of essential oils and other plant extracts. Journal of Applied Microbiology 86:985–990. Helander IM, Alakomi HL, Latva-Kala K, Mattila-Sandholm T, Pol I, Smid EJ, Gorris LGM, Wright AV. 1998. Characterization of the action of selected essential oil components on Gram-negative bacteria. Journal of Agriculture and Food Chemistry 46:3590–3595. Hernández E. 1994. Edible coatings for lipids and resins. In: Edible Coatings and Films to Improve Food Quality, edited by Krochta JM, Baldwin EA, Nisperos-Carriedo MO, pp. 279–304. Technomic Publishing Co., Lancaster, PA. Inouye S, Abe S, Yamaguchi H, Asakura M. 2003. Comparative study of antimicrobial and cytotoxic effects of selected essential oils by gaseous and solution contacts. The International Journal of Aromatherapy 13(1)33–41. Inouye S, Uchida K, Maruyama N, Yamaguchi H, Abe S. 2006. A novel method to estimate the contribution of the vapour activity of essential oils in agar diffusion assay. Japanese Journal of Medical Mycology 47:91–98. Jagannath JH, Najappa C, Gupta DD, Bawa AS. 2006. Studies on the stability of an edible film and its use for the preservation of carrot (Daucus carota). International Journal of Food Science and Technology 41:498–506. Kellner W, Kober W. 1954. Possibilities of the use of ethereal oils for room disinfection. Arzneim-Forschung 4:319–325. Koide S, Shi J. 2007. Microbial and quality evaluation of green peppers stored in biodegradable film packaging. Food Control 18:1121–1125. Krochta JM, De Mulder-Johnston C. 1997. Edible and biodegradable polymer films: challenges and opportunities. Food Technology 51:61–74. Lin D, Zhao Y. 2007. Innovations in the development and application of edible coatings for fresh and minimally processed fruits and vegetables. Comprehensive Reviews in Food Science and Food Safety 6:60–75. Mali S, Groddmann MVE, García MA, Martino MN, Zaritzky NE. 2006. Effects of controlled storage on thermal, mechanical and barrier properties of plasticized films from different starch sources. Journal of Food Engineering 75:453–460. Mancini F, McHugh TH. 2000. Fruit-alginate interactions in novel restructured products. Nahrung 44(3):152–157. Martinez-Romero D, Alburquerque N, Valverde JM, Guillen F, Castillo S, Valero D, Serrano M. 2006. Postharvest sweet cherry quality and safety maintenance by Aloe vera treatment: A new edible coating. Postharvest Biology and Technology 39:93–100. Maruzzella JC, Sicurella NA. 1960. Antibacterial activity of essential oil vapors. Journal of the American Pharmaceutical Association—Science Edition 49:692–694. McHugh TH, Huxsoll CC, Krochta JM. 1996. Permeability properties of fruit puree edible films. Journal of Food Science 61:88–91. McHugh TH, Huxsoll CC, Robertson GH. 1997. Fruit puree-based films and coatings. In: Chemistry of Novel Foods, edited by Spanier AM, Tamura M, Okai H, Mills O, pp. 167–178. Allured Publishing Co., Carol Stream, IL. McHugh T, Krochta JM. 1994a. Permeability properties of edible films. In: Edible Coatings and Films to Improve Food Quality, edited by Krochta JM, Baldwin EA, Nisperos-Carriedo MO, pp. 139–187. Technomic Publishing Co., Lancaster, PA. ———. 1994b. Sorbitol- vs glycerol-plasticized whey protein edible films: integrated oxygen permeability and tensile property evaluation. Journal of Agriculture and Food Chemistry 42:841–845. ———. 1994c. Water vapour permeability properties of edible whey protein-lipid emulsion films. Journal of the American Oil Chemistry Society 71:307–312. McHugh TH, Olsen CW. 2001. Thermal mechanical and water vapor permeability properties of whey protein-peach and beta-lactoglobulin-peach films. Paper presented at Annual Meeting of the Institute of Food Technologists, Dallas, TX, June 19–23, 2001.

238

Section III. Postharvest Interventions

McHugh TH, Senesi E. 2000. Apple wraps: a novel method to improve the quality and extend the shelf life of fresh-cut apples. Journal of Food Science 65:480–485. Miller KS, Krochta JM. 1997. Oxygen and aroma barrier properties of edible films: A review. Trends in Food Science and Technology 8:228–237. Min S, Krochta JM. 2005. Antimicrobial films and coatings for fresh fruit and vegetables. In: Improving the Safety of Fresh Fruits and Vegetables, edited by Jongen W, pp. 455–492. CRC Press, New York. Nam SH, Choi SP, Kang MY, Koh HJ, Kozukue N, Friedman M. 2006. Antioxidative activities of bran extracts from twenty one pigmented rice cultivars. Food Chemistry 94:613–620. Olsen CW, Du WX, Avena-Bustillos RJ, McHugh TH, Levin CE, Friedman M. 2008. Bactericidal effect, storage stability, and physical properties of carvacrol added tomato edible films prepared by two different casting methods. Institute of Food Technologists Annual Meeting. New Orleans, LA. (Abstract 053-05). Ouattara B, Simard R, Piette G, Bégin A, Holley RA. 2000. Inhibition of surface spoilage bacteria in processed meats by application of antimicrobial films prepared with chitosan. International Journal of Food Microbiology 62:139–148. Padgett T, Han Ly, Dawson PL. 1998. Incorporation of food-grade antimicrobial compounds into biodegradable packaging films. Journal of Food Protection 61(10):1330–1335. Park HJ. 1999. Development of advanced edible coatings for fruits. Trends in Food Science and Technology 10:254–260. Pranoto Y, Salokhe VM, Rakshit SK. 2005. Physical and antibacterial properties of alginate-based edible film incorporated with garlic oil. Food Research International 38:267–272. Quintavalla S, Vicini L. 2002. Antimicrobial food packaging in meat industry. Meat Science 62(3):373–380. Rodríguez A, Batelle R, Nerín C. 2007. The use of natural essential oils as antimicrobial solutions in paper packaging. Part II. Progress in Organic Coatings 60:33–38. Rodov V, Ben-Yehoshua S, Fang DQ, Kim JJ, Ashkenazi R. 1995. Preformed antifungal compounds of lemon fruit: citral and its relation to disease resistance. Journal of Agriculture and Food Chemistry 43:1057–1061. Rojas-Graü MA, Avena-Bustillos RJ, Friedman M, Henika PR, Martin-Belloso O, Mchugh TH. 2006. Mechanical, barrier, and antimicrobial properties of apple puree edible films containing plant essential oils. Journal of Agriculture and Food Chemistry 54:9262–9267. Rojas-Graü MA, Avena-Bustillos RJ, Olsen C, Friedman M, Henika PR, Martin-Belloso O, Pan Z, Mchugh TH. 2007a. Effects of plant essential oils and oil compounds on mechanical, barrier and antimicrobial properties of alginate-apple puree edible films. Journal of Food Engineering 81:634–641. Rojas-Graü MA, Raybaudi-Massilia RM, Avena-Bustillos RJ, Martín-Belloso O, McHugh TH. 2007b. Apple puree-alginate edible coating as carrier of antimicrobial agents to prolong shelf life of fresh-cut apples. Postharvest Biology and Technology 45:254–264. Schreiner M, Huyskens-Keil S, Krumbein A, Prono-Widayat H, Ludders P. 2003. Effect of film packaging and surface coating on primary and secondary plant compounds in fruit and vegetable products. Journal of Food Engineering 56:237–240. Seydim AC, Sarikus G. 2006. Antimicrobial activity of whey protein based edible films incorporated with oregano, rosemary and garlic essential oils. Food Research International 39:639–644. Shahidi F, Naczk M. 2004. Phenolics in Food and Nutraceuticals. CRC Press, Boca Raton, FL. Smith-Palmer A, Stewart J, Fyfe L. 1998. Antimicrobial properties of plant essential oils and essences against five important food-borne pathogens. Letters in Applied Microbiology 26:118–122. Suyatma NE, Tighzert L, Copinet A. 2005. Effects of hydrophilic plasticizers on mechanical, thermal, and surface properties of chitosan films. Journal of Agriculture and Food Chemistry 53:3950–3957. Tanada-Palmu PS, Grosso CRF. 2005. Effect of edible wheat gluten-based films and coatings on refrigerated strawberry (Fragaria ananassa) quality. Postharvest Biology and Technology 36:199–208. Valverde JM, Valero D, Martinez-Romero D, Guillen FN, Castillo S, Serrano M. 2005. Novel edible coating based on aloe vera gel to maintain table grape quality and safety. Journal of Agriculture and Food Chemistry 53:7807–7813. Walsh C. 2003. Antibiotics: Actions, Origins, Resistance. ASM Press, Washington, D.C. Yurdugül S. 2005. Preservation of quinces by the combination of an edible coating material, Semperfresh, ascorbic acid and cold storage. European Food Research and Technology 220:579–586.

Antimicrobial Edible Films and Coatings

239

Zaika LL. 1988. Spices and herbs: their antimicrobial activity and its determination. Journal of Food Safety 9:97–118. Zivanovic S, Chi S, Draughon AF. 2005. Antimicrobial activity of chitosan films enriched with essential oils. Journal of Food Science 70:M45–M51.

13

Improving Microbial Safety of Fresh Produce Using Thermal Treatment Xuetong Fan, Bassam A. Annous, and Lihan Huang

Introduction Many food intervention technologies, including chemical (such as chlorine), physical (such as ionizing radiation), and biological (such as bacterial phage) treatment methods have been investigated for their effectiveness against human pathogens on foods. Each of those technologies, however, has certain drawbacks. Most chemical intervention methods are rather limited in their effectiveness to reduce the microbiological populations on the surfaces of fresh and fresh-cut produce, partially due to the inherent cracks, crevices, pockets, and other openings that provide a protective environment to microbes and hamper the access of many chemical sanitizers to these areas. The formation of biofilm and the internalization of pathogens also limit the effectiveness of chemical sanitizers. Furthermore, there are concerns about the residues or byproducts of certain chemical sanitizers. The reluctance in consumer acceptance and concerns about safety hamper the wide adoption of some physical intervention technologies, such as ionizing irradiation. Biocontrol agents may require the use of environmental conditions (such as elevated temperature) that shorten the shelf life of fresh produce. A consumer-friendly and effective technology is needed for fresh and freshcut produce. Thermal treatment, a relatively simple, nonchemical alternative, has been used by humans for thousands of years, even though the ways that heat is generated have changed. Heat treatments do not pose significant health risks from chemical residues and, as a result, are appealing to consumers and can be used to process organic produce. For use on fresh produce, the conditions (temperature and time) of thermal treatments have to be developed so that the quality of fresh produce is maintained while achieving reductions of foodborne pathogens. As an alternative to replace the use of chemical treatments, mild heat has been used for many years as a nonchemical postharvest treatment to control decay and to disinfest various fruits and vegetables (Couey 1989; Lurie 1998; Fallik 2004; Hong and others 2007; Hu and Tanaka 2007). Furthermore, thermal treatment also has been used to reduce the development of chilling injury for chilling-sensitive fruits and vegetables during storage (Porat and others 2000) and to delay the ripening process by reducing ethylene production and respiration (Paull and Chen 2000). In addition, heat treatment also extends the shelf life of lettuce by inhibiting activity of phenylalanine ammonia lyase and by inducing heat shock proteins, resulting in the reduction of phenolics accumulation and tissue browning (Loaiza-Velarde and others 1997). Thermal treatment is commonly applied in the form of hot water, high-temperature forced air, or steam (vapor heat) to control insects and fungi in postharvest fresh fruits 241

242

Section III. Postharvest Interventions

and vegetables (APHIS 1993; Couey 1989). Steam treatment has been applied mainly for insect control, and hot dry air has been used for both fungal and insect control (Lurie 1998). Steam and forced hot-air treatment systems are less damaging to commodities and more versatile than hot water; however, hot water is preferred for most applications because it is more efficient than air in delivering thermal energy to fresh produce. The cost of a typical commercial hot-water technology system is significantly less than that of a commercial steam treatment system (Fallik 2004). Hot-water treatment has been used commercially for controlling decay and insects in sweet peppers, mangos, melons, tomatoes, and many other fruits, most noticeably in Israel and Central and South America (Geysen and others 2006). In a typical application, hot water is applied at temperatures between 43 and 53 °C for periods of several minutes up to 2 h for quarantine treatments (Fallik 2004). In addition to hot-water immersion, a hot-water rinsing and brushing system has been developed (Fallik 2004). Because of relatively lower temperatures used in the systems for the purpose of disinfestation, ripening inhibition, and decay control, the systems may reduce human pathogens only marginally. However, the systems may be modified to enhance microbial safety of fresh fruits and vegetables. Other forms of thermal treatments such as microwave and radio frequency (RF) treatments have also been investigated for controlling insects in many fruits (Wang and others 2003). The effectiveness of thermal treatments to control decay, disinfest produce, and reduce physiological disorders has been reviewed previously (Fallik 2004; Geysen and others 2006; Lurie 1998; Paull and Chen 2000). This chapter focuses on use of thermal treatment to improve microbial safety and extend shelf life of fresh and fresh-cut fruits and vegetables.

Thermal Treatment Fundamentals Heat Transfer During thermal treatment of any kind of food, heat is transferred by a combination of conduction, convection, and radiation, depending on the source and types of the heating medium. The objective of thermal treatment is to increase the temperature of the product to a point that causes lethal damage to the bacteria of concern without causing irreversible physical damage to the product. Because most fruits and vegetables are solid in nature, internal conduction heating is the predominant mode of heat transfer during thermal treatment. When hot water, steam, or a mixture of steam/air is used as a heating medium, convection heating is involved. If an infrared heater is used, the heat would be transferred to the produce surface by radiation. In a typical industrial operation, hot water, steam, or a mixture of steam/air is used; the process of heat transfer is unsteady; and the transient heat transfer for any shape of food is governed by the following: ∂T k ⎛ ∂ 2T ∂ 2T ∂ 2T ⎞ = + + ⎜ ⎟ ∂t ρC p ⎝ ∂ 2 x ∂ 2 y ∂ 2 z ⎠

Eq. 13.1

T ( x, y, x, t0 ) = T0

Eq. 13.2

∂T k  = −h(T − Ta ) ∂n

Eq. 13.3

Improving Microbial Safety Using Thermal Treatment

243

In those equations, T is the temperature at any given location (x, y, z) at any given time (t) after the heat treatment is initiated. T0 and Ta are the initial and ambient temperatures. The transfer of heat to the interior of food is affected by its physical properties, k (thermal conductivity), ρ (density), and Cp (specific heat). All three parameters determine the thermal diffusivity, K/(ρCp), which describes how fast heat is transferred. The transfer of heat to fruits and vegetables is affected by the initial temperature (Eq. 13.1) and the physical condition of the heating medium (Eq. 13.2).  dT dn is the temperature derivative normal to the boundary. The parameter h is the surface heat transfer coefficient, governing the effectiveness of convective heat transfer from the heating medium to the surface of fruits and vegetables. The heat transfer equation (Eq. 13.3), together with its initial and boundary conditions, is usually solved by numerical methods (see case study below). Models may be developed to design a thermal process for a specific type of fresh produce to achieve effective surface decontamination with minimal quality damage (Scheerlinck and others 2004). Thermal Inactivation Kinetics The inactivation of microorganisms on fruits and vegetables by heat usually follows first-order kinetics, which means that the log counts, log(C), of bacteria decrease linearly at constant temperature conditions: log(C ) = log(C0 ) −

t D

Eq. 13.4

In Equation 13.4, D is a measure of time needed to achieve a 1-log reduction of the bacterial counts under a constant temperature condition. The D value is dependent on temperature, and the relation between the two can be expressed as the following (Eq. 13.5): log ( D ) = log ( D0 ) −

T z

Eq. 13.5

The parameter z depicts the effect of temperature on the D values, and is a measure of the increase in temperature needed to cause a 90% reduction in the D values. Together with the temperature history, the D and z values are used to estimate the extent of bacterial destruction during a thermal process: t

LRD = ∫ 0

10

T − Tref z dt Dref

Eq. 13.6

In Equation 13.6, LRD represents the total log reduction in bacterial counts during a thermal process, where T is temperature, and Dref is the D value at a reference temperature Tref. It is apparent that the total bacterial reduction is affected by the D and z values of the target microorganism and the temperature history at the location of interest in the food. One major disadvantage of thermal processing is that heat applied to kill foodborne pathogens and spoilage microorganisms usually also causes quality damage to fresh

244

Section III. Postharvest Interventions

fruits and vegetables. In most cases, a short thermal treatment (usually referred to as blanching) is used as a pretreatment to inactivate enzymes in fresh fruits and vegetables before dehydration or further thermal processing (canning). Therefore, only a limited amount of thermal resistance data is available concerning foodborne pathogens in fresh fruits and vegetables. Some data were actually obtained from juices of fruits (Mazzotta 2001a). Table 13.1 lists the D and z values for Escherichia coli O157 : H7, Salmonella spp., and Listeria monocytogenes in selected fruits and vegetables. It is readily noted from Table 13.1 and Figure 13.1 that the common foodborne pathogens (E. coli O157 : H7, Salmonella spp., and L. monocytogenes) are in general not very heat resistant, and can be destroyed easily by mild heat. Figure 13.1 illustrates the time needed to achieve a 5-log reduction in E. coli O157 : H7 in single-strength orange juice (the most heat resistant shown in Table 13.1). Apparently, a very short time (<1 s) is needed to achieve a 5-log reduction in the bacterial counts if the temperature of the products can be increased to above 70 °C. If the bacteria survive after a food has been heated at or above 70 °C for more than 1 s, it is usually an indication that bacteria are located inside of the food at an internal temperature substantially lower than the surface temperature, which can be easily illustrated in the following case study.

Time to achieve 5-log reduction (min)

1000 100 10 1 0.1 0.01 0.001 0.0001 0.00001 0.000001 50

60

70

80

90

100

T (°C) Figure 13.1. Time (min) needed to achieve a 5-log reduction of E. coli O157 : H7 in single-strength orange juice (calculated based on Table 13.1 and Eq. 13.6).

Table 13.1. Thermal resistance of certain foodborne spoilage and pathogenic microorganisms frequently associated with fruits and vegetables (modified from Mazzotta 2001a1,b2) Organisms

Substrate/Condition

Escherichia coli O15:H71 (stationary phase)

Single-strength apple juice

Single-strength orange juice

Single-strength white grape juice

Salmonellas1 (stationary phase composite of S. Typhimurium and S. Enteritidis) )

Single-strength apple juice

Single-strength orange juice

Single-strength white grape juice

Listeria monocytogenes1 (stationary phase)

Single-strength apple juice

Single-strength orange juice

Single-strength white grape juice

Listeria monocytogenes2 (stationary phase composite)

Onion

Broccoli

Peppers

Mushroom

Peas

D Values (Min) D56 °C 4.1 D58 °C 1.9 D60 °C 0.8 D56 °C 7.5 D58 °C 3.2 D60 °C 1.1 D56 °C 4.0 D58 °C 1.6 D60 °C 0.7

z (°4C)

D56 °C 1.21 D60 °C 0.23 D62 °C 0.11 D56 °C 2.52 D60 °C 0.45 D62 °C 0.22 D56 °C 1.38 D60 °C 0.28 D62 °C 0.10

5.7 5.4 5.3

D56 °C 1.59 D60 °C 0.45 D62 °C 0.17 D56 °C 2.05 D60 °C 0.43 D62 °C 0.21 D56 °C 2.29 D60 °C 0.59 D62 °C 0.29

6.3 6.0 6.6

D56 °C 0.8 D60 °C 0.23 D62 °C 0.10 D56 °C 2.3 D60 °C 0.62 D62 °C 0.39 D56 °C 3.9 D60 °C 0.92 D62 °C 0.31 D56 °C 5.0 D60 °C 0.69 D62 °C 0.30 D56 °C 5.2 D60 °C 1.04 D62 °C 0.41

6.7 7.8 5.5 4.9 5.5

5.6 5.9 5.3

245

246

Section III. Postharvest Interventions

Case Study This case study evaluated the feasibility of using thermal treatment to eliminate Salmonella Poona on the rind of cantaloupe melons (Solomon and others 2006). Finite different analysis was used to simulate the temperature distribution within cantaloupe melons subject to hot-water submersion heating at 65, 75, and 85 °C. Figures 13.2 and 13.3 show the simulated temperature history at different locations within cantaloupe melons during hot-water treatment. Figure 13.4 shows the average temperature (measured) at different depths from the surface of melons during heating. It is obvious that although the surface temperature can reach the hot-water temperature very rapidly, the internal temperature would rise at a much lower rate (depending on depth). For hot-water treatment at 65, 75, and 85 °C for 90 s, an average of 1.0-, 3.3-, and 4.5-log reductions, respectively, in the counts of S. Poona was achieved. The temperature at 5 mm below the surface of cantaloupes remained relatively cold (<40 °C), thus without causing damage to the edible portion of the melons (Annous and others 2004; Solomon and others 2006; Fan and others 2008).

70 60

T (°C)

50 40 30 20 10

65°C

0 0

100

200

300

400

500

600

t (s) Hot Water

Simulated 0 mm

Simulated 1 mm

Simulated 2 mm

Simulated 10 mm

Measured 10 mm

Figure 13.2. Simulated and measured temperature histories at different locations of a cantaloupe melon subjected to hot water immersion heating at 65 °C (Solomon and others 2006).

80 70 60

T (°C)

50 40 30 20 10

75°C

0 0

100

200

300

400

500

600

t (s)

90 80 70

T (°C)

60 50 40 30 20 10

85°C

0 0

100

200

300

400

500

600

t (s) 0.0 mm

1.0 mm

2.5 mm

5.0 mm

10 mm

Figure 13.3. Computer simulation of temperature histories at different locations of a cantaloupe melon subjected to hot-water immersion heating at 75 and 85 °C (Solomon and others 2006).

247

80 70

T (°C)

60 50

65°C

40

75°C 85°C

1 mm

30 20 10 0 0

50

100

150

200

t (s)

70 60

T (°C)

50 40 30 20 5 mm

10 0 0

50

100

150

200

t (s)

60 50

T (°C)

40 30 20 10

10 mm

0 0

50

100

150

200

t (s)

Figure 13.4. The average temperature at different depths from the rind of cantaloupe melons subjected to hot-water submersion heating at 65, 75, and 85 °C (Solomon and others 2006).

248

Improving Microbial Safety Using Thermal Treatment

249

Hot-Water Treatment Vegetables Lettuce Several studies have been conducted to investigate the effectiveness of warm-water treatment either alone or in combination with chlorine to reduce populations of bacteria in lettuce. Delaquis and others (1999) washed shredded lettuce for 3 min in 100 ppm chlorine at 4 or 47 °C. They reported a 3- and 1- log CFU/gm reduction in microbial loads of lettuce following washing in chlorine at 47 or 4 °C, respectively, as compared to the unwashed lettuce. The surviving microbial population, mainly psychotropic bacteria, was significantly lower on lettuce washed at 47 °C compared to unwashed lettuce or lettuce washed at 4 °C throughout 15 days of storage. In a follow-up experiment (Delaquis and others 2000), the effect of the two treatments on the population of inoculated E. coli O157 : H7 and L. monocytogenes and the fate of the two pathogens on shredded lettuce during storage at 1 and 10 °C were investigated. Results showed that washing in cold, chlorinated water immediately after inoculation reduced the E. coli O157 : H7 population by 1 log CFU/g. The antimicrobial effect of chlorinated water was enhanced at 47 °C, resulting in a reduction of the pathogen by more than 2 log CFU/g. The population of L. monocytogenes was reduced by 1 log CFU/g by chlorinated water, either at 4 or 47 °C. Although the cold chlorinated water was detrimental to the survival of E. coli O157 : H7 and L. monocytogenes at both storage temperatures, washing in warm chlorinated water favored the growth of both pathogens in lettuce stored at 10 °C, suggesting the importance of appropriate handling and storage temperature during retail display and distribution. Warm-water washing seemed to improve the quality of shredded lettuce because the texture and appearance of lettuce washed at the elevated temperature were better than those washed in warm chlorinated water (Delaquis and others 2000). However, a chlorinaceous off-odor was detected by some panelists, and the authors speculated that residual chlorine in the lettuce could account for the off-odor. Li and others (2001a) reported that dipping fresh-cut iceberg lettuce for 1.5 min in water at 50 °C or water containing 20 ppm free chlorine at 20 or 50 °C significantly reduced the initial population of psychrotropic and mesophilic aerobic microflora by 1.73–1.96 log CFU/g as compared to nontreated samples. Similar treatments of freshcut lettuce, artificially inoculated with E. coli O157 : H7, resulted in similar reductions in E. coli populations (Li and others 2001b). Dipping lettuce at 50 °C did not result in significantly greater reductions in the populations of E. coli O157 : H7 compared to the treatments at 20 °C. The bacterial populations, independent of the temperature and the chlorine concentration, increased during the subsequent 18-day storage at 15 °C. However, warm-water treatment (50 °C) enhanced microbial growth at 4 and 7 days. During subsequent storage at 5 °C, the population of E. coli O157 : H7 decreased steadily in all treated samples. It seems that the bacteria grow faster in samples treated with warm water during storage at harsh temperatures. Two factors may have contributed to the increase in the population of E. coli O157 : H7 during storage. First, the warm-water treatment may have eliminated some competing background microflora, which favors the growth of E. coli O157 : H7. Secondly, the leakage of

250

Section III. Postharvest Interventions

electrolyte after warm-water treatment may provide an easier access of the nutrients for the growth of bacteria. Fan and others (2003) studied the effect of combining warm-water treatment with irradiation treatment on reducing the adverse affect of irradiation treatment alone on lettuce quality. Lettuce dipped at 47 °C for 2 min and irradiated at 0.5 and 1 kGy had better overall visual quality and less tissue browning than the corresponding irradiated samples dipped at 5 °C. A dipping temperature of 50 °C resulted in injury to lettuce tissue. Hydrogen peroxide (H2O2) and organic acids have been applied at an elevated temperature to enhance treatment efficacy. Lin and others (2002) found that the combinations of 1.5% lactic acid and 1.5% H2O2 at 40 °C for 15 min reduced E. coli O157 : H7 and S. Enteritidis by 4 log CFU/g and L. monocytogenes by 3 log CFU/g on fresh-cut iceberg lettuce. However, the sensory quality of lettuce deteriorated following the treatment. However, treating lettuce leaves with 2% H2O2 at 50 °C (1 or 1.5 min) was effective not only in reducing pathogenic bacteria but also in maintaining sensory quality for up to 15 days. Up to 4-log reduction in E. coli O157 : H7 and S. Enteritidis were achieved with the 2% H2O2 treatment, whereas a 3-log reduction in L. monocytogenes was observed. Hydrogen peroxide residues were undetectable (the minimum level of sensitivity was 2 ppm) on lettuce surfaces after the treated lettuce was rinsed with cold water and centrifuged with a salad spinner. The authors concluded that the treatment of lettuce with 2% H2O2 at 50 °C for 1 min is effective in reducing the populations of foodborne pathogens and maintaining the product quality. A consumer acceptance study showed that lettuce treated with 2% H2O2 at 50 °C for 1 min compared favorably to untreated lettuce with regard to sensory properties (McWatters and others 2002). Wei and others (2005) reported a study in which a field salad vegetable (Valerianella locusta) inoculated with L. innocua and S. Typhimurium was treated with hot water (45–50 °C), and H2O2 (1–4%) for treatment times of 1–3.5 min. It was observed that H2O2 at concentrations of more than 2% caused deterioration of overall visual quality of the salad. The populations of L. innocua and S. Typhimurium in the salad were reduced by 2.5 and 2.6 log CFU/g, respectively, immediately after treatment with 2% H2O2 at 49 °C for 1 min). A 3.0-log reduction in L. innocua and 2.5 log in S. Typhimurium was observed after 7 days storage at 4 °C (Wei and others 2005). Broccoli and Beans Stringer and others (2007) reported a study in which broccoli florets and beans were washed for 90 s in tap water at either 20 °C or 52 °C and then stored at 7 and 10 °C. The hot-wash treatment improved the initial appearance of the vegetables and resulted in a significant reduction in the populations of all groups of endogenous microflora, including bacteria, yeast, and mold counts on the vegetables. However, three pathogens (L. monocytogenes, B. cereus, and E. coli O157 : H7), inoculated after treatment, grew faster in hot-washed (52 °C, 90 s) broccoli than in the equivalent product washed at 20 °C (Stringer and others 2007). Therefore, even though hot-water washing of broccoli florets and beans can be used to reduce bacteria, yeast, and mold populations, it can also lead to more rapid and extensive growth by pathogens and spoilage organisms if posttreatment contamination occurs after the heat treatment.

Improving Microbial Safety Using Thermal Treatment

251

Green Onions Cantwell and others (2001) studied the effects of heat treatment on quality and microbial population of minimally processed green onions (leaf blades and stalks). They found that hot-water treatment for 2 min at 55 °C or 4 min at 52.5 °C effectively controlled postharvest leaf extension growth (a product defect) and reduced the microbial population of green onions, while maintaining the visual quality of the leaves during storage. Kim and others (2005) treated fresh-cut green onions (leaf blades) with warm water (50 °C for 20 s) alone, and in combination with irradiation at 0, 0.5, 1.0 and 1.5 kGy. The treated green onions were stored at 4 °C for 14 days. The warm-water treatment alone reduced the total plate count by 0.9 log CFU/g initially, but the beneficial effect disappeared during storage. Significantly higher scores of overall visual quality and greenness and lower off-odor scores for irradiated samples were recorded during sensory tests, regardless of the warm-water dip. With the test conditions used in the study, the warm-water treatment did not provide added benefits for quality improvement. Sprouts and Watercress Park and others (1998) reported reductions of microbial populations by 2.2 and 4.4 log CFU/g in soybean sprouts and watercress, respectively, after a 30 s hot-water dip at 60 °C. In summary of the studies on vegetables, it is obvious that warm-water treatment alone in the range of 45–50 °C reduced populations of E. coli O157 : H7 and L. monocytogenes by only 2 log CFU/g at the most, even when combined with chlorine. Therefore, to achieve greater reductions of foodborne pathogens on produce, higher temperatures are required. However, higher temperatures may cause tissue damage (Loaiza-Velarde and others 1997; Fan and others 2003). Therefore, the sensitivity of vegetables to heat injury prohibits the use of hot-water treatment to obtain dramatic reductions in populations of foodborne pathogens on contaminated fresh-cut vegetables. Hot water in combination with other sanitizers (such as hydrogen peroxide) or irradiation may achieve greater reductions of foodborne pathogens while maintaining product quality. Fruits Cantaloupe Pathogens, when present on the surface of whole fruits or vegetables, can be transferred to the fresh-cut produce during processing (cutting, peeling, etc.). Melons (mostly cantaloupes) are one of the groups of produce that are most frequently associated with outbreaks and contamination with foodborne pathogens (Fan and others 2006, 2008). The high rates of pathogen contamination associated with melons highlight the need for effective interventions for both whole and cut melons. Lamikanra and others (2005) immersed whole cantaloupes into 50 °C water for 60 min before being processed into cubes. Sensory evaluations indicated that the heat treatment increased intensities of desirable attributes, such as fruity melon and sweet aromatic flavors, and reduced undesirable flavors such as musty, sour, bitter, chemical, and fermented flavors. The treatment also reduced the initial total microbial count and

252

Section III. Postharvest Interventions

population of lactic acid bacteria by 3 log CFU/g after 8 days storage at 10 °C. Ukuku and others (2004b) reported that treating whole cantaloupes for 1 min with hot water (70 or 97 °C) caused 2.0 or 3.4 log CFU/cm2 reductions of Salmonella, respectively, on whole cantaloupe surfaces, and treatment with 5% H2O2 (70 °C, 1 min) led to a 3.8 log CFU/cm2 reduction of Salmonella. Vacuum Steam Vacuum (VSV), developed by USDA scientists, employs a short exposure of food to vacuum to remove insulating fluids on the surface, followed by a quick burst of steam that rapidly transfers energy to the food (Morgan 1994; Morgan and others 1996). Then a second exposure to vacuum cools the product surface to prevent thermal damage. The process time is on the order of 1 to 2 s. Ukuku and others (2004a) treated whole cantaloupes with a VSV processor, and fresh-cut pieces were prepared from the treated samples. The VSV treatment resulted in about 1.0-log reduction of aerobic mesophilic bacteria, 2.0-log reductions of yeasts and molds, and 1.5-log reductions of Pseudomonas spp. on cantaloupe surfaces. VSV treatment significantly reduced the transfer of yeasts and molds and Pseudomonas spp. from whole cantaloupe surfaces to fresh-cut pieces during preparation (P <0.05). The texture and color of fresh-cut pieces prepared from VSV-treated whole melons were similar to those of nontreated controls. The results indicated that using VSV process to reduce the surface populations of yeasts and molds and Pseudomonas spp. on whole cantaloupes will reduce subsequent transfer of these microbes to fresh-cut pieces and enhance the microbial quality of the fresh-cut product. Sapers and Sites (2003) reported that treating cantaloupes with 1% H2O2 at 40 °C was ineffective against E. coli NRRL B-766, a surrogate for S. Poona. Ukuku (2006) sanitized whole cantaloupes with 200 ppm chlorine, 2.5% H2O2 solution, or hot water (96 °C) for 2 min, and then stored the fruit at 5 °C for 24 h. The hot-water treatment reduced the microbial populations on the cantaloupe surface by approximately 4.9 log; only 2.6-log reductions were observed in the cantaloupes treated with H2O2 or chlorine solutions. Cantaloupes were reinoculated with Salmonella spp. following sanitation or hot-water treatments and were stored for up to 7 days at 5 °C. Higher counts of Salmonella were recovered from hot-water–treated cantaloupes than those from cantaloupes treated with chlorine or hydrogen peroxide. The results showed that hotwater–treated cantaloupes are susceptible to recontamination during subsequent handling (Ukuku 2006). Solomon and others (2006) found that treatment with water at 85 °C for 60 and 90 s resulted in reductions of up to 4.7 log CFU/cm2 of Salmonella Poona on the surface of cantaloupes. However, melons treated at 85 °C for 90 s were noticeably softer than those treated for 60 s. Thermal penetration profiles were measured, and computer simulations were conducted to verify the effect of hot-water treatment conditions on the internal temperatures of cantaloupe melons. Experimental and simulation data indicated that the internal temperature of melons treated with hot water did not increase as rapidly as the surface temperature (Figs. 13.3, 13.4). Regardless of the process temperature used or the temperature of the edible flesh, 10 mm from the cantaloupe surface remained at least 40 °C cooler than the surface temperature of the melon Annous and others (2004) developed a pilot-scale surface pasteurization process for treatment of whole cantaloupes. Using this process, they reported that surface

Improving Microbial Safety Using Thermal Treatment

253

Table 13.2. The effect of hot-water pasteurization on total plate count (TPC) and yeast and mold count, as well as the population of inoculated S. enterica serovar Poona and E. coli ATCC 25922 on the fruit rind Treatments Not washed Water wash Chlorine (20 μg/ml) 76 ° water wash

TPC 5.1 ± 0.6 a 5.0 ± 0.5 a 4.8 ± 0.8 a 3.9 ± 0.4 b

Yeast and Mold 4.4 ± 0.4 a 4.4 ± 0.5 a 3.7 ± 0.7 b 1.5 ± 0.7 c

S. Poona 6.2 ± 0.6 a 6.3 ± 0.2 a NT 0.8 ± 1.0 b

E. coli 5.9 ± 0.5 a 5.5 ± 0.1 a NT 0.3 ± 0.3 b

For the study of S. Poona and E. coli, cantaloupes were dip-inoculated with E. coli and Salmonella Poona for 5 min, allowed to air dry for 2 h, and stored at room temperature for 24 h prior to treatment at room temperature or 76 °C for 3 min (Annous and others 2004). For the study of TPC and yeast and mold, whole fruit were submerged into 10 °C water for 20 min, 20 ppm chlorine at 10 °C for 20 min, or 76 °C water for 3 min (Fan and others 2008). E. coli and S. Poona populations were enumerated on MacConkey and xylos lysine tergitol-4 agar medium, respectively. Data represent the means ± standard deviations (n = 4–9). NT: not tested.

pasteurization with hot water at 76 °C for 3 min resulted in more than 5 log CFU/cm2 reductions in S. enterica serovar Poona and E. coli populations (Table 13.2). Cantaloupes that were surface pasteurized and stored at 4 °C for 21 days were firm and had no visible mold growth; the control samples became soft and moldy. These results indicated that the hot-water treatment enhanced the microbial safety of cantaloupes and extended the shelf life of this commodity. Fan and others (2008) compared the quality and shelf life of fresh-cut cantaloupe prepared from hot-water–treated (76 °C, 3 min) whole fruit with that prepared from cold (10 °C, 20 min) or chlorinated water (20 ppm, 10 °C, 20 min). The quality and microbial populations of fresh-cut cantaloupes prepared from whole fruit were also analyzed after 1, 6, 8, 10, 13, 16 and 20 days of storage at 4 °C. Results showed that the yeast and mold count on the rinds of whole fruits were reduced by 2.9 log CFU/ cm2 (Table 13.2). However, the chlorine or cold-water washes did not result in significant reductions in the microbial population. Although the hot-water treatment of whole cantaloupe lowered the total plate count (TPC) on the rind, it did not reduce TPC on fresh-cut cubes (Table 13.3) at day 1. However, in the subsequent storage periods (day 13–20), the fresh-cut fruit prepared from hot-water–treated cantaloupes were about 2 log CFU/g lower in TPC than that obtained from the other two treatments. This observation was directly attributable to the lower counts and slower growth of bacteria on the fresh-cut cantaloupes prepared from hot-water–treated samples during storage. Soluble solids content, ascorbic acid content, fluid loss, and aroma and appearance scores were not consistently affected by either hot water or chlorine treatment. The results suggested that hot-water pasteurization of whole cantaloupes resulted in a longer shelf life of fresh-cut fruit without negatively effecting quality. A combination of whole fruit surface sanitation by hot water (76 °C for 3 min) with low dose (0.5 kGy) gamma irradiation of fresh-cut cantaloupes further reduced the microflora of cut fruit, compared with either treatment alone (Fan and others 2006).

254

Section III. Postharvest Interventions

Table 13.3. Total plate count and yeast and mold count of fresh-cut cantaloupes prepared from hot-water–treated fruit TPC Storage Time (Day) 0 1 6 8 10 13 16 20 LSD

20 °C Water 2.0 ± 0.7 2.9 ± 0.7 3.8 ± 0.8 4.1 ± 0.9 4.7 ± 1.2 5.6 ± 1.8 5.9 ± 1.9 6.7 ± 1.9 1.5

Chlorine 2.4 3.1 3.6 4.2 4.7 5.3 6.0 6.3 2.0

± ± ± ± ± ± ± ±

0.7 0.9 0.7 1.3 1.7 1.5 1.9 1.9

Yeast and Mold 76 °C Water 2.4 ± 1.0 3.1 ± 1.0 2.7 ± 1.3 3.4 ± 0.6 2.9 ± 1.1 3.2 ± 0.5 3.4 ± 0.5 4.4 ± 1.7 1.1

20 °C Water 1.5 ± 0.4 2.2 ± 0.9 2.6 ± 0.6 2.5 ± 0.5 3.0 ± 0.8 2.8 ± 0.9 2.8 ± 0.7 3.4 ± 0.8 0.8

Chlorine 1.8 ± 2.1 ± 2.2 ± 2.3 ± 2.8 ± 2.4 ± 2.6 ± 2.8 ± 1.0

0.2 0.8 0.8 0.8 1.0 1.1 0.9 1.1

76 °C Water 1.4 ± 0.5 1.9 ± 0.8 1.8 ± 1.0 2.1 ± 1.1 2.0 ± 1.0 1.6 ± 0.9 2.0 ± 0.8 1.8 ± 1.2 1.0

Data represent the means ± standard deviations (n = 9). Whole cantaloupes were submerged into 10 °C water for 20 min, 20 ppm chlorine at 10 °C for 20 min, or 76 °C water for 3 min (Fan and others 2008). Pieces prepared from whole cantaloupe were stored in clamshell containers at 4 °C.

Apples Hot water (80 or 95 °C) treatment of whole apples for 15 sec reduced E. coli O157 : H7 by more than 5 log CFU/g (Fleischman and others 2001). The treated apples were used to produce cider. Heat damage to the apple skin did not negatively affect quality of the apple cider. The efficacy of H2O2 solutions at different concentrations and temperatures in decontaminating apples inoculated with foodborne pathogens was also investigated (Sapers and others 2002; Sapers and Sites 2003). Apples inoculated with a nonpathogenic E. coli ATCC 25922 and E. coli O157 : H7 were washed with 1% H2O2 at 20, 40 or 50 °C for up to 30 min. Population reductions approaching 3 log were obtained with all treatments. Comparable reductions were obtained with apples inoculated directly with 3 strains of E. coli O157 : H7 associated with cider outbreaks. Venkitanarayanan and others (2002) spot-inoculated apples with E. coli O157 : H7, Salmonella Enteritidis, and L. monocytogenes near the stem end and treated the inoculated fruit in 1.5% lactic acid plus 1.5% hydrogen peroxide solutions for 15 min at 40 °C. The treatment reduced bacterial pathogens on fruits by more than 5.0 log CFU per fruit, whereas washing in water decreased the pathogens by only 1.5 to 2.0 log CFU per fruit. The sensory and qualitative characteristics of apples treated with the chemical wash were not adversely affected by the treatment. Oranges Oranges inoculated with E. coli ATCC 25922 were immersed into hot water (80 °C) for 1 or 2 min (Pao and others 2001). Greater than 5-log reductions in the population of E. coli were achieved in the juice processed from the hot-water–treated fruit in comparison to the E. coli level detected in the juice directly prepared from inoculated fruit. Hot-water treatment can be used to pasteurize whole oranges for the further processing of juice or fresh-cut oranges. However, it is unclear whether the treatment

Improving Microbial Safety Using Thermal Treatment

255

can be used for the marketing of whole fruit due to potential thermal damage to the rind. Venkitanarayanan and others (2002) applied a process to treat orange fruit with 1.5% lactic acid plus 1.5% hydrogen peroxide solutions for 15 min at 40 °C and found that more than 5 log of reduction per fruit in the population of surface-inoculated E. coli O157 : H7, S. Enteritidis, and L. monocytogenes was achieved by such treatment. Grapes Mlikota Gabler and others (2005) found that treating grapes with water at 50 °C for 1 min significantly reduced grey mold incidence, and treatment with 35% (v/v) ethanol at 25 °C was better in reducing mold incidence than the hot-water treatment (50 °C, 1 min). Treatment with 35% ethanol at 50 °C was more effective in reducing the mold incidence than the other two treatments. It is unclear how effective these treatments would be in reducing populations of foodborne pathogens. Kou and others (2007) found that a significantly lower decay rate was observed in samples treated with hot water (45 °C, 8 min) than that in the control or in the samples treated with hot air (55 °C, 5 min) throughout the entire 14 days of storage. Color and texture were not significantly (P > 0.05) affected by either heat treatment or stem removal. However, the mild heat treatment could not be used to enhance microbial safety of grapes because of the inadequacy of the lower temperature (45 °C) in inactivating foodborne pathogens. Tomatoes A hot-water–brushing system has been developed and commercialized for treatment of tomatoes and other commodities to reduce decay and chilling injury (Fallik and others 2002). Fresh tomatoes have been frequently associated with major outbreaks of salmonellosis. Sapers and Jones (2006) dipped Salmonella-inoculated tomatoes into 200 ppm chlorine at 20 °C for 3 min, water at 20 °C for 3 min or at 60 °C for 2 min, 1% H2O2 at 20 °C for 15 min or at 60 °C for 2 min, and 5% H2O2 at 60 °C for 2, 3, or 5 min. No greater than 1.3 log in the bacterial counts were observed with the chlorine, hot-water, and 1% H2O2 treatments. Slightly higher reductions (∼2 log CFU/g) in bacterial counts were observed with the application of 5% H2O2 at 60 °C. However, no additional reduction was observed by increasing the treatment time or adding surfactants. The treatment also caused damage (skin darkening) to the fruit. Venkitanarayanan and others (2002) spot-inoculated S. enteritidis, E. coli O157 : H7, and L. monocytogenes onto the smooth surface of tomato fruit and then immersed the inoculated fruit in 1.5% H2O2 plus 1.5% lactic acid at 40 °C for 15 min. More than a 5-log (CFU/fruit) reduction in the population of all three pathogens was achieved without damage to fruit. Spraying tomatoes, inoculated with S. typhimurium and E. coli O157 : H7 on the surface and in the stem scar area, with 2% lactic acid at 55 °C resulted in >2.9 log CFU/g reductions in the bacterial counts (Ibarra-Sánchez and others 2004). In the study conducted by Sapers and Jones (2006) the interval between tomato contamination and sanitizing was 24 h or longer; a much shorter interval between inoculation and treatment was used in the studies conducted by Venkitanarayanan and others (2002) (1 h) and Ibarra-Sánchez and others (2004) (40 min). Sapers and Jones (2006) argued that the short time used in the studies conducted by Venkitanarayanan and others (2002) and Ibarra-Sánchez and

256

Section III. Postharvest Interventions

others (2004) may not be sufficient for pathogens to become attached onto the surface, internalized in the stem scar area, or incorporated in resistant biofilms, which may explain the large difference in the reduction of bacterial population among those studies. Therefore, when reporting results about inactivation of pathogens by any intervention technologies, incubation conditions and incubation time after inoculation should be indicated. Mangoes Whole mangoes were washed in water with or without chlorine (100 ppm) at 11.7 °C and 50 °C before slicing and packaging (Ngarmsak and others 2006). The washing process reduced the populations of microorganisms in the stem scar area and on the whole fruit, and the effect was enhanced by the addition of chlorine and the elevated temperature. However, none of the treatments prevented the transfer of microbial contaminants to the flesh during slicing, and the evidence of spoilage in the form of discrete fungal colonies was observed in samples stored for 1 week at 5 °C. Pears A washing system using a high-pressure spray, warm water (30 °C), a wetting agent, and rotating soft brushes was found to be effective in decay control without causing internal or external damage to pear fruit (Bai and others 2006). It is doubtful that the system would significantly reduce the population of pathogens due to the low temperature (30 °C) applied. Heat injury was observed on pear fruit treated with 40 and 50 °C water spray. Pineapples Wilson Wijeratnam and others (2005) used a hot-water dip (54 °C for 3 min) to treat pineapples inoculated with Chalara paradoxa (104 spores/ml), a fungus causing black rot. The treated fruit was free of disease when stored at 10 °C for 21 days followed by 48 h at ambient temperature (28 ± 2 °C). As a contrast, characteristic symptoms of the disease were observed in fruit that was inoculated and held under similar storage conditions. No significant difference was observed between hot-water-dip–treated and untreated controls with respect to flesh and shell color of fruit, ascorbic acid levels, and titratable acidity. A significant difference (p < 0.05) in total soluble solids (mean Brix of 14 °) was observed between hot-water–treated and untreated fruit (mean Brix of 11.5 °), irrespective of storage temperature. Sprouting Seeds Decontamination of alfalfa seeds by hot-water treatment in combination with chemicals such as chlorine or ozonated water has been investigated. It was demonstrated that it is not possible to reduce the numbers of Salmonella and E. coli O157 : H7 by more than 5 log CFU/g without affecting seed germination (Jaquette and others 1996; Scouten and Beuchat 2002; Sharma and others 2002; Suslow and others 2002). However, Weiss and Hammes (2005) found that hot-water treatment of inoculated mung bean seeds for 2–20 min at 55–80 °C, and radish and alfalfa seeds for 0.5–8 min at 53–64 °C reduced Salmonella and E. coli O157 : H7 by more than 5 log CFU/g, with germination rates of more than 95%. The variations among the studies on the

Improving Microbial Safety Using Thermal Treatment

257

effectiveness of hot-water treatment in inactivating pathogens on seeds may be due to the use of different inoculation methods (naturally contaminated versus artificially inoculated, and the drying times after inoculation) Considerations for the Commercial Application of Hot-Water Treatment Several studies (Delaquis and others 2000; Li and others 2001b; Ukuku 2006) have demonstrated that microorganisms on fresh fruits and vegetables that survived or recontaminated after hot-water treatments grew faster than those on cold-water– treated fruit during storage, particularly under abusive (higher) storage temperature. Therefore, it is essential to avoid recontamination of fruits after hot-water treatment. Furthermore, the fruits should be stored at temperatures that prevent the growth of pathogens. In addition, when warm fruit is immersed in cold water, the transitory pressure difference resulting from the temperature differential between the fruit and water (i.e., warm fruit and cold water) can lead to infiltration and internalization of bacteria (if present) into fruit such as apple, orange, tomato, and mango (Bordini and others 2007). Therefore, if the fruit is cooled with water after heat treatment, the cooling water must be treated with a sanitizing agent to assure that it is free of pathogens. If the hotwater–treated fruit is processed into fresh-cut products without cooling, the peelers, knives, cutting boards, and other contact surfaces have to be sanitized to avoid recontamination. The sensitivity of produce to heat injury will limit the application of thermal treatments. The sensitivity of fresh fruits and vegetables varies with the insulating properties; thickness and resistance to heat damage of the skin; and the growing conditions, cultivar, maturity, rate of heating, and subsequent storage (Paull and Chen 2000). For example, mature tomatoes are more susceptible to heat injury than the less mature fruit (Sapers and Jones 2006). As a result, whether thermal treatment can be applied to a specific commodity has to be studied individually with consideration of growing condition, maturity, etc.

Microwave, RF, and Infrared Microwave, RF, and infrared are systems of electromagnetic radiation that can be used to generate heat in a material. Therefore, energy can be transferred from the generators to a material without direct physical contact. The microwave frequencies allocated by the Federal Communications Commission (FCC) for the purposes of heating are 2450 and 915 MHz. The 2450 MHz frequency is primarily used for home ovens; 915 MHz is often used in commercial applications. RF is allowed at 13.56, 27.12, and 40.68 MHz. Two mechanisms are primarily responsible for microwave and RF heating: dielectric heating and ionic heating. The primary benefit of microwave and RF heating is that they can penetrate into the foods and require less time to come up to the desired process temperature than conventional heating in which heat is mainly transferred by conduction. Other advantages of microwave and RF heating systems are that the energies can be turned on or off instantly and the product can be pasteurized after being packaged (Datta and Davidson 2000). Microbial inactivation kinetics for microwave and RF are essentially the same as those of conventional thermal processing.

258

Section III. Postharvest Interventions

Infrared radiation has a wavelength longer than that of visible light, but shorter (longer frequencies) than that of microwaves. The wavelength of infrared radiation ranges from 780 nm to 1 mm. Near infrared wavelengths range from 780 nm to 1400 nm, medium infrared wavelengths are between 1400 nm and 3000 nm, and far infrared is above 3000 nm. Practically, the efficiency of infrared heating can be increased by matching the emitted wavelength and the absorption spectrum of the material to be heated. For example, the absorption spectrum for water has its maximum at around 3000 nm. Emission from medium-wavelength infrared is much better absorbed by water than near infrared or short-wave infrared radiation. Bacteria have a different cell wall composition (peptidoglycan) than plants (mainly polysaccharide, such as cellulose). The infrared absorbance spectrum of peptidoglycan is different from that of polysaccharide (Naumann and others 1982). Therefore, it may be possible to develop an infrared heater that emits at certain wavelengths that are mainly absorbed by the bacteria and other unique bacterial molecules anchored on the cell wall, resulting in selective heating of the bacteria. Another way of selective heating might be achieved by using a food-grade chemical that selectively binds to bacteria and has a unique absorption of infrared, microwave, or RF radiation. When exposed to certain wavelength(s) of radiation, only bacteria would heated up and plant cells would not be significantly affected, thus preserving the quality of treated fresh produce. RF Treatment Alfalfa seeds inoculated with Salmonella, E. coli O157 : H7, and L. monocytogenes were exposed to RF at 39 MHz and different electric field intensities (Nelson and others 2002). Although it was possible to achieve significant reductions in the populations of all three pathogens without affecting seed germination, it was impossible to obtain the desired levels (5 log) of pathogen reduction without significant damage to seed germination. RF thermal treatment was compared with chlorine wash and hot-water dipping in a study to improve the storability of carrot sticks (Orsat and others 2001). Carrot sticks were heated to 60 °C in less than 2 min in a parallel plate RF applicator. Results showed that the quality of the RF treated samples was greater than those treated with chlorine (100 ppm) or hot water (by immersing carrots in boiling water until the internal temperature reached 60 °C). The color and fresh taste were maintained in RFtreated carrot samples, in contrast to the control and hot-water–treated carrots. However, merely reducing the initial microbial load did not maintain the quality of carrot sticks for 14 days at 6 °C. In another study, apple fruit was initially exposed to 27.12-MHz RF energy at 12 kw for 2.75 min and then subjected to hot-water dips (48–50 °C) for different durations (Hansen and others 2006). The combined heat treatment resulted in peel and flesh discoloration, along with reduced firmness and increased external and internal damage. Microwave Microwave radiation has been extensively studied to disinfest fruits. For example, microwave heating, followed by steam treatment, has been successfully applied to

Improving Microbial Safety Using Thermal Treatment

259

treat mangoes to eliminate oriental fruit fly eggs (Varith and others 2007). Changes in physiochemical properties of mangoes, including color, titratable acid, soluble solid content, and firmness of the treated mangoes were not significantly different (P > 0.05) from untreated mangoes. The combined treatment resulted in less heat damage to mangoes than conventional steam treatment and greater than 90% reduction in preheating time. It would be interesting to study the combined effect of microwave heating with steam on the reduction of microorganisms in produce. Infrared There have been no publications about the use of infrared heating to inactivate foodborne pathogens on fresh and fresh-cut fruits and vegetables. Gomez Galindo and others (2005) treated carrot slices with far-infrared for 7 s (from a radiant surface of 538 °C) and found that the far-infrared radiation damaged only the first halfmillimeter of cells from the surface, and the carrot slices exhibited the texture characteristics of the raw tissue.

Conclusion There is no question that thermal treatment can reduce or even eliminate the population of foodborne pathogens on or in fresh produce. However, such treatments may cause damage to fresh produce if not optimized. The goal of any food safety intervention technology is to decontaminate food products while maintaining or minimizing the adverse effects to the treated product in terms of quality loss. Understanding the fundamental mechanism of thermal treatment and application of mathematical calculations and modeling will help design better and more effective thermal processes to inactivate the targeted pathogens and to minimize quality loss. In general, if the temperature at the site of pathogen attachment to produce reaches 70 °C, vegetative bacterial pathogens will definitely be inactivated. However, whether fresh produce can tolerate such a high temperature even for a brief period depends on the type of fresh fruit or vegetable and whether the produce will be further processed. For fresh produce (such as lettuce leaves) that is rather tender and delicate, thermal treatment may have a limited application because heat treatment can cause quality deterioration. For this type of produce, warm-water treatment may be combined with other intervention treatments (such as hydrogen peroxide and organic acids) to enhance microbial safety and to maintain product quality. If the produce has a thick protective layer (such as cantaloupe rind), damage caused by heating may be minimal. For products in which the surface of a fruit or vegetable will be removed during fresh-cut processing, damage to the surface may not be an issue, simply because the damaged surface layer will be discarded. However, prolonged surface thermal treatment may cause the internal temperature to increase to a point high enough to induce internal tissue damage in the product. Therefore, the commercial application of thermal processing to enhance fresh produce safety needs to be evaluated on a case-by-case basis. Furthermore, more studies are needed to evaluate the feasibility of microwave, RF, and infrared treatments in inactivating foodborne pathogens on/in fresh and fresh-cut fruits and vegetables.

260

Section III. Postharvest Interventions

References Annous BA, Burke A, Sites JE. 2004. Surface pasteurization of whole fresh cantaloupes inoculated with Salmonella Poona or Escherichia coli. J Food Prot 67:1876–1885. APHIS. 1993. Plant Protection and Quarantine Treatment Manual. United States Department of Agriculture. Animal and Plant Health Inspection Service. Bai J, Mielke EA, Chen PM, Spotts RA, Serdani M, Hansen JD, Neven LG. 2006. Effect of high-pressure hot-water washing treatment on fruit quality, insects, and disease in apples and pears Part I. System description and the effect on fruit quality of “d’Anjou” pears. Postharv Biol Technol 40:207–215. Bordini MEB, Ristori CA, Jakabi M, Gelli DC. 2007. Incidence, internalization and behavior of Salmonella in mangoes, var. Tommy Atkins. Food Contr 18:1002–1007. Cantwell MI, Hong G, Suslow TV. 2001. Heat treatments control extension growth and enhance microbial disinfection of minimally processed green onions. HortScience 36:732–737. Couey HM. 1989. Heat treatment for control of postharvest diseases and insect pests of fruits. HortScience. 24:198–202. Datta AK, Davidson PM. 2000. Microwave and radio frequency processing. J Food Sci Suppl: Kinetics of microbial inactivation for alternative food processing technologies 65:32–41. Delaquis PJ, Stewart S, Cliff M, Toivonen PMA, Moyls AL. 2000. Sensory quality of ready-to-eat lettuce washed in warm, chlorinated water. J Food Qual 23:553–563. Delaquis PJ, Stewart S, Toivonen PMA, Moyls AL. 1999. Effect of warm, chlorinated water on the microbial flora of shredded iceberg lettuce. Food Res Intern 32:7–14. Fallik E. 2004. Prestorage hot water treatments (immersion, rinsing and brushing). Postharv Biol Technol 32:125–134. Fallik E, Ilic Z, Tuvia-Alkalai S, Copel A, Polevaya Y. 2002. A short hot water rinsing and brushing reduces chilling injury and enhances resistance against Botrytis cinerea in fresh harvested tomato. Adv Hortic Sci 16:3–6. Fan X, Annous BA, Beaulieu JC, Sites JE. 2008. Effect of hot water surface pasteurization of whole fruit on shelf-life and quality of fresh-cut cantaloupe. J Food Sci 73:M91–M98. Fan X, Annous BA, Sokorai KJ, Burke AM, Mattheis JP. 2006. Combination of hot water surface pasteurization of whole fruit and low dose irradiation of fresh-cut cantaloupe. J Food Prot 69:912–919. Fan X, Toivonen PMA, Rajkowski KT, Sokorai K. 2003. Warm water treatment in combination with modified atmosphere packaging reduced undesirable effects of irradiation on the quality of fresh-cut iceberg lettuce. J Agric Food Chem 51:1231–1236. Fleischman GJ, Bator C, Merker R, Keller SE. 2001. Hot water immersion to eliminate Escherichia coli O157 : H7 on the surface of whole apples: thermal effects and efficacy. J Food Prot 64:451–455. Geysen S, Verlinden BE, Nicolaï BM. 2006. Thermal treatments of fresh fruit and vegetables. In: Improving the Safety of Fresh Fruit and Vegetables. Jongen W, ed. Boca Raton, FL: CRC Press, pp. 429–453. Gomez Galindo F, Toledo RT, Sjoholm I. 2005. Tissue damage in heated carrot slices. Comparing mild hot water blanching and infrared heating. J Food Engin 67:381–385. Hansen JD, Heidt ML, Watkins MA, Tang J, Wang S. 2006. Radio frequency hot water dips for postharvest codling moth control in apples. J Food Proc Preserv 30:631–642. Hong S, Lee H, Kim D. 2007. Effects of hot water treatment on the storage stability of satsuma mandarin as a postharvest decay control. Postharv Biol Tech 43:271–279. Hu W, Tanaka S. 2007. Effects of heat treatment on the quality and storage life of sweet potato. J Sci Food Agric 87:313–319. Ibarra-Sánchez LS, Alvarado-Casillas S, Rodríguez-García MO, Martínez-Gonzáles NE, Castillo A. 2004. Internalization of bacterial pathogens and their control by selected chemicals. J Food Prot 67:1353–1358. Jaquette CB, Beuchat LR, Mahon BE. 1996. Efficacy of chlorine and heat treatment in killing Salmonella Stanley inoculated onto alfalfa seeds and growth and survival of the pathogen during sprouting and storage. Appl Environ Microbiol 62:2212–2215. Kim HJ, Hao F, Stoyan TA, Fan X. 2005. Effect of sequential treatment of warm water dip and low dose gamma irradiation on the quality of fresh-cut green onions. J Food Sci 70:M179–185.

Improving Microbial Safety Using Thermal Treatment

261

Kou L, Luo Y, Wu D, Liu X. 2007. Effects of mild heat treatment on microbial growth and product quality of packaged fresh-cut table grapes. J Food Sci 72:S567–S573. Lamikanra O, Bett Garber KL, Ingram DA, Watson MA. 2005. Use of mild heat pre-treatment for quality retention of fresh-cut cantaloupe melon. J Food Sci 70:C53–C57. Li Y, Brackett RE, Chen J, Beuchat LR. 2001b. Survival and growth of Escherichia coli O157 : H7 inoculated onto cut lettuce before and after heating in chlorinated water, followed by storage at 5 or 15 °C. J Food Prot 64:305–309. Li Y, Brackett RE, Shewfelt RL, Beuchat LR. 2001a. Changes in appearance and natural microflora on iceberg lettuce treated in warm, chlorinated water and then stored at refrigeration temperature. Food Microbiol 18:299–308. Lin CM, Moon S, Doyle MP, McWatters KH. 2002. Inactivation of Escherichia coli O157 : H7, Salmonella enterica Serotype Enteritidis, and Listeria monocytogenes on lettuce by hydrogen peroxide and lactic acid and by hydrogen peroxide with mild heat. J Food Prot 65:1215–1220. Loaiza-Velarde JG, Tomas-Barbera FA, Saltveit ME. 1997. Effect of intensity and duration of heat shock treatments on wound-induced phenolics metabolism in iceberg lettuce. J Am Soc Hortic Sci 122:872–977. Lurie S. 1998. Postharvest heat treatment. Postharv Biol Technol 14:257–269. Mazzotta AS. 2001a. Thermal inactivation of Stationary-phase and acid-adapted Escherichia coli O157 : H7, Salmonella, and Listeria monocytogenes in fruit juices. J Food Prot 64:315–320. Mazzotta AS. 2001b. Heat resistance of Listeria monocytogenes in vegetables: evaluation of blanching processes. J Food Prot 64:385–387. McWatters K, Chinnan MS, Walker SL, Doyle MP, Lin CM. 2002. Consumer acceptance of fresh-cut Iceberg lettuce treated with 2% hydrogen peroxide and mild heat. J Food Prot 65:1221–1226. Mlikota Gabler F, Smilanick JL, Ghosoph JM, Margosan DA. 2005. Impact of postharvest hot water or ethanol treatment of table grapes on gray mold incidence, quality, and ethanol content. Plant Dis 89:309–316. Morgan AI. 1994. Method and apparatus for treating and packaging raw meat. United States Patent 5,281,428. Morgan AI, Radewonuk ER, Scullen OJ. 1996. Ultra high temperature, ultra short time surface intervention of meat. J Food Sci 61(6):1216–1218. Naumann D, Barnickel G, Bradaczek H, Labischinski H, Giesbrecht P. 1982. Infrared spectroscopy, a tool for probing bacterial peptidoglycan: potentialities of infrared spectroscopy for cell wall analytical studies and rejection of models based on crystalline chitin. Euro J Biochem 125:505–515. Nelson SO, Lu CY, Beuchat LR, Harrison MA. 2002. Radio-frequency dielectric heating of alfalfa seed for reduction of human pathogens. Paper Number 026001. 2002 ASAE Annual Meeting. Ngarmsak M, Delaquis P, Toivonen P, Ngarmsak T, Ooraikul B, Mazza G. 2006. Microbiology of fresh-cut mangoes prepared from fruit sanitized in hot chlorinated water. Food Sci Technol Intern 12:95–102. Orsat V, Gariépy Y, Raghavan GSV, Lyew D. 2001. Radio-frequency treatment for ready-to-eat fresh carrots. Food Res Intern 34:527–536. Pao S, Davis CL, Parish ME. 2001. Microscopic observation and processing validation of fruit sanitizing treatments for the enhanced microbiological safety of fresh orange juice. J Food Prot 64:310–314(5). Park WP, Cho SH, Lee DS. 1998. Effect of minimal processing operations on the quality of garlic, green onion, soybean sprouts and watercress. J Sci Food Agric 77:282–286. Paull RE, Chen NJ. 2000. Heat treatment and fruit ripening. Postharv Biol Technol 21:21–37. Porat R, Weiss B, Cohen L, Daus A, Aharoni N. 2000. Reduction of postharvest rind disorders in citrus fruit by modified atmosphere packaging. Postharv Biol Technol 33:35–43. Sapers GM, Jones DM. 2006. Improved sanitizing treatments for fresh tomatoes. J Food Sci 71(7):M252–M256. Sapers GM, Miller RL, Annous BA, Burke AM. 2002. Improved antimicrobial wash treatments for decontamination of apples. J Food Sci 67:1886–1891. Sapers GM, Sites JE. 2003. Efficacy of 1% hydrogen peroxide wash in decontaminating apples and cantaloupe melons. J Food Prot 68:1793–1797. Scheerlinck N, Marquenie D, Jancsók PT, Verboven P, Moles CG, Banga JR, Nicolaï BM. 2004. A modelbased approach to develop periodic thermal treatments for surface decontamination of strawberries. Postharv Biol Technol 34:39–52.

262

Section III. Postharvest Interventions

Scouten AJ, Beuchat LR. 2002. Combined effects of chemical, heat and ultrasound treatments to kill Salmonella and Escherichia coli O157 : H7 on alfalfa seeds. J Appl Microbiol 92:668–674. Sharma RR, Demirci A, Beuchat LR, Fett WF. 2002. Inactivation of Escherichia coli O157 : H7 on inoculated alfalfa seeds with ozonated water and heat treatment. J Food Prot 65:447–451. Solomon EB, Huang L, Sites JE, Annous BA. 2006. Thermal inactivation of Salmonella on cantaloupes using hot water. J Food Sci 71(2):M25–30. Stringer SC, Plowman J, Peck MW. 2007. The microbiological quality of hot water-washed broccoli florets and cut green beans. J Appl Microbiol 102:41–50. Suslow TV, Wu J, Fett WF, Harris LJ. 2002. Detection and elimination of Salmonella Mbandaka from naturally contaminated alfalfa seed by treatment with heat or calcium hypochlorite. J Food Prot 65:452–458. Ukuku DO. 2006. Effect of sanitizing treatments on removal of bacteria from cantaloupe surface, and recontamination with Salmonella. Food Microbiol 23:289–293. Ukuku DO, Fan X, Kozempel MF. 2004a. Effect of hot water and hydrogen peroxide treatments on survival of Salmonella and microbial quality of whole and fresh-cut cantaloupe. J Food Prot 69:1623–1629. Ukuku DO, Pilizota V, Sapers GM. 2004b. Effect of hot water and hydrogen peroxide treatment on survival of Salmonella and microbial quality of whole cantaloupe and fresh-cut cantaloupe. J Food Prot 67:432–437. Varith J, Sirikajornjaru W, Kiatsiriroat T. 2007. Microwave-vapor heat disinfestation on oriental fruit fly eggs in mangoes. J Food Proc Preserv 31:253–269. Venkitanarayanan KS, Lin CM, Bailey H, Doyle MP. 2002. Inactivation of Escherichia coli O157 : H7, Salmonella Enteritidis, and Listeria monocytogenes on apples, oranges, and tomatoes by lactic acid with hydrogen peroxide. J Food Prot 65:100–105. Wang S, Tang J, Cavalieri RP, Davis D, 2003. Differential heating of insects in dried nuts and fruits associated with radio frequency and microwave treatments. Trans ASAE 46(4):1175–1182. Wei H, Wolf G, Hammes WP. 2005. Combination of warm water and hydrogen peroxide to reduce the numbers of Salmonella Typhimurium and Listeria innocua on field salad (Valerianella locusta). Europ Food Res Tech 221:187–191. Weiss A, Hammes WP. 2005. Efficacy of heat treatment in the reduction of salmonellae and Escherichia coli O157 : H7 on alfalfa, mung bean and radish seeds used for sprout production. Europ Food Res Tech 221:180–186. Wilson Wijeratnam RS, Hewajulige IGN, Abeyratne N. 2005. Postharvest hot water treatment for the control of Thielaviopsis black rot of pineapple. Postharv Biol Tech 36:323–327.

14

Enhanced Safety and Extended Shelf Life of Fresh Produce for the Military Peter Setlow, Christopher J. Doona, Florence E. Feeherry, Kenneth Kustin, Deborah Sisson, and Shubham Chandra

Introduction Fresh fruits and vegetables (FFVs) provide rich sources of bioavailable nutrients, vitamins, minerals, phytochemicals, and dietary fiber for healthy diets, and they also provide the consumer with crisp and crunchy food textures and refreshing mouthfeel. As the consumption of fresh produce continues to increase and gain an increasing share of the ready-to-eat (RTE) foods market, the incidence of reported outbreaks of foodborne illnesses associated with eating fresh produce has also increased. In recent years, salads, lettuce, spinach, tomatoes, melons, sprouts, onions, apples, berries, and fruit juices have been implicated as the sources of illnesses, with the pathogens in these cases having been identified as Escherichia coli O157:H7, Salmonella spp, Listeria monocytogenes, viruses, and parasites. The outbreaks of E. coli O157:H7 in leafy greens (prepackaged fresh-cut lettuce, salad blends, and spinach) that caused >400 illnesses and 2 deaths and the recurrent outbreaks related to Salmonella in tomatoes (Morbidity and Mortality Weekly Report 2007) and jalapeño peppers have heightened the public’s awareness of the safety of fresh produce and prompted intensification of the dedicated efforts of growers, processors, and federal and state government agencies to enhance farm-to-fork agrimanagement practices and ensure fresh produce safety for the consumer. Microbial contaminants can be transmitted to fresh produce from a variety of sources, such as contaminated soils, contaminated irrigation water and lines, improperly composted animal manure used as fertilizer, bird droppings, intrusions of feral animals into growing fields, insufficient pesticide use, poor hygiene practices of employees, and unsanitary conditions during shipping, handling, processing, and distribution. Accordingly, perspectives to enhance the safety of fresh produce have produced a variety of initiatives ranging from improving farm management to exploring alternative processing technologies such as irradiation and chemical sanitizing agents. These enhancements include conducting independent third-party audits and improving farming practices, controlling the proximity of animal pastures and growing fields, avoiding or treating contaminated water sources for irrigation, increasing the monitoring of microbial hazards and traceability throughout the supply chain, enhancing sanitization strategies in processing plants and using potentially more efficient chemical agents (in flume washes and for cleaning processing and handling equipment) such as chlorine, hypochlorite, peroxyacetic acid, ozone, hydrogen peroxide, or chlorine dioxide, the agent of particular interest in this chapter. Another response to these outbreaks, however, has been a heightened sense of the vulnerability of the public to potential bioterrorist threats that could infiltrate the 263

264

Section III. Postharvest Interventions

nation’s food and water supplies and be promulgated through the supply chain. The increased complexity and globalized character of today’s supply chain also raises concerns that FFVs could act as vehicles for transmitting bioterrorist threats through the food supply that could endanger public safety. Adopting the safety practices mentioned above enhances the safety of fresh produce, and also helps safeguard the nation’s food supply from these potential sources of bioterrorism. The military faces a unique set of challenges in providing FFVs to personnel in different types of environments across distant global deployments on land or on the sea. FFVs are a highly valued part of a well-balanced diet for the soldier, and landbased foodservice operations often have the option of purchasing fresh produce from local vendors in host nations. Produce procured in these circumstances pose potential risks from microbiological hazards and agriterrorist threats, and fresh produce is routinely treated with chlorine rinses to inactivate microorganisms. FFVs are also an important part of all meals served on board U.S. Navy ships. Due to the inexorable process of ripening, fresh produce tends to be highly perishable, especially in improperly controlled storage conditions. In the past, transoceanic shipments of produce to distant global deployments incurred substantial economic losses due to improper storage and handling that contributed to premature spoilage. To ensure that military personnel have available the freshest and most nutritious produce, scientists at the U.S. Army Natick Soldier RD&E Center, Department of Defense Combat Feeding Directorate (NSRDEC/CFD) are investigating methods to ensure the microbiological safety and to extend the fresh shelf life of FFVs. They explore modified atmosphere packaging (MAP) to retard postharvest quality losses and extend the shelf life of select FFVs. They focus on a commercial MAP technology whose effectiveness was recently demonstrated on board the USS Ronald Reagan (CVN-76) for iceberg lettuce, bananas, broccoli, and romaine lettuce. They also explore novel technologies based on chemically generated chlorine dioxide, which has the capability of eliminating microbial hazards on FFVs, preventing enzymatic discoloration of certain fresh-cut fruits, and sanitizing food-processing and -handling equipment. The Portable Chemical Sterilizer (PCS, see Doona and others 2005) is a revolutionary device originally intended for the energy-independent sterilization of medical equipment in austere environments. The PCS could be used in place of chlorine rinses in military deployments or on-site during harvest at farming operations to ensure the safety of fresh produce. The Disinfectant-sprayer for Foods and ENvironmentally friendly Sanitation (D-FENS, Doona and others 2008a) is a convenient spray-and-wipe device that generates aqueous chlorine dioxide on-site to sanitize food contact surfaces. Chlorine dioxide is EPA-registered for uses in food-processing, -handling, and -storage plants, and it is approved for washing fruits and vegetables (Food and Drug Administration 1998). The PCS and D-FENS are “green” technologies that produce fewer dichlorinated by-products than technologies using chlorine rinses (bleach, OCl−) or chlorine gas (Cl2). The application of chlorine dioxide to foods requires strong enough doses to eliminate the most resistant microorganisms present, while also avoiding overprocessing or unwanted side reactions that could compromise appearance, flavor, or food quality (Rico et al. 2007). This type of balance was recently achieved using chlorine dioxide to eliminate spoilage microflora and >5-log reductions of pathogens (E. coli O157:H7,

Safety and Shelf Life of Fresh Produce for the Military

265

L. monocytogenes, and Salmonella spp) on a variety of FFVs without degrading product quality (Kim and others 2008; Mahmoud and Linton 2008; Park and others 2008; Gómez-López and others 2007; Mahmoud and others 2007). Tough-to-kill bacterial spores are especially resistant, and a number of spore-forming bacteria of Bacillus and Clostridium species inhabit soils, are commonly found on fresh produce, and have the potential to cause food spoilage and foodborne disease (Setlow and Johnson 2007). There is significant interest in devising methods for overcoming spore resistance to chemical agents, not only for disinfecting fresh produce, but also for sanitizing food-processing or -handling equipment, including removing biofilms, and for decontaminating biothreats (e.g., Bacillus anthracis). We focus our attention here on understanding the nature of the interactions of chlorine dioxide (and other chemical sanitizing agents) with bacterial spores and vegetative pathogens in the context of enhancing fresh produce safety using the PCS, D-FENS, and MAP technologies, all of which have potential dual-use applicability to benefit military and civilian consumers.

Bacterial Spore Resistance to and Inactivation by Chemical Agents Bacillus and Clostridium species can form bacterial spores, one of the most resistant of all life forms. Although such spores can remain metabolically dormant for long periods of time in environments with significant nutrients or appropriate stimuli present, spores can rapidly germinate and return to active growth. Spore resistance to and inactivation by particular chemical agents such as chlorine dioxide is due to a variety of factors and is best understood in terms of the novel structure of the spore (Fig. 14.1), which is very different from that of a growing cell, because a number of

Figure 14.1. Schematic representation of the structure of a dormant Bacillus spore. The various layers are not drawn to scale, and the exosporium is not present, or at most vestigial, in spores of many species.

266

Section III. Postharvest Interventions

spore layers have no counterparts in growing cells (Driks 2002a,b; Henriques and Moran 2007). Spore Morphology Spore layers, starting from the outside, begin with the exosporium, a balloonlike structure composed of proteins, some of which are glycoproteins (Driks 2002a; Henriques and Moran 2007). A large exosporium is not present in spores of all species, although it is present in spores of a number of species that can cause foodborne disease (e.g., B. anthracis and B. cereus). As far as is known, the exosporium plays little if any role in spore resistance to chemical agents. Beneath the exosporium is the spore coat. This structure is composed largely of protein, with ≥30 spore-specific proteins often in several distinct layers, and a number of coat proteins are cross-linked to one another (Driks 2002a; Driks 2002b; Henriques and Moran 2007). The coat plays an important role in imparting spore resistance to many chemical agents, including chlorine dioxide, sodium hypochlorite, ozone and other oxidizing agents, because spores in which the majority of coat protein is absent due either to a defect in coat assembly or chemical extraction are much more sensitive to these chemicals than are spores with intact coats (Fig. 14.2; see Genest and others 2002; Loshon and others 2001; Paul and others 2006, 2007; Setlow 2006; Young and Setlow 2003, 2004a,b). No specific individual protein or proteins have been implicated in spore chemical resistance, and it has been suggested that the large amount of coat protein (30–50%

Figure 14.2. Effect of the spore coat on spore resistance to sodium hypochlorite (A) and chlorine dioxide (B). B. subtilis spores either intact or after chemical removal of ≥50% of coat protein were treated in water with chlorine dioxide or sodium hypochlorite, the reagents were neutralized, and spore viability was determined. Similar results have been obtained when spores with a genetic defect in coat assembly are tested (Young and Setlow 2003). The symbols used are 䊉, decoated spores; and 䊊, intact spores. The data for this figure were taken from Young and Setlow (2003).

Safety and Shelf Life of Fresh Produce for the Military

267

of total spore protein) may simply act as “reactive armor,” detoxifying reactive chemicals before they penetrate to more sensitive and essential targets further within the spore (Setlow 2006). Notwithstanding the important role of the spore coat in spore resistance to some chemical agents, the coat plays a much smaller role in spore resistance with other chemical agents, such as hydrogen peroxide, sodium nitrite, and DNA alkylating agents, (Cortezzo and Setlow 2005; Riesenman and Nicholson 2000). Even in the presence of chemicals for which the coat is known to play a large role in spore resistance, coat-defective spores generally remain much more resistant to these chemicals than are growing cells (Cortezzo and Setlow 2005; Riesenman and Nicholson 2000). Thus, either residual coat protein in coat-defective spores or some other spore component plays major roles in spore resistance to such chemicals. Underlying the coat layer is the spore’s outer membrane. Insofar as is known this membrane plays no major role in spore chemical resistance, and this membrane may not even be an intact permeability barrier in mature spores (Setlow 2006). However, there are no mutants that specifically lack the outer spore membrane, and there is also no method for selectively removing this membrane without removing the spore coat. Indeed, methods for chemically removing the spore coat also remove many components of the outer membrane (Buchanan and Neyman 1986). Consequently, it is theoretically possible that the outer membrane could play some role, although presently not defined, in spore resistance to chemicals such as chlorine dioxide and hypochlorite. Beneath the outer membrane are two layers of peptidoglycan (PG), first the cortex and then the germ cell wall. The cortex contains the great majority of spore PG, and cortical PG has a number of structural features different from the PG in growing cells (Popham 2002). Although the cortex is almost certainly largely responsible for the reduction in the water content of the spore core that takes place late in spore formation (see below), it is not thought to play a major role in spore chemical resistance. The cortex is degraded in the first minutes of spore germination, and this degradation is essential for the germinated spore to develop into a growing cell (Setlow 2003). The second PG layer in spores is the germ cell wall (Popham 2002). The PG in this layer comprises only a small fraction of total spore PG and its structure appears identical to that of growing cell PG. The germ cell wall is not degraded during spore germination, and it becomes the cell wall of the outgrowing spore (Popham 2002). The second spore membrane, the inner membrane, is under the germ cell wall. Although the lipid composition of this membrane does not appear to be especially novel, it contains a number of proteins not found in growing cells, including a number of proteins needed for spore germination events (Setlow 2003). This membrane also has some novel properties. First, lipids in this membrane appear to be largely immobile as determined by measurement of the fluorescence redistribution after photobleaching of lipid probes in this membrane (Cowan and others 2004). However, shortly after initiation of spore germination, these lipid probes become fully mobile in the inner membrane. Second, this membrane appears to have an extremely low permeability in the dormant spore, not only to small charged molecules, but also to small hydrophobic molecules such as methylamine, and perhaps even to water (Cortezzo and others 2004; Cortezzo and Setlow 2005; Westphal and others 2003). The low permeability of this membrane is also lost shortly after the initiation of spore germination. It has been

268

Section III. Postharvest Interventions

suggested that the low permeability of this membrane to small molecules may be an important factor in spore resistance to chemical agents such as chlorine dioxide, hypochlorite, hydrogen peroxide, and other oxidizing agents (Setlow 2006). Indeed, increases in the permeability of this membrane result in increased sensitivity of spores to a variety of chemical agents that kill spores by acting on the spore core (see below; Cortezzo and Setlow 2005). To date the precise physical basis for the lipid immobility in and low permeability of the spore’s inner membrane are not known. However, it seems likely that these two phenomena are related. The final spore layer is the central spore core, the site of the spore’s DNA, transcription and translation machinery, and most spore enzymes. Novel features of the spore core include: 1. A relatively low water content. Growing cells have ∼80% (wet weight) as water, but this can be as low as 25% in spores, although other spore layers have the normal high water content. 2. An extremely high level (∼25% of dry weight) of pyridine-2,6-dicarboxylic acid (common name: dipicolinic acid, DPA) in a 1 : 1 chelate with divalent cations, primarily Ca2+ (Gerhardt and Marquis 1989). 3. A pH 1–1.5 units below that of growing cells (Magill and others 1994, 1996). The maximum removal of water from the core takes place late in spore formation, largely concomitant with the accumulation of DPA by the spore, and ∼90 min after the decrease in the developing spore’s pH (Magill and others 1994). The DPA and chelated cations are excreted early in spore germination, and the core water content and pH also rise rapidly during this period to the level found in growing cells (Setlow 2003). Available evidence indicates that the low core water content, low pH, and high DPA and divalent cation contents play only a minor role in spore resistance to chemicals, although more work is needed in this area (Setlow 2006). Other unique components of the spore core are a group of novel small, acid-soluble spore proteins (SASP) of the α/β-type that are named after the major proteins of this type found in B. subtilis spores (Setlow 2006, 2007). The α/β-type SASP are the products of a multigene family of up to 10 members, found in both Bacillus and Clostridium spores, synthesized only in the developing spore late in sporulation, and the sequences of these proteins are highly conserved both within and across species, but show no obvious sequence homology with other proteins. The α/β-type SASP are DNA-binding proteins, and there is sufficient protein of this type to saturate the spore’s DNA. This DNA binding provides tremendous protection to the DNA against heat, desiccation and UV radiation, and to some genotoxic chemicals such as hydrogen peroxide, formaldehyde, and sodium nitrite, although not against DNA alkylating agents (Setlow 2006). Indeed, spores lacking the majority of their α/β-type SASP (α−β− spores) are much more sensitive to hydrogen peroxide, sodium nitrite, and formaldehyde than are wild-type spores (Setlow 2006). The key role of the α/β-type SASP in DNA protection in spores is shown most strongly by the fact that wild-type B. subtilis spores are not killed by hydrogen peroxide through damage to spore DNA, although hydrogen peroxide treatment kills α−β− spores by DNA damage (Setlow and Setlow 1993).

Safety and Shelf Life of Fresh Produce for the Military

269

There is one other group of proteins in the spore core that could also possibly contribute to spore resistance. These are enzymes that can inactivate toxic chemicals, and include catalases and superoxide dismutases that can inactivate hydrogen peroxide and superoxide, respectively. However, while these enzymes are present in spores, including at least one spore-specific catalase, it appears most likely that these enzymes do not play a role in spore resistance (Casillas-Martinez and Setlow 1997; Setlow 2006). The most likely reason for the lack of effect of these enzymes in dormant spore resistance is that all spore core enzymes appear to be inactive, probably due to the low level of core water (Setlow 2006). Indeed, a normally soluble protein is immobilized in the dormant spore core, although it becomes freely mobile early in spore germination when the spore core’s water content rises to that of a growing cell (Cowan and others 2003). Mechanisms of Spore Killing by Chemical Agents Just as there are multiple factors that contribute to spore resistance against different chemical agents, there are also different mechanisms of spore killing by different chemicals as well, including DNA damage, inactivation of spore germination proteins, damage to the spore’s inner membrane, and, perhaps, inactivation of spore core enzymes. DNA Damage One group of chemicals, including formaldehyde, nitrous acid, and DNA alkylating agents, clearly kills spores by DNA damage, because survivors of treatments with these chemicals have a high level of mutations, and spore resistance to these agents is greatly decreased by the loss of DNA repair proteins (Setlow 2006). For the chemicals that kill spores by DNA damage, the spore coats provide minimal protection, while the low permeability of the spore’s inner membrane appears to be important in protection (Cortezzo and Setlow 2005; Setlow 2006). For some DNA-damaging chemicals (hydrogen peroxide, nitrous acid, formaldehyde), the α/β-type SASP are extremely important in spore resistance, but α/β-type SASP are not important in protection against DNA alkylating agents (Setlow 2006). The analysis of a recently determined high-resolution structure of a DNA saturated with α/β-type SASP has now provided the structural rationale for the differences in DNA protection against the effects of different DNA reactive chemicals by α/β-type SASP binding (Lee and others 2008). However, in some cases the specific DNA damage leading to spore death is not known. Spore Germination Disruption There are also at least a few chemicals that cause spore death by destroying one or more essential proteins of the spore germination apparatus (Setlow 2006). The beststudied example of such a chemical is strong alkali. Incubation of B. subtilis spores in 1 M NaOH quickly causes loss in spore viability on nutrient plates (Setlow and others 2002). This loss in spore viability is due to the inability of the treated spores to complete the process of spore germination, since the alkali treatment inactivates the enzymes required to degrade the spore’s PG cortex, an essential event in spore germination (Setlow and others 2002; Setlow 2006). Consequently these spores appear

270

Section III. Postharvest Interventions

to be dead. However, spores that have been killed in this fashion are actually not dead, and they can be recovered if they are germinated artificially through degradation of the spore cortex with exogenous lysozyme (Setlow and others 2002). This finding is important to keep in mind when assessing spore killing by a particular agent. Are the spores truly dead or only conditionally dead, and might they revive in alternative recovery media or with treatment by an appropriate stimuli? This point is noted again below. Inner Membrane Damage In addition to the mechanisms of spore inactivation considered above, a number of chemicals that kill spores appear to damage the spore’s inner membrane in some fashion such that this membrane ruptures, either directly or upon spore germination (Setlow 2006). One agent that has this effect is relatively concentrated mineral acids. For example, incubation of B. subtilis spores in 1 M HCl rapidly causes dramatic rupture of the dormant spore’s inner membrane (Setlow and others 2002). Although such a treatment is extremely effective in spore inactivation, clearly this method is not likely to be used in the decontamination of foodstuffs. There are a number of other agents that appear to kill spores by damaging the spore’s inner membrane and that are, or could be, used in the decontamination of fresh produce and other foodstuffs. Agents that appear to kill spores by this mechanism include chlorine dioxide, hypochlorite, ozone, and some oxidizing agents (Cortezzo and others 2004; Genest and others 2002; Loshon and others 2001; Paul and others 2006, 2007; Young and Setlow 2003, 2004a,b). These agents do not kill spores by DNA damage, and α/β-type SASP are not important in spore resistance to them. Upon extended treatment with these agents, spores exhibit large decreases in germination efficiency. However, when spores are killed only 90–99%, the treated spores do still germinate, albeit often slowly. In addition, artificial germination of partially killed spore populations with lysozyme does not increase spore recovery (in contrast to the situation with NaOH-treated spores described above). Thus, the reason for the killing of spores by this group of agents is not the inactivation of one or more components of the spore’s germination apparatus. There is evidence indicating that these three agents, and perhaps others, kill spores by some type of inner membrane damage (Cortezzo and others 2004; Genest and others 2002; Loshon and others 2001; Paul and others 2006, 2007; Young and Setlow 2003, 2004a,b): 1. Spore populations killed 90–99% by these agents do not release their small molecule pools, in particular DPA. Thus the spore’s inner membrane, the spore’s main permeability barrier to small molecules, must remain intact in the treated dormant spore. 2. Spore populations killed 90–99% by such agents are sensitized to DPA release upon a normally sublethal heat treatment (Cortezzo and others 2004), suggesting that the spore’s main permeability barrier to small molecules, the inner membrane, is damaged by treatment with these agents. 3. Spore populations killed 90–99% by these agents exhibit much lower recovery when plated on media with relatively high salt content in comparison to untreated

Safety and Shelf Life of Fresh Produce for the Military

271

spores (Cortezzo and others 2004). Such sensitivity to high salt content is often ascribed to damage to membrane functions (Hurst 1977, 1983). 4. Although spore populations killed 90–99% by these agents will germinate, these germinated “killed” spores: a. Often exhibit little if any metabolic activities b. Appear to have a severely damaged plasma membrane as determined by staining with a dye that is excluded from untreated germinated spores c. Will lyse d. Even when germinated artificially with exogenous lysozyme in a hypertonic medium, many of these treated spores lyse. While the available data for killing spores with chlorine dioxide, hypochlorite, or ozone are most consistent with some sort of damage to the spore inner membrane, the identity of this damage is not clear, although it is not damage to unsaturated fatty acids (Cortezzo and others 2004). However, it is clear that spores killed by these agents are indeed truly dead. Spore killing by hydrogen peroxide shares some features with spore killing by chlorine dioxide, hypochlorite, and ozone as noted above, and spore killing by hydrogen peroxide may again be in part due to some sort of inner membrane damage. However, in contrast to the latter three agents that do not kill spores that lack the majority of their α/β-type SASP (α−β− spores) by DNA damage, hydrogen peroxide clearly kills α−β− spores by DNA damage (Setlow and Setlow 1993). Hydrogen peroxide can enter the spore core and damage core components; the other three agents do not appear to do this at rates sufficient to cause appreciable spore killing. Once in the spore core, hydrogen peroxide, and at least some other hydroperoxides efficiently inactivate spore core enzymes at rates that are comparable to spore killing (Palop and others 1996, 1998). Although there is no evidence implicating the inactivation of any specific enzyme by hydrogen peroxide as the cause of spore killing, this mechanism is consistent with recent work suggesting that the inactivation of one or more core proteins is the cause of spore killing by wet heat (Coleman and others 2007). There are also data suggesting that hydrogen peroxide–treated spores can initiate germination, but they do not proceed into spore outgrowth, consistent with the postulated mechanism of inactivation of one or more core enzymes (Melly and others 2002). Although the mechanism of spore killing by hydrogen peroxide has not been specifically identified in detail, it is clear that spores killed by this agent are indeed dead, since artificial germination of such killed spores does not increase spore recovery (Melly and others 2002).

Novel Chlorine Dioxide Technologies for Eliminating Microbial Hazards from Fresh Produce and Food-Handling Environments Aqueous chlorine dioxide provides an effective method for washing minimally processed fruits and vegetables and eliminating spoilage microflora, vegetative pathogens, and bacterial spores of soilborne Bacillus and Clostridium species to prevent potential foodborne disease and food spoilage (Rico and others 2007; Gómez-López and others 2007; Park and others 2008; Mahmoud and Linton 2008; Mahmoud and

272

Section III. Postharvest Interventions

others 2007; Setlow and Johnson 2007). Although commercial chlorine dioxide technologies exist, they are not necessarily well suited to meet the specific, and often unique, requirements of deployed military personnel. Research scientists at NSRDEC/ CFD and in academia have collaborated to invent, develop, and validate novel technologies directed primarily toward military applications that generate chlorine dioxide at point-of-use for eliminating microbial hazards from fresh produce and foodhandling environments. The Portable Chemical Sterilizer The Portable Chemical Sterilizer (PCS) is an innovative surface disinfectant technology based on the broad-based biocidal activity of chlorine dioxide and can be used to eliminate bacterial spores, vegetative pathogens, and spoilage microflora in crosscutting applications ranging from the sterilization of medical equipment and surgical instruments to ensuring the safety of fresh produce (Feeherry and others 2007). The PCS is a portable, energy-independent, almost waterless device that uses a novel chemical combination to controllably produce copious amounts of chlorine dioxide and readily inactivate vegetative pathogens (E. coli, L. monocytogenes, Staphylococcus aureus), resistant bacterial spores (Bacillus stearothermophilus), and bacterial spore bioindicators (B. stearothermophilus or B. atrophaeus spores). The essential design of the PCS is based on a Pelican case, a commercially available hard plastic suitcase made of lightweight, chemically resistant material that is rugged, durable, capable of multiple uses, and molded into a compact, stackable design for simplified logistics and convenient transport. This particular design is well suited to accommodate requirements for rapid mobility in the high-intensity, demanding environments of far-forward military deployments (Fig. 14.3). Comparison of PCS with “Bertha” With its ability to sterilize contaminated surgical instruments, the PCS is a modern field autoclave that signifies a major technological improvement over conventional

Figure 14.3. The PCS is a lightweight, easy-to-carry plastic suitcase embellished with design features to control the activity of chlorine dioxide sterilant and offers several technological advantages over conventional steam autoclaves.

Safety and Shelf Life of Fresh Produce for the Military

273

“Bertha” steam autoclaves used by the military. Steam autoclaves are common laboratory equipment and undoubtedly are reliable in effecting sterilization through the use of wet heat, which is the introduction of water/steam at temperatures above 121 °C and slightly elevated pressures around 17 psi inside a closed, heavily reinforced metal chamber. Bertha steam autoclaves used by the military weigh over 450 lbs (4-man lift), occupy a large footprint (60.1 ft3), require 9 kw of electricity and 5 gallons (640 oz) of water per sterilization cycle, and cost around $27K. Bertha autoclaves are behemoths that are difficult to move and transport, encumber logistics, and require high maintenance, which causes further difficulties for the user in far-forward areas). In contradistinction, as a modern field autoclave, the PCS is low-maintenance, lightweight (at 20 lbs achieves a 95% reduction in weight compared to Bertha), energyindependent (a 100% reduction in electricity usage), almost waterless (uses 10 oz of water, a 98% reduction in water consumption), and compact (at 2.1 ft3, a 96% reduction in volume). The compressed size of the PCS does not compromise throughput: either 2 PCSs or 1 Bertha autoclave can sterilize 4 surgical trays in 1 hour. As a revolutionary breakthrough for energy-independent sterilization, the PCS closed a critical capability gap for far-forward Army Surgical Teams and Special Forces medical detachments by providing an improved medical sterilization technology that is truly portable, power-free, uses a proven sterilant, and functions at point-of-use to meet the high-intensity, rapid-mobility demands of far-forward areas. Commercial industry is currently undertaking development efforts under Patent License Agreements with NSRDEC/CFD to manufacture, commercialize, and obtain FDA approval of the PCS for military and commercial use. Ontogeny of the PCS The military uses chemical heaters as a convenient, lightweight technology capable of heating rations (Meals, Ready-to-Eat—MRE) or for self-heating meals. The basic premise requires balancing in equipoise at least two goals. One is to use exothermic chemical reactions to react quickly and generate an intense surge of heat without inducing explosions, and the second is to prolong the reaction such that the higher temperature does not dissipate heat to the surroundings, but sustains thermal output for protracted periods sufficient to heat the intended consumable by conductive heating. The currently used chemical heater (called the Flameless Ration Heater, FRH) consists of the iron-activated magnesium-water chemical reaction embedded in a polymeric support matrix and balances the goals mentioned above. The primary drawback of the Mg(Fe)-water reaction is the cogeneration of large quantities of flammable hydrogen gas that can be hazardous when produced in the confined spaces of tents, storage containers, or underwater shelters. Previous research efforts at NSRDEC/CFD determined that scavengers of certain chemical precursors could reduce or suppress entirely the production of hydrogen without compromising heater performance (Taub and others 2002; Taub and Kustin 1996). To circumvent the potential hazards associated with the FRH, select exothermic inorganic oxidation-reduction chemical reactions have been investigated as potential environmentally friendly (“green”) chemical heaters. These alternative heaters form benign end-products that are safe to human health and the environment. These studies produced one such chemical heater based on the complex oxidation-reduction reaction

274

Section III. Postharvest Interventions

chemistry of formate and persulfate. The chemical reaction system involved the use of a unique chemical effector and was described using a 20-step mechanism capable of predicting the controlled heat released by this system (Curtin and others 2004). Through the similar use of a chemical effector, a second alternative chemical heater based on oxyhalogen oxidation-reduction chemistry was found. The chlorite–sulfite reaction generates abundant heat exothermically, but it could also be adjusted to produce the disinfectant chlorine dioxide at near-neutral pH without the use of caustic acids (Doona and others 2004, 2007). Understanding the complex reaction kinetics and mechanism of this novel chemical combination has furnished insight that was used to control the rate and quantity of chlorine dioxide produced and which provided the conceptual basis for designing the PCS. That is to say, the PCS is designed to accommodate this unique reaction chemistry and unlocks its inherent capabilities for producing chlorine dioxide disinfectant and efficiently inactivating bacterial spores and vegetative pathogens on foodstuffs and contaminated surfaces. Table 14.1 shows a simplified version of the chemical reaction mechanism that produces chlorine dioxide in the PCS. The core chemical species are an oxidant (chlorite, chemical symbol ClO2−), a reductant (sulfite, chemical symbol SO32−), and a chemical effector (called either ascorbic acid negative ion or sodium ascorbate, chemical symbol C6H7O6Na) that induces the reaction between the oxidant and reductant while also undergoing oxidation. In a typical formulation, the symbols shown in Table 14.1 are the following: Ox− = ClO2−, OxA− = ClO−, Eff− = AH−, Eff • = AH• (ascorbyl free radical), Eff = A (dehydroascorbic acid), Salt1 = SO42−, Salt2 = Cl−. A similar scheme would complete the mechanism by reduction of ClO− to Cl−. The PCS is an innovative device that represents a major technological breakthrough (Doona and others 2005) for portable, energy-independent, point-of-use medical sterilization, where no preexisting commercial technologies were available to fill this need. For the purposes of sterilizing surgical instruments with chlorine dioxide generated by the novel chemical combination specified above, the Pelican case was embellished with special features designed to accommodate the chemical reaction, prevent the accumulation of excess heat and pressure inside the sterilization chamber, flush, and safely filter chlorine dioxide during evacuation of the chamber poststerilization using a disposable, homemade scrubber device, thereby ensuring the health and safety of the user and the environment (Fig. 14.4). Microbiological Validation of the PCS The easy-to-operate PCS units were configured to achieve sterility by inactivating resistant bacterial spores (live cultures and bioindicators) in 30 min. Microbiological validation studies verified that the PCS achieved sterility by inactivating bioindicators

Table 14.1. Simplified 3-step reaction mechanism proposed for the production of chlorine dioxide by the novel chemical combination of chlorite (Ox−), sulfite (Red2−), and ascorbate (Eff−) i. ii. iii.

2Ox− Ox− 2Ox−

+ + +

Eff − Red2− OxA−

→ → +

OxA− OxA− Eff •

+ + →

ClO2 Salt1 Eff

+

Eff •

+

2ClO2

+

Salt2

Safety and Shelf Life of Fresh Produce for the Military

275

(b)

(a)

(c)

Figure 14.4. Easy three-step procedure for operating the PCS: a) mix appropriate quantities of reagents, b) generate ClO2 for 30–60 min, c) air flush and filter remaining ClO2. The process kills pathogenic bacteria on FFVs or sterilizes resistant bacterial spores contaminating medical instruments.

containing spores (105 spores/mL) of either B. stearothermophilus or B. atrophaeus, that are intended to indicate sterilization by steam autoclaves (wet heat) or ethylene oxide gas, respectively. The PCS also sterilized live cultures of resistant bacterial spores of B. stearothermophilus configured as suspensions or dried-on, hard, nonporous surfaces made of glass or metal (and recovered using commercial HY-Check swab kits). In all cases, conditions were achieved in the PCS such that exposure to the chlorine dioxide eliminated the resistant bacterial spores. Aqueous suspensions of B. stearothermophilus spores exposed to the chlorine dioxide treatments were not recoverable but retained their phase bright character after treatment, indicating that a different mechanism was responsible for effecting spore inactivation by chlorine dioxide than by wet heat or high pressure (Young and Setlow 2003). The PCS for Fresh Fruits and Vegetables As indicated above, FFVs are popular food items desired by the soldier, and soldiers in global deployments face potential emerging and more virulent agriterrorist threats

276

Section III. Postharvest Interventions

from fresh produce purchased in host nations with lax standards for hygiene, sanitation, and pesticide use that could contaminate produce. The PCS can also be configured to generate less stringent conditions of chlorine dioxide for treating fresh produce to achieve food safety without compromising food quality (Feeherry and others 2007). The ability of the PCS to kill harmful foodborne pathogens (E. coli and L. monocytogenes) on fresh produce was tested using a spot-inoculation method of high levels of pathogens on the exterior surfaces of tomatoes. Specifically, the exterior surfaces of 25 g samples of tomato wedges were inoculated with either 105 CFU/g L. monocytogenes OSY-8578 (obtained from The Ohio State University) or with 106 CFU/g E. coli ATCC 11229 and air-dried in a sterile bio-hood for 15 min. After the inoculum dried, the tomato wedges were placed inside the PCS and tested under various conditions of chlorine dioxide concentration and exposure time (Fig. 14.5). In some instances, B. stearothermophilus and B. atrophaeus spore bioindicators were also placed inside the PCS during the treatment. In all cases, treatment conditions were found that inactivated the target microorganisms to render tomatoes safe from these foodborne pathogens without compromising the red color of the exterior tomato surfaces (Table 14.2). The color of whole tomatoes was only slightly diminished by the chlorine dioxide treatment. Similar treatments with uninoculated apple slices also tended not to discolor the skin color. The pulp

Figure 14.5. A typical PCS setup to inactivate E. coli or L. monocytogenes on exterior tomato surfaces is shown.

Safety and Shelf Life of Fresh Produce for the Military

277

Table 14.2. Inactivation of spores of B. stearothermophilus (BS) and B. atrophaeus (BA) and vegetative pathogens by different ClO2 treatments with the Portable Chemical Sterilizer (PCS) ClO2 Treatment I II III III a b

Time (Min)

BS Indicator

BA Indicator

30 30 30 60

neg. neg. neg. neg.

neg. neg. neg. neg.

Listeriaa (105)

E.colib (106) neg.

neg. neg.

25 g tomato wedges inoculated with Listeria monocytogenes OSY- 8578. 25 g tomato wedges inoculated with Escherichia coli ATCC 11229.

(a)

(b)

Figure 14.6. The D-FENS sprayer (left) generates aqueous ClO2 at point-of-use and easily sprays ClO2 on food contact surfaces and wipes away harmful pathogens. D-FENS has the potential to sanitize U.S. Army field kitchens and protect soldier health from secondary contamination.

tissue of the sliced apple resisted discoloration and remained white for at least a week after the chlorine dioxide treatment, despite exposure to ambient air. Clearly, chlorine dioxide treatments in the PCS are efficacious in eliminating pathogenic cells, bacterial spores, and the polyphenoloxidase enzyme that causes enzymatic browning in cut apple tissue. D-FENS Another invention derived from this unique chemical combination is a novel process for conveniently generating aqueous chlorine dioxide solutions at point-of-use in a collapsible handheld bottle equipped with a spray-trigger device and called D-FENS (Disinfectant-sprayer for Foods and ENvironmentally-friendly Sanitization, see Fig. 14.6). D-FENS was designed to exploit chlorine dioxide’s ability to eliminate vegetative pathogens, viruses, resistant bacterial spores, and biofilms on surfaces and meet

278

Section III. Postharvest Interventions

the important U.S. Army needs of sanitizing and disinfecting food contact surfaces, food-handling equipment, and field feeding equipment in U.S. Army field kitchens and sanitation centers (Doona and others 2008a,b), thereby protecting soldier health by preventing the spread of foodborne illness through secondary contamination. The preparation of the chlorine dioxide solution in D-FENS takes 2–3 minutes and is intended to remain stable at a minimum for an 8-hour shift. The flexible plastic pouch has a gusseted bottom that opens for the bottle to stand upright when full of disinfectant solution, and the material is chemically resistant to afford multiple reuses. Since the sprayer is collapsible, it reduces the logistics burden for the Army by eliminating the need to transport the weight of water in these units, thereby benefiting the environment by decreasing the consumption of fossil fuel required for shipment, reducing CO2 emissions, and decreasing landfill wastes. The microbiological efficacy of chlorine dioxide as produced in the D-FENS system was determined using the following test protocol and appropriate combinations of the chlorine dioxide concentration and exposure times needed to inactivate certain infectious microorganisms. The D-FENS was used to generate chlorine dioxide solutions at concentrations ≤500 ppm (500 mg chlorine dioxide per 1 liter H2O), and the disinfectant solution was sprayed onto porous agar surfaces in Petri dishes inoculated with 105 S. aureus. A consistent, steady force was used to dispense approximately equal volumes of solution per spray-trigger pulse for each experiment, and the plates were rotated 90 ° between pulses to ensure nearly uniform coverage of the complete agar surface. This technique was used with or without mechanical abrasion (equivalent to wiping or scrubbing) using a glass hockey stick to spread the chlorine dioxide solution. These treatments completely inactivated the infectious bacteria S. aureus. In similar tests, 100 ppm chlorine dioxide solutions effected a >7-log reduction of a 3-strain cocktail of S. aureus inoculated onto stainless steel surfaces after 1, 3, and 5 minutes of contact time. These results demonstrate the efficacy of the D-FENS system as an aqueous disinfectant rinse capable of eliminating microbial contaminants on the nonporous surfaces of fresh produce or as a disinfectant spray to reduce microbes from hard contact surfaces (e.g., countertops, cutting boards, utensils, etc.) in the food preparation and handling environment. A number of small businesses involved in food and food surface sanitation are competing to license the patent for D-FENS and commercialize it for military and civilian consumers in the shortest time practicable. Although it is technically feasible that chlorine dioxide could inactivate biological weapons (e.g., Bacillus anthracis or Anthrax,), some chemical agents used as weapons, or mold in building remediation, D-FENS and the PCS have not yet been tested for these specific applications.

Modified Atmosphere Packaging (MAP) FFVs are continually supplied to U.S. Navy ships, submarines, and deployed ground forces. Improper storage and handling of perishable commodities can contribute to premature spoilage and expensive losses. In fiscal year 2005, the U.S. Navy estimated spending over $26M on FFVs with losses attributable to spoilage costing as much as $3M (12%). Periods of resupplying U.S. Navy ships can be as long as 21 days, a period that exceeds the average useful shelf life of many commodities. Often as the

Safety and Shelf Life of Fresh Produce for the Military

279

food supplies dwindle, Culinary Specialists resort to peeling away wilted, brown leaves of lettuce and other salad greens to find an edible portion. Thus, extending the shelf life of FFVs shipboard is a high priority to ensure the availability of high-quality, fresh, wholesome foods for consumers. As demonstrated in the test aboard the USS Ronald Reagan mentioned earlier, MAP systems can retard postharvest quality losses and extend the shelf life of iceberg lettuce, bananas, broccoli, and romaine lettuce. The U.S. Navy estimates that implementing MAP systems will prevent premature spoilage of FFVs shipboard and could achieve cost-savings of up to $10.6M over the next 5 years. MAP Technologies The basic premise of MAP technology (Werner and Hotchkiss 2006) relies on understanding the fundamental nature of respiration. Minimally processed fruits and vegetables are living, respiring tissues that consume oxygen and release carbon dioxide, water vapor, and heat through normal metabolic processes, with individual produce types having different respiratory requirements. Modifying the atmosphere in which the produce is stored can reduce the metabolic rate and prolong shelf life, maintain postharvest quality, and retain nutritive content. The commercial MAP technology used in the study onboard the USS Ronald Reagan was based on perforate membrane substrate coated with a side chain crystallizable (scc) polymer (provided courtesy of Apio, Inc, Guadalupe, CA). The properties of the membrane-scc polymer construct can be adjusted to vary its gas permeability by changing the polymer composition, the thickness of the scc coated on the polymer, and the number or diameter of holes in the film, and thereby accommodate different types of produce. In general, the sscpolymer membrane is designed to provide higher permeability rates than conventional fresh produce packaging (Table 14.3). The physical characteristics of the ssc polymer are temperature-dependent. At higher temperatures, the ssc-coated membrane becomes more amorphous, which imparts higher gas permeability by a phenomenon called moderate temperature compensation character. In general, a temperature increase from 0–10 °C will induce an increase in the permeability of the membrane 1.8-fold, while concomitantly increasing the respiration rate of a typical produce commodity by 2.0-fold. The shelf life of the produce is improved because the membrane permeability changes in response to the fluctuations in storage temperature and accommodates the changes occurring in the respiratory requirements of the produce.

Table 14.3. Permeabilities of ssc-coated membrane vs. polyethylene O2 permeability (cc/100 in2-atm-day) CO2 permeability (cc/100 in2-atm-day) Ethylene permeability (cc/100 in2-atm-day) H2O permeability to (g/m2-day)

ssc-Coated Membrane 280,000

Polyethylene (Low Density, 2 mm) 254

1,120,000

1,102

1,080,000

508

849

16

280

Section III. Postharvest Interventions

In-House Testing The efficacy of individual MAP systems was tested in an actual environment using 8 of the 14 highest-volume FFVs consumed in the U.S. Navy: bananas, broccoli (crowns and florets), cantaloupes, tomatoes, honeydew melons, iceberg lettuce, and romaine lettuce (Table 14.4). Individual ssc polymer membranes have to be developed with appropriate properties to induce ideal storage atmospheres for each respective commodity as stored in standard, commercial-size case quantities according to ideal respiratory requirements available from the University of California–Davis Postharvest Research and Technology Center (UC–Davis 2007). We will examine specific instances of membrane development and shelf life extension for iceberg lettuce and bananas below, but the end result is that these select FFVs all showed a significant (≥25%) extension in shelf life (Table 14.5). A minimum 21day shelf life was achieved for all of the FFVs (with the exception of bananas), which comports more closely with the resupply time frames of most fleet operations. Iceberg Lettuce Iceberg lettuce is a popular salad bar item and consumed in high volumes on ships. It needs the development of an appropriate membrane capable of sustaining quality Table 14.4. Target ideal packaging atmospheres for select commodities* Produce Item Iceberg lettuce Romaine lettuce Bananas Broccoli Broccoli florets Cantaloupe Honeydew melon

Ideal Temperatures (°C) (°F) 0 32 0 32

Natick Test Temp. (°F) 34–37 34–37

Ideal MAP Atmosphere %O2 %CO2 1–3 >2 1–3 >5

Respiration Rate (RR) mgCO2/kg·hr 6–17 at 0°C 18–24 at 0°C

Back Calculated RR <2% <5%

15 0 0 2.2 7

34–37 34–37 34–37 34–37 34–37

2–5 1–2 1–2 3 3

26–140 at 15°C 20–22 at 0°C 60–100 at 0°C 5–6 at0 °C 14 at 10°C

— 32.4 47 — —

59 32 32 36 45

2–5 5–10 5–10 10 10

* Provided with permission courtesy of Apio, Inc. 2005. Table 14.5. Shelf-life extension obtained with MAP technology* Commercial Shelf Life (Days)† Bananas Broccoli crowns Broccoli florets Cantaloupe Tomatoes (pink stage) Honeydew melon Iceberg lettuce Romaine lettuce * †

3–5 10–14 10–14 10–14 7–14 14–21 14–21 14–21

MAP Shelf Life (Days) 13–15 45 28 35 21 35 36 35

Data provided courtesy of Apio, Inc. Shelf-life estimates are available from http://thepacker.com/theguide/theguidehyphen;home.asp.

Safety and Shelf Life of Fresh Produce for the Military

281

Table 14.6. Iceberg lettuce testing configurations* Iceberg Lettuce Weight Recommended atmosphere Respiration rate (RR) (mg CO2/kg·hr) Storage temperature Package hole size

*

14.35 lb 1–3% O2, minimize CO2 to prevent browning 10.8–18.0 5 °C A = 0.11 in2 (1 × 0.375˝ diameter) B = 0.20 in2 (1 × 0.500˝ diameter)

Data provided courtesy of Apio, Inc. 2005.

and freshness for 6 weeks of storage at 2.2 °C. Control iceberg lettuce in a standard 24 head case made of corrugated cardboard (approximately 40–45 lb per box) was stored exposed to air at 2.2 °C. Two different MAP membrane designs were evaluated for iceberg lettuce. Design A used a polyethylene membrane with an uncovered 0.375in diameter hole, and design B used polyethylene with a 0.500-in diameter hole (Table 14.6). The %O2 and %CO2 were measured in the internal atmospheres created by each membrane using a Bridge Analyzer, Model 9001-2 or a Mocon PAC Series 360 gas analyzer (Fig. 14.7). The %O2 in designs A and B were comparable, with design A generating a slightly lower average %O2 over the 6-week storage period. The control samples were exposed to higher levels of %O2 in the ambient atmosphere. The %CO2 in each membrane design was comparable. At the end of the 6-week trial, the heads of lettuce were visually inspected by trained quality assurance personnel for browning and spoilage as indicators of suitability for consumption. Individual heads of lettuce were deemed “spoiled” if ≥15% of the sample surface showed extensive browning and decay. All heads of iceberg lettuce stored in the MAP systems retained freshness and edibility until week 6, when some samples showed indications of decay (Fig. 14.8). In contradistinction, control samples began showing extensive decay by Week 4, and 100% of the control samples were “spoiled” and inedible by Week 5. Results from the iceberg lettuce trial demonstrate that the shelf life is directly related to the ambient gaseous environment in the package. An appropriate MAP design technology lowered the %O2 in the atmosphere and helped retain the fresh and salubrious nature of iceberg lettuce for the extended shelf life of up to 6 weeks. Control lettuce was exposed to considerably higher levels of O2 in the ambient atmosphere and decayed more readily. Bananas A polyethylene membrane was investigated as a candidate for maintaining the quality and freshness of bananas stored in 40-pound containers at room temperature for ≥15 days (Table 14.7). Control samples used no membrane. The %O2 and %CO2 in each configuration were measured as described above. The MAP membrane produced an average of 6–7 %O2 and of 6–10 %CO2. Control banana samples were exposed to a higher %O2 atmosphere exhibited some browning

MAPS Testing - 24 Heads Lettuce, Oxygen % 14.00 12.00

Oxygen, %

10.00 8.00

A 0.375˝ Hole

6.00

B 0.5˝ Hole

4.00 2.00

39

35

33

28

26

24

20

18

10

6

4

Day

0.00

Days

(a)

MAPS Testing - 24 Heads Lettuce, Carbon Dioxide %

Carbon Dioxide, %

6.00 5.00 4.00

A 0.375˝ Hole

3.00 B 0.5˝ Hole

2.00 1.00

(b)

41

39

35

33

28

26

24

20

18

10

6

4

Day

0.00

Days

Figure 14.7. Percent oxygen (a) and percent carbon dioxide (b) for iceberg lettuce stored in membrane designs a and b stored at 2.2 °C.

Table 14.7. Bananas testing configuration* Bananas Weight Recommended atmosphere Respiration rate (mgCO2/kg·hr) Storage temperature *

40 lb 2–5% O2. Exposure to <1% O2 and/or >7 %CO2 may cause undesirable texture and flavor. 26–140 15 °C

Data provided courtesy of Apio, Inc. 2005.

282

Safety and Shelf Life of Fresh Produce for the Military

(a)

283

(b)

Figure 14.8. Representative samples of iceberg lettuce at day 41 of storage at 2.2 °C. Conrol samples (a) show browning and spoilage, whereas samples stored in MAP (b) are fresh and edible.

Figure 14.9. Comparison of bananas stored in MAP (top) versus control (bottom) at days 6 (left), 9 (center), and 13 (right) stored at ambient temperature.

spots by Day 6, significant browning by Day 13, and mold growth by Day 16 (shelf life estimated as 3–5 days). In contradistinction, bananas stored in the polymer membrane remained yellow, firm, and good-tasting at Day 13 (Fig. 14.9). As demonstrated with iceberg lettuce and bananas, MAP systems based on the ssc polymer membrane provides lower O2 atmospheres to extend the shelf life of fresh

284

Section III. Postharvest Interventions

produce toward meeting the goals of prolonged storage required for U.S. Navy ships. Extending the shelf life of bananas to 13–15 days is significantly more compatible with meeting Navy feeding needs than the conventional 3–5-day shelf life. This MAP technology has been used successfully to extend the shelf life of broccoli (crowns and florets), cantaloupes, tomatoes, honeydew melons, and romaine lettuce (Table 14.5). In addition to meeting military storage requirements, this MAP technology can also be used by growers, packers, and shippers who use good agricultural and good manufacturing practices to achieve the maximum extension of shelf life of high-quality fresh produce, or to minimize food safety hazards.

Acknowledgments The work of Professor Peter Setlow on spore resistance and inactivation has been supported by grants from the Army Research Office and the National Institutes of Health (GM19698). Ms. Deborah Sisson and Mr. Subham Chandra acknowledge with gratitude the work of Dr. Ray Clarke, Cali Tanguay, and Apio, Inc. and their support in completing this chapter. (More information is available at http://www.apioinc.com, accessed September 30, 2008). Helpful information is also available on the UC–Davis Postharvest Research and Technology Center ’s 2007 Produce Facts (available at http://postharvest.ucdavis.edu, accessed September 30 2008). Dr. Christopher Doona, Ms. Florence Feeherry, and Professor Kenneth Kustin acknowledge their gratitude to the U.S. Army Environmental Quality 6.1 Basic Research program for funding this work. They would also like to express their appreciation for the contributions of Ms. Mona Bray of NSRDEC, Professor of Chemistry; Maria Curtin of Stonehill College, in Easton, MA; and Edmund Powers (Food Safety and Defense Team), Claire Lee (Product Optimization and Ration Team), and Chad Haering (Equipment and Energy Technology Team) for providing results relating to chlorine dioxide disinfection of nonporous surfaces.

References Apio, Inc. 2005. Lettuce Shelf Life Extension Project, Phase 2-Test 2. Apio, Inc., CA. Contract NO. W911QY-05-P-0249, Technical Report, U.S. Army Natick Soldier Research Development & Engineering Center, Natick, MA 01760. Buchanan, C.E. and S.L. Neyman. 1986. Correlation of penicillin-binding protein composition with different functions of two membranes in Bacillus subtilis forespores. Journal of Bacteriology 165(2):498–503. Casillas-Martinez, L. and P. Setlow. 1997. Alkyl hydroperoxide reductase, catalase, MrgA, and superoxide dismutase are not involved in resistance of Bacillus subtilis spores to heat or oxidizing agents. Journal of Bacteriology 179(23):7420–7425. Coleman, W.H., D. Chen, Y.-q. Li, A.E. Cowan and P. Setlow. 2007. How moist heat kills spores of Bacillus subtilis. Journal of Bacteriology 189(23):8458–8466. Cortezzo, D.E., K. Koziol-Dube, B. Setlow and P. Setlow. 2004. Treatment with oxidizing agents damages the inner membrane of spores of Bacillus subtilis and sensitizes the spores to subsequent stress. Journal of Applied Microbiology 97(4):838–852. Cortezzo, D.E. and P. Setlow. 2005. Analysis of factors influencing the sensitivity of spores of Bacillus subtilis to DNA damaging chemicals. Journal of Applied Microbiology 98(3):606–617.

Safety and Shelf Life of Fresh Produce for the Military

285

Cowan, A.E., D.E. Koppel, B. Setlow and P. Setlow. 2003. A soluble protein is immobile in dormant spores of Bacillus subtilis but is mobile in germinated spores: implications for spore dormancy. Proceedings of the National Academy of Sciences USA 100(7):4209–4214. Cowan, A.E., E.M. Olivastro, D.E. Koppel, C.A. Loshon, B. Setlow and P. Setlow. 2004. Lipids in the inner membrane of dormant spores of Bacillus species are immobile. Proceedings of the National Academy of Sciences USA 101(20):7733–7738. Curtin, M.A., I.A. Taub, K. Kustin, N. Sao, J.R. Duvall, K.I. Davies, C.J. Doona and E.W. Ross. 2004. Ascorbate-induced oxidation of formate by peroxodisulfate: product yield, kinetics, and mechanism. Research on Chemical Intermediates 30(6):647–661. Doona, C.J., F.E. Feeherry, D.G. Baer, M.A. Curtin, S. Kandlikar, K. Kustin, I.A. Taub and A.T. McManus. 2005. Portable Chemical Sterilizer, 2005 (USPTO Serial No. 11/105,211). Doona, C.J., F.E. Feeherry, and K. Kustin. 2008a. “D-FENS” chlorine dioxide sanitizing sprayer system, 2008 IFT Annual Meeting, New Orleans, LA. Doona, C.J., F.E. Feeherry, K. Kustin, and M.A. Curtin. 2008b. Process for Producing Aqueous Chlorine Dioxide for Surface Disinfection and Decontamination, 2008 (USPTO Serial No. 12/008,035). Doona, C.J., F.E. Feeherry, K. Kustin, M.A. Curtin, J. Barcus and L. Bonvini. 2007. Kinetics and mechanism of effector-driven oxyhalogen reactions for the inactivation of foodborne microorganisms by chlorine dioxide. 234th American Chemical Society Annual Meeting, Boston, MA. Doona, C.J., M.A. Curtin, I.A. Taub and K. Kustin. 2004. Chemical Combination for the Generation of Disinfectant and Heat. 2004 (USPTO Serial No. 10/988,442). Driks, A. 2002a. Overview: development in bacteria: spore formation in Bacillus subtilis. Cellular and Molecular Life Science 59(3):389–391. ———. 2002b. Maximum shields: the assembly and function of the bacterial spore coat. Trends in Microbiology 10(6):251–254. Feeherry, F.E., C.J. Doona, K. Kustin, and M.A. Curtin. 2007. Novel chlorine dioxide technology for eliminating pathogens on tomato surfaces. 2007 IFT Annual Meeting, Chicago, IL. Food and Drug Administration, Department of Health and Human Services. 1998. 21 CFR Part 173 Secondary Direct Food Additive for Human Consumption, Section 173.300 Chlorine dioxide (available at http://edocket.access.gpo.gov/cfr_2008/aprqtr/21cfr173.300.htm, accessed January 15, 2009). Genest, P.C., B. Setlow, E. Melly and P. Setlow. 2002. Killing of spores of Bacillus subtilis by peroxynitrite appears to be caused by membrane damage. Microbiology 148(1):307–314. Gerhardt, P. and R.E. Marquis. 1989. Spore thermoresistance mechanisms. In: Regulation of Prokaryotic Development, edited by I. Smith, R.A. Slepecky and P. Setlow, pp. 43–63. Washington, D.C.: ASM Press. Gómez-López, V.M., F. Devlieghere, P. Ragaert, and J. Debevere. 2007. Shelf-life extension of minimally processed carrots by gaseous chlorine dioxide. International Journal of Food Microbiology 116:221–227. Henriques, A.O. and C.P. Moran, Jr. 2007. Structure, assembly and function of the spore surface layers. Annual Reviews of Microbiology 61:555–588. Hurst, A. 1977. Bacterial injury: a review. Canadian Journal of Microbiology 23(8):935–944. ———. 1983. Injury. In: The Bacterial Spore, vol. II, edited by A. Hurst and G. W. Gould, pp. 255–274. London: Academic Press. Kim, Y.-J., S.-H. Lee, Ji. Park, Jo. Park, M. Chung, K Kwon, K Chung, M. Won and K.B. Song. 2008. Inactivation of Escherichia coli O157:H7, Salmonella typhimurium, and Listeria monocytogenes on stored iceberg lettuce by aqueous chlorine dioxide treatment. Journal of Food Science 73(9):M418–M422. Lee, K.S., D. Bumbaca, J. Kosman, P. Setlow and M.J. Jedrzejas. 2008. Structure of a protein-DNA complex essential for DNA protection in spores of Bacillus species. Proceedings of the National Academy of Sciences USA 105(8):2806–2811. Loshon, C.A., E. Melly, B. Setlow and P. Setlow. 2001. Analysis of killing of spores of Bacillus subtilis by a new disinfectant, Sterilox®. Journal of Applied Microbiology 91(6):1051–1058. Magill, N.G., A.E. Cowan, M.A. Antonio-Leyva, M. Brown, D.E. Koppel and P. Setlow. 1996. Analysis of the relationship between the decrease in pH and accumulation of 3-phosphoglyceric acid in developing forespores of Bacillus species. Journal of Bacteriology 178(8):2204–2210.

286

Section III. Postharvest Interventions

Magill, N.G., A.E. Cowan, D.E. Koppel and P. Setlow. 1994. The internal pH of the forespore compartment of Bacillus megaterium decreases by about 1 pH unit during sporulation. Journal of Bacteriology 176(8):2252–2258. Mahmoud, B.S.M., A.R. Bhagat and R.H. Linton. 2007. Inactivation kinetics of inoculated Escherichia coli O157:H7, Listeria monocytogenes and Salmonella enterica on strawberries by chlorine dioxide gas. Food Microbiology 24(7–8):736–744. Mahmoud, B.S.M. and R.H. Linton. 2008. Inactivation kinetics of inoculated Escherichia coli O157:H7 and Salmonella enterica on lettuce by chlorine dioxide gas. Food Microbiology 25(2), 244–252. Melly, E., A.E. Cowan and P. Setlow. 2002. Studies on the mechanism of killing of Bacillus subtilis spores by hydrogen peroxide. Journal of Applied Microbiology 93(2):316–325. Morbidity and Mortality Weekly Report. 2007. Synopsis for September 6, 2007. Multistate Outbreaks of Salmonella Infections Associated with Eating Raw Tomatoes in Restaurants—United States, 2005–2006 (available at http://www.cdc.gov/od/oc/media/mmwrnews/2007/n070906.htm#2.) Palop, A., G.C. Rutherford and R.E. Marquis. 1996. Hydroperoxide inactivation of enzymes within spores of Bacillus megaterium ATCC 19213. FEMS Microbiology Letters 142(2–3): 283–287. ———. 1998. Inactivation of enzymes within spores of Bacillus megaterium ATCC 19213 by hydroperoxides. Canadian Journal of Microbiology 44(5): 465–470. Park, E.-J., P.M. Gray, S.-W. Oh, J. Kronenberg and D.-H. Kang. 2008. Efficacy of FIT produce wash and chlorine dioxide on pathogen control in fresh potatoes. Journal of Food Science 73(6):M278–M282. Paul, M., B. Setlow and P. Setlow. 2007. Killing of spores of Bacillus subtilis by tert-butyl hydroperoxide plus a TAML® activator. Journal of Applied Microbiology 102(4):954–962. Paul, M., S. Atluri, B. Setlow and P. Setlow. 2006. Mechanisms of killing of spores of Bacillus subtilis by dimethyldioxirane. Journal of Applied Microbiology 101(5):1161–1168. Popham, D.L. 2002. Specialized peptidoglycan of the bacterial endospore: the inner wall of the lockbox. Cellular and Molecular Life Science 59(3):426–433. Rico, D., A.B. Martín-Diana, J.M. Barat and C. Barry-Ryan. 2007. Extending and measuring the quality of fresh-cut fruit and vegetables: a review. Food Science and Technology 18, 373–386. Riesenman, P.J. and W.L Nicholson. 2000. Role of the spore coat layers in Bacillus subtilis spore resistance to hydrogen peroxide, artificial UV-C, UV-B, and solar radiation. Applied and Environmental Microbiology 66(2):620–666. Setlow, B., C.A. Loshon, P.C. Genest, A.E. Cowan, C. Setlow and P. Setlow. 2002. Mechanisms of killing of spores of Bacillus subtilis by acid, alkali and ethanol. Journal of Applied Microbiology 92(2):362–375. Setlow, B. and P. Setlow. 1993. Binding of small, acid-soluble spore proteins to DNA plays a significant role in the resistance of Bacillus subtilis spores to hydrogen peroxide. Applied and Environmental Microbiology 59(10):3418–3423. Setlow, P. 2003. Spore germination. Current Opinion in Microbiology 6(6):550–556. ———. 2006. Spores of Bacillus subtilis: their resistance to radiation, heat and chemicals. Journal of Applied Microbiology 101(3):514–525. ———. 2007. I will survive: DNA protection in bacterial spores. Trends in Microbiology 15(4):172–180. Setlow, P. and E.A. Johnson. 2007. Spores and their significance. In: Food Microbiology, Fundamentals and Frontiers, 3rd edition, edited by M.P. Doyle and L.R. Beauchat, pp. 35–68. Washington, D.C.: ASM Press. Taub, I.A. and K. Kustin. 1996. Water-activated chemical heater with suppressed hydrogen (USPTO Patent number 5,517,981). Taub, I.A., W. Roberts, S. LaGambina and K. Kustin. 2002. Mechanism of dihydrogen formation in the magnesium-water reaction. Journal of Physical Chemistry A 106(35):8070–8078. UC–Davis (University of California–Davis Postharvest Research and Technology Center). 2007. Produce Facts (available at http://postharvest.ucdavis.edu, accessed September 30, 2008). Werner, B.G. and J.H. Hotchkiss. 2006. Modified atmosphere packaging. In: Microbiology of Fruits and Vegetables, edited by G.M. Sapers, J.R. Gorny and A.E. Yousef. Boca Raton, FL: CRC Press, pp. 437–459.

Safety and Shelf Life of Fresh Produce for the Military

287

Westphal, A.J., B.P. Price, T.J. Leighton and K.E. Wheeler. 2003. Kinetics of size changes of individual Bacillus thuringiensis spores in response to changes in relative humidity. Proceedings of the National Academy of Sciences USA 100(6):3461–3466. Young, S.B. and P. Setlow. 2003. Mechanisms of killing of Bacillus subtilis spores by hypochlorite and chlorine dioxide. Journal of Applied Microbiology 95(1):54–67. ———. 2004a. Mechanisms of killing of Bacillus subtilis spores by Decon and OxoneTM, two general decontaminants for biological agents. Journal of Applied Microbiology 96(2):289–301. ———. 2004b. Mechanisms of Bacillus subtilis spore resistance to and killing by aqueous ozone. Journal of Applied Microbiology 96(5):1133–1142.

Section IV Produce Safety during Processing and Handling

15 Consumer and Food-Service Handling of Fresh Produce Christine M. Bruhn

Introduction The recognition that eating fruits and vegetables contributes to good health is widely held. In the last decade, 70% or more people say they are increasing their consumption of fruits and vegetables (Research International 1999, 2004). Improvements in handling and distribution systems and the increased availability of locally grown products in supermarkets and farmer ’s markets have increased consumer exposure to, and demand for, quality produce. In fact, high-quality produce, a clean, neat store, and high-quality meat are the three factors most often ranked as very important to consumers in selecting a supermarket (Food Marketing Institute 2007). The importance of high-quality fruits and vegetables is consistent across the range of urban, suburban, rural, and small town environments. Although the consumption of fruits and vegetables enhances health, these products may also increase the risk of foodborne illness if not handled properly from production through consumption.

Selecting Produce The Food Marketing Institute’s annual survey of 1,000 households consistently indicates good taste as the most important factor influencing purchases (Research International 2000). The perception of association with health does have an influence, and 88% of consumers indicate they are somewhat or very concerned about the nutritional content of their diet (Food Marketing Institute 2007). Produce is viewed as a healthy food choice. In each of the last 10 years, 70% or more consumers responding to the Food Marketing Institute’s annual survey indicated they have increased their consumption of produce to achieve a healthier diet (Research International 2004). When asked to disclose how they are changing their diet to improve or maintain health, respondents most frequently mentioned eating more fruits and vegetables (International Food Information Center 2007). Americans responding to Parade’s Annual Study of the Nation’s Shopping Practices say they are eating more complex carbohydrates including vegetables 50%, salads 49%, and fruits 47% (Hales 2004). Consumers view fruits and vegetables as good sources of vitamins, minerals, and fiber, and helpful in controlling calories. Focus groups conducted over the Internet by Vance Research Services found consumers were aware of the link between eating produce and reducing the risks of cancer (Nelson 2004). When asked what health-related reasons led them to eat more produce, people responded that they were cutting back on calories, reducing cholesterol, following a diet, or following suggestions from a health professional. More households with 291

292

Section IV. Produce Safety during Processing and Handling

children said they were increasing their produce consumption compared to households without children. The U.S. Department of Agriculture (USDA) Economic Research Service (2005) uses marketplace disappearance data to indicate consumption of foods over time. This measure is calculated by measuring the total amount of a commodity produced in the United States plus imports and beginning stocks, and then subtracting exports, ending stocks, and nonfood uses. Per capita estimates are then calculated using population estimates for the year. These data are used for economic analysis to indirectly measure trends in food consumption. The results were consistent with an increase in fruit and vegetable consumption, with the greatest increase in fresh compared to frozen or canned categories (USDA Economic Research Service 2005). Apples, bananas, carrots, bagged salads, and broccoli are the top five fruit and vegetable commodities consumers report eating more of in 2004 (Nelson 2004). Attitude studies indicate consumers tend to prefer locally grown produce because it is considered fresher and riper, it is transported with less fuel, and it supports the local economy, although many people are not aware of what produce is grown locally (Bruhn 1992; Lockeretz 1986). At this time, branding does not appear to be a major factor related to consumer perceptions of quality. When branding was addressed in 2000, almost 90% of consumers believe branded and nonbranded items were about the same in nutritional value, and about 80% considered them comparable in storage life and taste. With regard to safety, 75% of consumers consider branded and nonbranded items comparable (The Packer 2000). When selecting raw produce, ripeness and freshness are rated as the most important factors in the initial purchase (The Packer 2001). Having the appropriate color, ripeness, shape, and size are also important factors (Govindasamy and others 1997). Although color varies by produce type and variety, red blush is preferred in some products, like peaches and nectarines (Bruhn 1995). A characteristic odor is desirable as an indicator of ripeness and high quality. Generally, larger-sized products are priced at a premium; however, some consumers may prefer medium or smaller sizes, depending on who will be eating the item and how it will be consumed. Produce with scars, scratches, and other marks are considered lower quality (Kader 2002); however, some consumers will purchase lower-grade produce if the price is sufficiently low and other factors are present to indicate good quality. When asked how produce could be promoted to encourage purchases, consumers suggested focusing on quality, holding products at proper temperatures, removing spoiled produce from the display, providing sample tasting to demonstrate the quality to consumers, and offering new recipe ideas and preparation tips (Bruhn 1995; The Packer 1996a,b).

Consumer Perception of Produce Safety Even with the widely publicized outbreaks of Escherichia coli in leafy greens and Salmonella in tomatoes, the Food Market Institute’s annual survey of consumer attitudes indicates that generally between 80–86% of consumers were completely or mostly confident that food in the supermarket is safe (Research International 2004). Perception varies over time, depending on news coverage of food safety issues. Consumer confidence significantly decreased between 2006 and 2007, dropping from

Consumer and Food-Service Handling of Fresh Produce

293

82% mostly or completely confident in 2006 to 66% in 2007 (Food Marketing Institute 2007). Confidence in the safety of restaurant food was even lower, with only 42% of respondents completely or mostly confident in safety. The primary sources of food safety information (Food Marketing Institute 2007) are television for 62% of consumers, followed by the Internet (47%), newspapers (44%), and magazines (33%). A significant number of people, 27%, get food safety information from friends and family. The radio is an important source for 21%, followed by supermarkets and the doctor, each cited by 17% of consumers. Bacteria or germs are recognized as a serious health risk by 49% of consumers (Food Marketing Institute 2007). This category has consistently been recognized as a threat by more people than any other category. When potential food safety problems were specifically identified in the 2007 survey, 43% considered tampering a serious risk, 38% considered avian influenza and terrorist tampering a serious threat, and 37% considered residues such as pesticides and herbicides a serious hazard. Irradiation, which could be used to control some bacterial contamination, was considered a serious hazard by only 18% of consumers. In 2001, The Packer ’s survey found 30% of consumers were concerned about residues on fresh fruits compared to 22% concerned about residues on fresh vegetables. Disease and contamination were the primary food safety concerns associated with vegetables, specified by 36% of consumers. The concerns about pesticide residue were highest in 1989 at the time of the controversy over use of the growth regulator, Alar, on apples (Opinion Research 1990). Over time, confidence in the safety of produce and the belief in the health-enhancing value of produce increased due to concerted efforts by the produce industry and health professionals to educate the public. Some supermarkets advertise the use of a certification system to verify that produce satisfies legal requirements for pesticide residue or contains no residues, as detected by sensitive testing methods. Many supermarkets also offer organic produce that circumvents the use of chemical pesticides. Organic Food The United States 1990 Farm Bill established the Organic Foods Production Act. The goal of this legislation was to establish a system for marketing foods produced under organic standards. By following specific production methods, organic production aims to promote biodiversity and soil biological activity and enhance ecological harmony. U.S. regulations, established by the Organic National Standards Board (USDA-AMS 1990), specified that organic foods are produced and handled without the use of synthetic chemicals. The Board prohibits the use of growth hormones, antibiotics, and modern genetic engineering techniques (including genetically modified crops), irradiation, or sewage sludge, even if these approaches are shown to be sustainable or environmentally beneficial (USDA-AMS 1990). The organic market grew from $2.3 billion in 1994, to 6.7 billion in 2000, to $10.38 billion in 2003 (Organic Trade Association 2005a,b). Organic foods are found in foodservice operations and are a mainstay in supermarkets and specialized markets. In 2003, 51% of U.S. women indicate they have seen the USDA organic seal where they do most of their shopping (Burfields 2003). As an indicator of the growing availability of organic foods, The Hartman Group’s report in 2008 found that 69% of U.S.

294

Section IV. Produce Safety during Processing and Handling

consumers use organic products at least occasionally (Organic Trade Association 2008). This is an increase from 55% in 2000. The largest segment of the organic market is fresh produce (NBJ 2001), with 92% of organic consumers having purchased fruits and vegetables (Hartman Group 2001). Price is the greatest barrier to purchasing organic produce, with 63% of consumers associating higher prices with organic produce compared to conventional produce (Nelson 2002). According to the Organic Trade Association (2005a), Hispanic consumers are significantly more interested in natural and organic products than the general population, and more likely to want their stores to carry natural and organic products. Consumers with children under 18 years of age are less likely to buy organic produce than those without young children (Produce Merchandising Staff 2002). Consumers in the West are more likely to buy organic produce than other regions, and households with incomes of $75,000 or more have the highest likelihood to purchase organic produce (Produce Merchandising Staff 2002). The Hartman Group (2001) found that consumers who select organic products do so primarily for health and nutrition reasons, followed by taste, assurance of food safety, and environmental concerns. Using organic food is mentioned as a practice used to maintain health by 37% of consumers (Food Marketing Institute 2001). Those selecting organic foods believe conventionally produced food is risky. They believe organic products are grown without the use of pesticides, are chemical free, and are safer for the environment (Hartman Group 2001; HealthFocus 2003; Natural Marketing Institute 2001; Zehnder and others 2003). In 2007, 10% of consumers indicated that they kept organic food separated from nonorganic (Food Marketing Institute 2007). When asked what attributes of organic foods are of greatest importance, 63% of those who select organic foods indicated that they are grown without pesticides (Natural Marketing Institute 2001). This statement may be the most important characteristic of organic foods and has been rated very important to more consumers than “certified organic” (HealthFocus 2003). Integrated Pest Management (IPM) Virtually every land grant university in the United States has a research group developing environmentally responsive pest control strategies known as Integrated Pest Management (IPM). These methods include the use of beneficial insects to attack harmful ones, the use of insect-resistant varieties of plants, and production management techniques. If pests reach an economically significant level, and other measures are not effective, pesticides may be employed. Prior to selecting a pesticide, the impact on the worker, the environment, and the target pest needs to be evaluated. When consumers hear about the IPM approach, their attitudes toward farming practices and food safety are positive (Bruhn and others 1992: Diaz-Knauf and others 1995). Furthermore, Govindasamy and others 1997) found that people indicate strong support for IPM through both a high willingness to purchase and willingness to pay a premium for produce grown according to IPM approaches. Once informed about IPM, consumers surveyed were more willing to pay a premium and more willing to switch supermarkets to obtain IPM rather than organic produce. An economic analysis of purchase intent found that those with higher income, who were younger, who were

Consumer and Food-Service Handling of Fresh Produce

295

frequent purchasers of organic produce, and who live in suburban areas were more likely to purchase IPM-grown produce and to pay a premium (Govindasamy and Italia 1997). If advertised, the IPM approach may more accurately reflect the environmental and food safety attributes consumers currently associate with organic foods. Waxes Waxing is a concern for many consumers. A national survey in 1995 indicated that only 35% of women and 43% of men said they definitely would eat fresh produce knowing that it is coated with an approved food-grade wax. Since many produce items already appearing in supermarkets are waxed, and nonwaxed fruit is afforded little shelf space, the appearance of waxed fruit must appeal to most consumers. Consumer attitudes appear to indicate a preference for natural (unmodified) produce, a concern regarding the long-term health effects of ingesting wax, and the perception that wax affects product taste (The Packer 1995). Microbiological Hazards Consumers consistently volunteer that the greatest threat to food safety is microbiological (Food Marketing Institute 2007; Research 2004; Research International 2004). Several recent highly publicized outbreaks of foodborne illness have identified fresh produce as the source of the pathogen. Outbreaks have been associated with cantaloupes as the source of Salmonella; frozen strawberries contaminated with Hepatitis A virus; and spinach, lettuce, sprouts, fresh apple juice, and fresh basil implicated in outbreaks of E. coli O157:H7. Consumers responded by avoiding the implicated product. Historically, after these incidences, as many as 60% of consumers indicated they were more concerned about bacterial contamination of fresh produce than in the previous year (The Packer 1998). After the spinach outbreak and other illnesses associated with fresh produce in 2006, 84% of consumers said they had stopped purchasing the products, with 74% specifically identifying spinach as an item they did not buy. (Food Marketing Institute 2007). The trend to stop purchasing specific product items continued as food safety concerns remained in the news. In 2008, 34% of consumers indicated they had stopped purchasing some produce item the previous year (Food Marketing Institute 2008). Although consumers believe produce grown in the U.S. is safer than imported produce, the USDA Economic Research Services has noted that outbreaks occurred from domestic and imported produce (Zepp and others 1998). Care must be taken in the production and processing of fresh produce to avoid contamination or to destroy potential pathogenic organisms. Perception of Produce Processed for Convenience Convenience is highly valued among today’s consumers. A review of successful new product introductions confirms that foods with increased convenience are well received (Grocery Manufacturers of America 2004). Consumers report an increased use of pretrimmed, washed, and bagged fresh produce. Overall, 32% indicated they purchased convenience products more frequently than 5 years ago. More households with children purchased convenience products (31%) compared to households without children (25%) (see The Packer 2004).

296

Section IV. Produce Safety during Processing and Handling

Consumer Handling Practices Studies of consumer attitudes and self-reported behavior indicate most people usually follow safe handling practices; however, members of every demographic group report mishandling, which can result in an increased likelihood of foodborne illness (Redmond and Griffith 2003). Generally, a larger percentage of people over 45 compared to those younger than 45 report following safe handling practices (Albrecht 1995; Altekruse and others 1995; Jay and others 1999; Klontz and others 1995; LiCohen and Bruhn 2002; Williamson and others 1992), and men are less likely than women to follow recommended kitchen sanitation procedures (Albrecht 1995; Altekruse and others 1995; Jay and others 1999; Li-Cohen and Bruhn 2002). Individuals with higher socioeconomic status or with at least some college are least likely to follow safe handling guidelines than those with lower status or compared to those with 12 years or less of schooling, respectively (Altekruse and others 1995; Li-Cohen and Bruhn 2002). This difference relates to the greater likelihood of non– college-bound students taking high school courses where skills in food safety and meal preparation are presented. Much of the safe food handling research is based upon self-reported behavior. People often overstate their compliance to safe handling guidelines. For example, Audits International (1999) found that 79% of consumers correctly identified instances in which hand washing was necessary during food preparation; however, 20% were observed to neglect recommended hand-washing practices. Similarly, 97% of consumers believed that eating lettuce that had been moistened by raw poultry drippings was a “risky” food-handling practice; yet, 98% of these consumers were observed during food preparation to cross-contaminate ready-to-eat (RTE) foods with raw meat or raw egg (Anderson and others 2004). Selecting Produce Consumers avoid produce with cuts, bruises, or obvious blemishes. Although some blemishes on produce surfaces are merely cosmetic, consumers also ensure their own safety by avoiding products in which the tissue shows signs of damage, and bacteria has been shown to invade more readily fruit with cuts and bruises. Cross-contamination can occur in the grocery shopping cart. Although some consumers routinely separate meat and poultry from raw produce while they shop, focus group discussions indicate this practice is not ubiquitous among consumers (Li-Cohen and Bruhn 2002). If people are buying lots of items, adequately separating meat and poultry items from produce in the shopping cart may be difficult. Bringing Produce Home Although the Fight BAC food safety guidelines emphasize the need to separate raw meat and poultry from foods to be eaten raw, many consumers do not realize the potential for fluids to commingle and cause cross-contamination in shopping bags. Less than 30% of consumers in a nationwide mailing indicated that they ask for meat, poultry, and fish to be bagged separately from fresh produce (Li-Cohen and Bruhn 2002). More than half of consumers surveyed indicated that they had no special requirements for produce packaging. Based upon personal observations, some super-

Consumer and Food-Service Handling of Fresh Produce

297

market clerks separate fresh produce and meat in shopping bags, but others are not aware of the importance of this practice. Home Storage Most consumers store produce in the refrigerator; however some items are stored at room temperature (Li-Cohen and Bruhn 2002). Room temperature storage is appropriate for optimum quality of tomatoes, bananas, and unripe climacteric fruits, although refrigeration temperatures lengthens the freshness and slows bacterial growth in produce infected by harmful microorganisms. In a nationwide survey, 42% of consumers stored apples and 24% of consumers stored melons at room temperature (Li-Cohen and Bruhn 2002). Most consumers store fresh produce either in the refrigerator produce drawer or on a shelf. Produce can be contaminated by meat/poultry drippings, if these items are stored in close proximity or above the produce. This potential exists for 30% of consumers, who either store produce wherever there is room or who place meat and poultry on a shelf above the produce (Li-Cohen and Bruhn 2002). This practice is not unique to the United States. Sammarco and others (1997) reported that 75% of Italian households stored raw meat and poultry in the upper refrigerator shelves. Contamination can also occur from produce contacting unclean surfaces in the refrigerator. Frequent cleaning of the home refrigerator is not a universal practice. Although 50% of consumers indicate they clean their refrigerators at least once a month, the remainder cleaned 2–3 times a year or less frequently (Li-Cohen and Bruhn 2002). Some volunteer that they clean only when they see visible dirt or stains. Preparation Although people should wash their hands before beginning food preparation, consumers do not always wash their hands, overstate their behavior, and might not wash with soap for the recommended 20 seconds. Most consumer observation studies in the United States noted that hand washing had taken place, even if the people merely rinsed their hands with water. Although 94% of adults say they wash their hands after using the restroom, an observational study of 6,333 adults found only 68% actually washed (FDA—Center for Food Safety and Applied Nutrition 1998). Between 20– 60% of consumers reported not washing their hands before starting meal preparation or after handling raw meat or poultry (Altekruse and others 1996, 1999; Cody and Hogue 2003; Food Safety Inspection Staff and others 2000; Yang and others 1998). With regard to handling produce, almost half of the respondents acknowledged that they do not always wash their hands before handling produce (Li-Cohen and Bruhn 2002). When food handling was observed, only 45% of consumers attempted to wash their hands before food preparation; of those who washed, only 84% used soap (Anderson and others 2004). Kitchen Sanitation Slightly more than half of consumers report washing the sink before handling fresh produce and about half wash the sink after handling (Li-Cohen and Bruhn 2002). Most (69%) indicated using a cleanser or cleaning solution for washing, 40% used

298

Section IV. Produce Safety during Processing and Handling

dishwashing liquid, 27% used bleach, 19% used antibacterial soap, and 11% indicated they washed with water only. Food safety guidelines developed by the Partnership For Food Safety Education (2006), enumerate four practices that consumers should follow to protect themselves from bacteria and reduce the risk for foodborne illness: 1. Clean by washing hands and surfaces often. 2. Separate potentially hazardous foods from ready to eat, thereby preventing cross-contamination. 3. Cook foods to proper temperatures. 4. Chill foods promptly and at the appropriate temperature. Consumers do not always follow these Fight BAC recommendations. Between 13–70% report using the same utensils and/or cutting board without intervening washings between preparing raw meat or poultry and preparing ready to eat raw produce, thereby risking cross-contamination (Altekruse and others 1996, 1999; Bruhn and Schutz 1999; Cody and Hogue 2003; Food Safety Inspection Staff and others 2000; Jay and others 1999; Klontz and others 1995; Li-Cohen and Bruhn 2002; Williamson and others 1992; Yang and others 1998). Consumers may actually contaminate a food preparation and serving surface, instead of cleaning it. In a 1,000-person survey, Cody and Hogue (2003) found only 29% of respondents reported changing the kitchen cleaning sponge or cloth daily or several times per week. Washing Produce Consumer respondents to survey questions about washing produce indicated that 81% wash just before eating or cooking, 21% wash before placing produce in the refrigerator, and 6% acknowledge that they seldom or never wash produce (Li-Cohen and Bruhn 2002). The item washed least frequently is melon, with 36% indicating they never wash this fruit. Many consumers thought it was not necessary to wash melon since the rind is not consumed. Consumers also indicated they thought it unnecessary to wash homegrown or organic produce. In actual practice, consumers may not wash produce, even when it will be eaten raw. Anderson and others (2004) observed that vegetable washing was inadequate. When preparing salad, 6% of 99 subjects made no effort to clean vegetables, 70% rinsed lettuce, 93% rinsed tomato, 47% rinsed carrots, and 55% rinsed cucumber. Rinsing times ranged from 1 second for lettuce, cucumber, and carrot to 55 seconds for tomatoes, with an overall average of less than 12 seconds. Consumers use different techniques to wash produce, such as peeling, rubbing with hands, scrubbing with a brush, and, most commonly, washing under running water (LiCohen and Bruhn 2002). As many as 20%, however, soak produce in a container. This method is not recommended because contamination can be spread to other produce items. Some consumers, 4%, indicated that they wash produce with dish detergent. This method is not recommended either, because detergent residues can remain on the product (FDA Talk Paper 2001; Food Safety Inspection Staff and others 2000). Storage of Leftovers Most consumers recognize that cut fruit should be refrigerated; however, some volunteered that they stored cut melons at room temperature. When responding to the

Consumer and Food-Service Handling of Fresh Produce

299

development of a safe handling brochure, consumers advised that the words, “always” be added to the advice to refrigerate leftovers immediately, because the participants knew people who stored melons and other cut fruit at room temperature (Li-Cohen and Bruhn 2002).

Food-Service Workers Handling Practices That Affect Produce Safety Safe food handling in the food-service sector may be expected to exceed handling in the home setting. McElroy and Cutter (2004) found that 86% of food-service workers indicated they are moderately likely to follow food safety techniques prior to training and 93% are very likely to follow these techniques after adequate training. In reality, risky practices by food safety workers are commonly reported (Green and others 2005). More than half of respondents (60%) indicated that while at work, they did not always wear gloves while touching RTE food, 23–33% did not always wash their hands or change their gloves between handling raw meat and RTE foods, and 5% said they had worked while sick with vomiting or diarrhea. Food-service workers have frequently been implicated in the spread of foodborne disease by the consumption of raw produce the food-service workers could have accidentally contaminated (Todd and others 2007). The annual food safety audit in Los Angeles County by The Steritech Group (2004) found that the results of restaurant inspections were predictive of foodborne illness investigations. Those factors significantly associated with foodborne illness include incorrect storage of food, reuse of food, lack of employee hand washing, lack of thermometers to monitor and control temperature, and the occurrence of any foodprotection violations. In 2003, the U.S. Food and Drug Administration (FDA) collected data from site visits to over 900 establishments representing nine distinct facility types, including restaurants, institutional food-service operations, and retail food stores (FDA 2004). Direct observations of produce-handling practices were supplemented with information gained from discussions with management and food workers and were used to document the establishments’ compliance status based on provisions in the 1997 Food Code (FDA 1997). Failure to control product holding temperatures (49%), poor personal hygiene (22%), use of contaminated equipment/failure to protect food-handling equipment from contamination (20%), and risk of potential chemical contamination (13%) were the risk factors found to be most often out of compliance with the 1997 Food Code. For the improper holding time and temperature risk factor, 70% of the observations were not at or below 5 °C for produce items classified as potentially hazardous foods (PHF). Holding PHFs at or below 5 °C is critical to preventing the potential growth of human pathogens, which may rapidly proliferate on inadequately refrigerated PHFs. Date marking of refrigerated RTE food is also an important component of any food safety system, and it is designed to promote proper food rotation and limit the potential growth of Listeria monocytogenes during cold storage. However, appropriate date marking of RTE, PHF produce items made on-site did not occur in 34% of the observations.

300

Section IV. Produce Safety during Processing and Handling

Hands are a very common vehicle for the transfer of human pathogens to food products, and food-handlers’ hands may become contaminated when they engage in activities such as handling raw meat products, using the restroom, coughing, or handling soiled tableware. The personal hygiene risk factors associated with produce that are most in need of attention at retail and food-service operations include adequate, available, and accessible hand-washing facilities. These personal hygiene risk factors were found in the survey to be not in compliance with the 1997 Food Code (FDA 1997) 33, 26, and 21% of the time, respectively. Proper cleaning and sanitization of food contact surfaces is essential to preventing cross-contamination. Food safety procedures for cleaning and sanitizing food contact surfaces and utensils for handling produce were found in noncompliance with the 1997 Food Code (FDA 1997) in 44.4% of the observations in this study.

Staying Healthy, Eating Healthy Research on health-related behavior indicates that people make rational decisions when they are aware of and have some knowledge about health problems (McIntosh and others 1994). However, the acquisition of knowledge alone does not automatically produce compliant behavior (Ajzen 1991). People commonly approach decision making under the influence of optimistic bias (Frewer and others 1994). In this case, optimistic bias may lead people to consider the risks of foodborne illness unlikely, and therefore relax the vigilance in their practices to increase safety. Convincing consumers to change unsafe handling practices may be difficult. Most people consider their knowledge of safe handling very good, and about one-third of those responding to a mail questionnaire on safe handling of produce indicated that they were not interested in receiving information (Li-Cohen and Bruhn 2002). Therefore, it is important to recognize that fruits and vegetables promote health and are tasty, but fruits and vegetables must be handled appropriately to avoid illnesses. Barriers to safe handling appear to be similar among food-service personnel and consumers. People are often aware of the steps that should be taken, but do not always follow them due to time pressures, or lack of resources, such as a clean cutting board or thermometer (Clayton and others 2002). Some consumers think the steps are unnecessary, especially if a product already appears clean (Audits International 2001; LiCohen and Bruhn 2002). Because some foodborne illness takes time to develop, consumers may not recognize the consequences of inappropriate handling. Food workers also indicate they would be more likely to follow recommended practices if management and co-workers emphasized food safety, provided education and training, and initiated negative consequences for those who do not prepare food according to safe practices (Green and others 2005). Educational materials must be presented through convenient sources. Among consumers willing to receive information, 54% indicated they preferred brochures in supermarkets, and 46% preferred information on individual produce containers (LiCohen and Bruhn 2002). Safe handling guidelines must be practical, easy to incorporate into daily activities, and presented in simple, direct terms. Consumers indicate they prefer pictures with a minimum of words, and they want to know the reason behind any recommendations (Li-Cohen and others 2002). Consumers suggest that

Consumer and Food-Service Handling of Fresh Produce

301

safe handling be available in English and other languages. Consumers would like to see safe food-handling practices discussed in the news and practiced on cooking shows. They also suggested that the media could advertise the availability of guidelines on the Web. Information should also be targeted toward children, both as part of school and in conjunction with any special activity. For example, safe handling should be a required component of any school gardening program. Recommendations on safe handling of produce should remind consumers and food-service workers that eating fruits and vegetables is healthy, but care should be taken to clean products and minimize any risks that may be present. Consumers believe produce is safe and are reluctant to take extraordinary steps to further enhance safety (Li-Cohen and others 2002). Additionally, people say they prefer not to purchase special cleaning solutions, but would consider using materials already in the home, such as vinegar or a scrub brush. People prefer as many pictures as possible in written guides and most acknowledge that they will just skim text. Recommendations then should be concise, practical, and rely on materials readily available at home. Because of the potential risk for cross-contamination in the home, retail, and food-service establishment, rewashing leafy greens in an intact package is not recommended (Palumbo and others 2007). Guidelines for washing in the commercial setting include thorough washing of hands, using clean intact gloves, cleaning and sanitizing the preparation area including the sink and any equipment that will come in contact with leafy greens, washing the product under cold running water, and then draining in a clean colander or dried in a spinner (Palumbo and others 2007). Produce should be stored under refrigeration and placed in an area to prevent crosscontamination (especially from raw meat or poultry). Produce should be discarded if it looks spoiled. Bagged products should be discarded similarly or if they exceed the use-by date. Recommended handling practices for washing leafy greens by the consumer are also detailed by Palumbo and others (2007). The following are consumer guidelines for washing fruits and vegetables. These were refined through consumer focus groups (Li-Cohen and others 2002), peer-reviewed, and are available at http://anrcatalog.ucdavis.edu/InOrder/Shop/Shop.asp. Type Safe Handling in the search box. This publication provides guidelines for protecting you from harmful bacteria.

Safe Handling of Fruits and Vegetables Eating a Variety of Fruits and Vegetables Is Healthy However, care must be taken to be sure fruits and vegetables do not become contaminated with harmful bacteria. In the United States, one out of four people suffers from foodborne illness each year. Some of these illnesses have been traced to eating raw, unwashed fruits or vegetables. Everyone Is at Risk for Foodborne Illness However, people who are younger than 5, older than 50, diabetic, take antibiotics or antacids, and whose immunity is compromised are at higher risk.

302

Section IV. Produce Safety during Processing and Handling

Bacteria Are Everywhere Harmful bacteria may be on fruits and vegetables, hands, kitchen counters, and sinks, even when they look, feel, or smell clean. At the Supermarket • In the grocery cart, separate fruits and vegetables from meat, poultry, and fish to avoid cross-contamination. • When bagging fresh fruits and vegetables to take home from the supermarket, put fresh produce and meat, poultry, and fish in separate bags. Home Storage • All cut or prepared fruits and vegetables should be stored in the refrigerator along with many types of whole fruits and vegetables. • When using the refrigerator, place produce in the produce drawer or on a clean refrigerator shelf. • Store meat, poultry, and fish in the clean meat drawer or on a tray on the bottom shelf below other refrigerated foods. This prevents meat, poultry, or fish juices from dripping on other foods. Prepare the Kitchen • Clean the sink with hot, soapy water or cleanser before and after washing and preparing fresh fruits and vegetables. • Always wash cutting boards and preparation areas before and after food preparation. Wash preparation areas and utensils especially well after preparing meat, poultry, or fish and before preparing foods that will be eaten without cooking. • If possible, use one cutting board and preparation area for fruits and vegetables and a different cutting board and preparation for meat, poultry, and fish. • Always wash knives with hot soapy water after cutting meat, poultry, or fish before cutting fresh fruits and vegetables; or, use different knives for cutting meat products and fresh produce. • Washing with soap or detergent removes soil and food, but it removes only some bacteria. For additional safety, always sanitize cutting boards and food preparation areas after cutting meat, poultry, or fish, or any produce item with visible dirt or any produce item that grows on or in the ground. Sanitize by one of the following methods: • Pour boiling water over the clean wood or plastic cutting boards for 20 seconds. • Rinse clean wood or plastic cutting boards with a solution of 1 teaspoon chorine bleach in 1 quart (4 cups) of water. • Place plastic cutting boards in the dishwasher and run, using normal cleaning cycle. Wash Your Hands Always wash hands with hot, soapy water for at least 20 seconds before and after handling fresh fruits and vegetables.

Consumer and Food-Service Handling of Fresh Produce

303

Wash All Fruits and Vegetables • Always wash fruits and vegetables, including those that are organically grown, come from a farmer ’s market, or were grown in your own garden. • Wash fruits and vegetables just before cooking or eating. • Wash under running water. • When possible, scrub fruits and vegetables with a clean scrub brush or with hands. • For melons, scrub the rind with a brush under running water before cutting or peeling. This removes bacteria before it is spread by the knife when slicing. Sanitize the brush by putting it in the dishwasher, placing it in boiling water for 20 seconds, or rinsing it in a bleach solution of 1 teaspoon chlorine bleach in 1 quart (4 cups) of water: • Dry fruits and vegetables with disposable paper towels. • Do not use antibacterial soaps or dish detergents to wash fruits and vegetables because soap or detergent residues can remain on the produce. The FDA has not evaluated the safety of the residues that remain on the produce. However, the effectiveness of these washes is not currently standardized. • Soaking fruits and vegetables in water is not recommended because of the potential for cross-contamination. • Remove outer green leaves from items like lettuce or cauliflower before washing. Trim the hull or stem from items like tomatoes, strawberries, and peppers after washing. • Ready-to-eat, prewashed, bagged produce can be used without further washing if it has been kept refrigerated and is used by the “use-by” date. If desired, produce can be washed again under running water. • Precut or prewashed produce sold in open bags or containers should always be washed under running water before using. Refrigerate All Leftovers • Peel leftover melons and store fruits in the refrigerator. Store all cut produce in a clean container in the refrigerator. Useful Websites for More Information • USDA/FDA Foodborne Illness Education Information center • http://www.nal.usda.gov/foodborne/index.html • U.S. FDA/Center for Food Safety and Applied Nutrition • http://vm.cfsan.fda.gov/list.html • Gateway to Government Food Safety Information • http://www.foodsafety.gov

References Ajzen I. 1991. The theory of planned behavior. Organizational Behavior and Human Decision Processes 50:179–211. Albrecht JA. 1995. Food safety knowledge and practices of consumers in the U.S.A. Journal of Consumer Studies and Home Economics 19:119–134. Altekruse SF, Street DA, Fein SB, Levy AS. 1995. Consumer knowledge of foodborne microbial hazards and food-handling practices. Journal of Food Protection 59:287–294.

304

Section IV. Produce Safety during Processing and Handling

———. 1996. Consumer knowledge of foodborne microbial hazards and food handling practices. Journal of Food Protection 59:287–294. Altekruse SF, Yang S, Timbo BB, Angulo FJ. 1999. A multi-state survey of consumer food handling and food-consumption practices. American Journal of Preventive Medicine 16:216–221. Anderson JB, Shuster TA, Hansen KE, Levy AS, Vok A. 2004. A camera’s view of consumer food-handling behaviors. American Dietetic Association 104(2):186–191. Audits International. 1999. Subject: Consumer food handling practices. Available at http://www.audits.com, accessed June 30, 2000. Audits International. 2001. Subject: 2000 Home Food Safety Study. Available at http://www.audits.com, accessed December 12, 2001. Bruhn C. 1992. Consumer attitude toward locally grown produce. California Agriculture 46(4):13. Bruhn C, Peterson S, Phillips P, Sakovich N. 1992. Consumer response to information on integrated pest management. Journal of Food Safety 12:315–326. Bruhn C, Schutz H. 1999. Consumer food safety knowledge and practices. Journal of Food Safety 19:73–87. Bruhn CM. 1995. Consumer and retailer satisfaction with the quality and size of California peaches and nectarines. Journal of Food Quality 18(3):241–256. Burfields T. 2003. Subject: Customers accepting organics. Available at www.thepacker.com/icms/dtaa2/ content/2003-16620-290.asp, accessed September 30, 2008. Clayton DA, Griffith CJ, Price P, Peters AC. 2002. Food handlers’ beliefs and self-reported practices. Environmental Health Research 12:25–39. Cody MM, Hogue MA. 2003. Results of the home food safety—It’s in your hands 2002 survey: Comparisons to the 1999 benchmark survey and health people 2010 food safety behavior objective. American Dietetic Association 103(9):1115–1125. Diaz-Knauf K, Lopez M, Ivankovich C, Aguilar F, Bruhn C, Schutz H. 1995. Hispanic consumer response to information on integrated pest management and food safety concerns. Journal of Sustainable Agriculture 5(1/2):137–149. FDA. 1997. Food Code. Available at http://www.cfsan.fda.gov/∼dms/fc-toc.html, accessed September 30, 2008. FDA—Center for Food Safety and Applied Nutrition. 1998. Handwashing-related research findings. FDA. 2004. FDA report on the occurrence of foodborne illness risk factors in selected institutional food service, restaurant, and retail food store facility types. Available at http://www.cfsan.fda.gov/∼dms/ retrs2a.html, accessed September 30, 2008. FDA Talk Paper. 2001. Subject: FDA advises consumers about fresh produce safety. Available at http:// vm.cfsan.fda.gov/∼lrd/tpproduce.html, accessed September 30, 2008. Food Marketing Institute. 2001. Reaching out to the whole health consumer. Food Marketing Institute, Washington D.C. Food Marketing Institute. 2007. U.S. Grocery Shopper Trends, Arlington, VA. Food Marketing Institute. 2008. U.S. Grocery Shopper Trends, Arlington, VA. Food Safety Inspection Staff, Food Safety and Inspection Service, and U.S. Department of Agriculture. 2000. Focus groups shed light on consumers’ food safety knowledge. The Food Safety Educator 5:1–8. Frewer L, Shepherd R, Sparks P. 1994. The interrelationship between perceived knowledge, control and risk associated with a range of food-related hazards targeted at the individual, other people and society. Journal of Food Safety 14(14):19–40. Govindasamy R, Italia J. 1997. Subject: Consumer response to integrated pest management and organic agriculture: an econometric analysis. Available at http://aesop.rutgers.edu/∼agecon/pub_index.htm, accessed February 8, 2006). Govindasamy R, Italia J, Liptak C. 1997. Quality of agricultural produce: consumer preferences and perceptions. Available at http://aesop.rutgers.edu/∼agecon/agmkt.htm, accessed November 10, 2004. Green L, Selman C, Banerjee A, Marcus R, Medus C, Angula F, Radke V, Buchanan S, EHS—New Working Group. 2005. Food service workers’ self-reported food preparation practices: and EHS—New study. International Journal of Hygiene and Environmental Health 208:27–35. Grocery Manufacturers of America. 2004. Convenience benefits continue to make life easier. Available at http://www.gmabrands.org, accessed December 3, 2005.

Consumer and Food-Service Handling of Fresh Produce

305

Hales D. 2004. What America really eats. Sacramento Bee, pp 6–7, November 14, 2004. Hartman Group. 2001. Healthy living: organic and natural products organic lifestyle study. Spring, Bellevue, Wash. HealthFocus. 2003. HealthFocus Trends Survey, Atlanta Georgia. International Food Information Center. 2007. Consumer attitudes toward functional foods/foods for health, Washington, D.C. Jay SL, Comar D, Govenlock LD. 1999. A national Australian food safety telephone survey. Journal of Food Protection 62:921–928. Kader AA. 2002. Quality and safety factors: definition and evaluation for fresh horticultural crops. In: Postharvest Technology of Horticultural Crops, Third Edition (A.A. Kader, Ed), pp 279–285. University of California Agricultural and Natural Resources, Oakland, CA. Klontz KC, Timbo B, Fein SB, Levy AS. 1995. Prevalence of selected food consumption and prevalence behaviors associated with increased risks of food-borne disease. Journal of Food Protection 58:927–930. Li-Cohen AE, Bruhn CM. 2002. Safety of consumer handling of fresh produce from the time of purchase to the plate: a comprehensive consumer survey. Journal of Food Protection 65(8):1287–1296. Li-Cohen AE, Klenk M, Nicholson Y, Harwood J, Bruhn CM. 2002. Refining consumer safe handling educational materials through focus groups. Dairy, Food and Environmental Sanitations 22(7):539–551. Lockeretz W. 1986. Urban consumer ’s attitudes toward locally grown produce. American Journal of Alternative Agriculture 1(2):83–88. McElroy DM, Cutter CN. 2004. Self-reported changes in food safety practices as a result of participation in a statewide food safety certification program. Food Protection Trends 24(3):150–161. McIntosh WA, Christensen RB, Acuff GR. 1994. Perceptions of risks of eating undercooked meat and willingness to change cooking practices. Appetite 22:83–96. Natural Marketing Institute. 2001. Health and wellness trends report. Natural Marketing Institute, Harleysville, PA. NBJ (National Business Journal). 2001. Organic foods report 2001, San Diego CA. Nelson A. 2002. Nonbuyers of organics prove to be a hard sell. Available at http://www.thepacker.com, September 12, 2002. ———. Subject: Nutrition news incites consumer purchase. Available at www.thepacker.com/icms/_dtaa2/ content/wrapper.asp?alink=2004-1411, accessed November 12, 2004. Opinion Research. 1990. Trends 1990, consumer attitudes and the supermarket. In: Trends. Food Marketing Institute, Washington, D.C. Organic Trade Association. 2005a. Past Present and Future of the Organic Industry. Available at http://www. ota.com/pics/documents/Forecasting2005.pdf, accessed October 12, 2008. Organic Trade Association. 2005b. Organic Food Facts. Available at http://ota.com/organic/mt/food.html, accessed March 2, 2008. Organic Trade Association. 2008. Consumer Profile Facts. Available at http://www.ota.com/organic/mt/ consumer.html?printable=1, accessed October 12, 2008. Palumbo MS, Gorny JR, Gombas DE, Beuchat LR, Bruhn CM, Cassens B, Delaquis P, Farber JM, Harris LJ, Ito K, Osterholm MT, Smith M, Swanson KM. 2007. Recommendations for handling fresh-cut leafy green salads by consumers and retail foodservice operators. Food Protection Trends 27(11):892–898. Partnership for Food Safety Education. 2006. Available at http://www.fightbac.org, accessed October 12, 2008. Produce Merchandising Staff. 2002. More consumers are buying organic produce. Redmond EC, Griffith J. 2003. Consumer food handling in the home: A review of food safety studies. Journal of Food Protection 66(1):130–161. Research. 2004. Subject: U.S. Consumer attitudes toward biotechnology. Available at http://www.Ific.org, accessed February 3, 2007. Research International. 1999. Trends in the United States, Consumer Attitudes and the Supermarket. Food Marketing Institute, Washington, DC. Research International. 2000. Trends in the Supermarket. Food Marketing Institute, Washington, DC. Research International. 2004. Trends in Consumer Attitudes and the Supermarket. Food Marketing Institute, Washington, DC.

306

Section IV. Produce Safety during Processing and Handling

Sammarco ML, Ripabelli G, Grasso GM. 1997. Consumer attitude and awareness towards food-related hygienic hazards. Journal of Food Safety 17:215–221. The Packer. 1995. Convenience craze. The Packer Fresh Trends 1995. CI(54), 56–63. The Packer. 1996a. Catching the shopper ’s eye. In: Fresh Trends, Volume 102, pp 50–52. Vance Publishing, Lenexa, KS. The Packer. 1996b. Turning consumers off. In: Fresh Trends, Volume 102, pp 84–86. Vance Publishing, Lenexa, KS. The Packer. 1998. Microbes grab the spotlight. In: Fresh Trends, Volume 104, ages 20–26. Vance Publishing, Lenexa, KS. The Packer. 2000. Brands. In: Fresh Trends, Volume 106, pp 24–32. Vance Publishing, Lenexa, KS. The Packer. 2001. Eye appeal influences shoppers. In: Fresh Trends, Volume 107, pp 38–42. Vance Publishing, Lenexa, KS. The Packer. 2004. By the numbers. In: Fresh Trends, Volume 110, pp 7–14. Vance Publishing, Lenexa, KS. The Steritech Group Inc. 2004. Subject: The annual Food Safety Audit: Trend Report. Available at http://www.steritech.com, accessed March 2, 2008. Todd ECD, Greig JD, Bartleson CA, Michaels BS. 2007. Outbreaks where food workers have been implicated in the spread of foodborne disease. Part 3. Factors contributing to outbreaks and description of outbreak categories. Journal of Food Protection 70(9):2199–2217. USDA-AMS. 1990. National Organic Program, Organic Foods Production Act of 1990. Available at http:// www.ams.usda.gov/AMSv1.0/getfile?dDocName=STELPRDC5060370&acct=nopgeninfo, accessed October 12, 2008. USDA—Economic Research Service. 2005. Food Consumption. Available at http://ers.usda.gov/data/food consumption, accessed September 30, 2008. Williamson D, Gravani R, Lawless H. 1992. Correlating food safety knowledge with home food-preparation practices. Food Technology 46(5):94–100. Yang S, Leff MG, McTauge D, Horvath KA, Jackson-Thomas J, Murayi T, Boeselager GK, Melkin TA, Gildemaster TC, Ridings DL, Altekruse SF, Angulo FJ. 1998. Multistate surveillance for food-handling, preparation, and consumption behaviors associated with foodborne diseases: 1995 and 1996 BRFSS food-safety questions. Morbidity and Mortality Weekly Report. Surveillance Summary 47SS:33–57. Zehnder G, Hope C, Hill H, Hoyle L, Blake J. 2003. As assessment of consumer preferences for IPM and Organically grown produce. Journal of Extension 41(2):1–5. Available at www.joe.org/joe/2003april/ rb3.shtml, accessed January 10, 2004. Zepp G, Kuchler F, Lucier G. 1998. Food safety and fresh fruits and vegetables: is there a difference between imported and domesticated produced products. USDA Economic Research Services, Washington, D.C.

16 Plant Sanitation and Good Manufacturing Practices for Optimum Food Safety in Fresh-Cut Produce Edith H. Garrett

Introduction Since the middle of the 20th century, fresh-cut vegetables have been staples for food preparation in restaurants. In the late 1980s, during a boom of new restaurant expansion in the U.S., several fast-food chains expanded their lines of fresh-cut produce by creating single-serve salads with multiple fresh vegetable ingredients combined in one bag. Not only was this product innovative and challenging to package to sustain a 2-week shelf life, but food-service buyers also demanded implementation of strict sanitation and food safety procedures to protect the public. Thus began the fresh-cut produce industry’s quest to find the “Holy Grail of sanitation.” Across the country, fresh-cut produce processors initiated an aggressive effort to find food safety procedures that could be applied to fresh-cut produce. They formed their own trade group, the International Fresh-cut Produce Association (IFPA), which had a mission to support the industry in the production of safe, wholesome products. Because there was no other one-stop shopping reference on food safety for produce, the association took on the project themselves. In 1990, Professor William Hurst at the University of Georgia, Department of Food Science, led a team of experts in writing the first version of the Food Safety Guidelines for the Fresh-cut Produce Industry for the IFPA. Before the association merged with the United Fresh Produce Association (UFPA) in 2007, the fourth edition of this document had been published and edited by Dr. James R. Gorny, the IFPA’s Technical Director, and other members of the association’s technical committee. To date, the fourth edition is the most comprehensive publication focusing on food safety in fresh-cut produce and it can be purchased by contacting the UFPA (Gorny 2001). This chapter is designed to cover the important topics, which have been identified by the industry and regulators, that must be included in a comprehensive food safety program for a fresh-cut produce facility. Many publications and research papers have been published on unique food safety technologies for the fresh-cut industry, but the basic standards for food safety for fresh-cut are no different from those for other foodprocessing environments. Good manufacturing practices and sanitation procedures are equally applicable in all food-processing plants.

Product Risk Assessment One of the first steps in determining what kind of food safety procedures apply in a fresh-cut facility is to determine the risk associated with the products being prepared, especially the incoming raw products and ingredients. The National Advisory 307

308

Section IV. Produce Safety during Processing and Handling

Committee on Microbiological Criteria for Foods (NACMCF) defined a hazard as “A biological, chemical, or physical agent that is reasonably likely to cause illness or injury in the absence of its control” (1997). Because we are focused on a specific product category—cut fruits or vegetables—documentable evidence of the hazards associated with these products are available from government or other public databases. Microbiological Hazards Outbreak data is available from the U.S. Food and Drug Administration (FDA) and the U.S. Centers for Disease Control and Prevention (CDC, see CDC 2008). This information has been compiled over a period of years to determine which commodities have been implicated with the most human pathogenic outbreaks. One survey of this data showed produce-related outbreaks were on the rise (Sivapalasingam 2004). In addition, many companies may conduct their own microbiological testing of a product and this data can be added to a risk assessment for the products being packaged in their facility. Physical Hazards Presently, there are no compiled databases on physical hazards associated with freshcut produce, but most processors have internal documentation on items found in processing-line grading reports, sample analyses, and customer complaint files. This kind of “industry knowledge” can be used to determine the commodities most often associated with physical contaminants such as rocks, wood pieces, plastic pieces, or other nonedible items. This type of information could be compiled from multiple processors to generate a more comprehensive risk assessment. Chemical Hazards One of the most frequently associated chemical contaminants for produce is pesticides, which can be monitored through government sampling studies. Products are randomly sampled from the retail marketplace and tested for pesticide residues to enforce the tolerances set by the U.S. Environmental Protection Agency (EPA). The FDA conducts annual pesticide residue monitoring and publishes the results on the web for easy access (U.S. FDA 2008). Many fruits and vegetables are tested each year, but some are only randomly included each year. Those commodities that have violative results (levels higher than the restrictions set by the EPA, including detected chemicals that have no allowable tolerance levels set by EPA for that commodity) should be listed as a more risky product. Once the hazards and their frequency have been identified there should be additional consideration given to the production/handling environment for the finished product and the preparation of the food product by the end-user. For instance, if the commodities will be cut and packaged in the raw form and then served as a raw salad, the risk would be higher than if the products were cooked before consumption. In another scenario, the commodities could be cut and packaged as a fresh salsa where the acidic pH could provide protection against a microbiological risk, even though the product would be uncooked. It becomes easy to see that each product must be analyzed carefully to determine the appropriate level of risk to be applied.

Plant Sanitation and Good Manufacturing Practices

309

The risk categories could be divided into thirds, much as a traffic signal is designed. For instance, red would indicate high risk, yellow for medium risk, and green for low risk. The highest-risk commodities would be identified with red labels and would demand the highest food safety attention. The risk assessment system should be applied throughout the entire line of production.

Worker Hygiene When it comes to preventing microbiological contamination of fresh-cut produce, there are several major sources of microbial contaminants that should be considered: animals in the fields, soil, water, equipment, and people. In the United States, good manufacturing practices (GMPs, see U.S. FDA 2007) are part of the regulations that the FDA has established for all food-manufacturing facilities. These mandatory minimum standards serve to identify subjects to cover in employee training programs. Everyone who works with the product needs education about the risks and potential sources of microbiological contaminants, as well as training on the measures that can be used to prevent and control that contamination. Common and standard practices to prevent contamination from employees include, but are not limited to, workers wearing hairnets, beard covers, lab coats, and gloves to prevent incidental contamination in the food-handling environment. Workers also wear goggles for eye protection and protective footwear to ensure their own safety in these environments. Management will further focus on individuals who exhibit outward symptoms of injury or contagious disease and prevent them from working in the production environment. Employees should be trained not to come to work with these kinds of problems, and they will be visually checked for open wounds, coughing, diarrhea, and other symptoms that may indicate they have a contagious disease or the possibility of transmitting infected bodily fluids to the food or other workers before they are allowed to enter the production area. Employee Training in GMPs Food safety training must be conducted for all newly hired employees from the owner to the janitor. Thereafter, food safety training should be conducted on a regular basis throughout the year as a refresher. Regular hand washing is the number one habit that all employees should practice to prevent the spread of germs from their bodies to the product. Proper hand-washing techniques could be demonstrated in person, videos, or pictures. Employees should also be instructed about daily bathing habits and the need for clean clothes in a food-processing environment. It may be surprising to discover that large numbers of people do not know this information. In keeping with the risk analysis mentioned so far, employees can also be sources of physical or chemical contaminants, so they should be trained and made aware of appropriate practices to protect the product. For instance, FDA’s GMPs do not allow jewelry in the processing plant because it can become a physical contaminant if dropped into the product. Employees benefit by not wearing jewelry—they won’t lose valued personal property or risk physical harm from possible entanglement with a piece of equipment.

310

Section IV. Produce Safety during Processing and Handling

Eating and drinking on the job are other areas that need to be addressed through employee training. To prevent contamination from an employee’s mouth and hand, no form of eating or drinking should be allowed in the production environment. All such activity should be confined to certain break times and within a certain physical area designated for such activities. Finally, employees need to understand proper product-handling techniques. Product that falls on the floor should be put in the trash can, not back on the line. Edible product should be contained only in edible product totes that are marked accordingly. It is very important that management analyze the entire system to identify equipment, containers, tools, and other items that come in contact with the product, and then formalize procedures to prevent contamination of the product. Color coding for tools and containers eliminates confusion, but it must be explained to all personnel for effectiveness.

Process Controls Fresh produce typically has some level of natural microflora with no significance to human health. In some circumstances, fresh produce can be contaminated with pathogenic microorganisms, even when grown and harvested in accordance with good agricultural practices (GAPs). When raw produce is delivered to the freshcut processing plant, there are also some steps that can be taken to prevent contamination during processing and handling of the finished product. Most of these steps are covered in a comprehensive food safety program designed specifically for fresh-cut produce. Hazard Analysis and Critical Control Point (HACCP) Plans Hazard Analysis and Critical Control Point (HACCP) plans have been used in the fresh-cut produce industry since the early 1990s to control hazards. According to the NACMCF HACCP Guidelines (NACMCF 1997), HACCP has the capacity to control microbiological, physical, and chemical hazards. In a fresh-cut plant, the contaminants most often cited are physical in nature and range from wood to metal, but microbiological contamination could be present. When it was first published, the Food Safety Guidelines for the Fresh-cut Produce Industry (Gorny 2001) contained a model HACCP plan for a typical fresh-cut operation. There was only one Critical Control Point (CCP) identified, where the hazard is metal contaminants and the control point was a metal detector. The massive amount of machinery, equipment, and tools used in these facilities could contribute to this potential hazard and provides the impetus to require metal detectors as a critical control point. To date, there is no economical approach that can detect foreign objects made of other materials, so these physical contaminants are best removed using a washing step and several points of inspection by workers. The current edition of the Food Safety Guidelines for the Fresh-cut Produce Industry has an updated plan (with no new CCPs) and is a great resource to use in developing an HACCP plan for any fresh produce operation. It is important to note that there are several prerequisites that need to be in place before the HACCP plan can be effective. Those include the GAPs outlined in the FDA’s Guide to Minimize

Plant Sanitation and Good Manufacturing Practices

311

Microbial Food Safety Hazards for Fresh Fruits and Vegetables (U.S. FDA 1998), GMPs as identified earlier, a documented pest control program, and a well-rounded sanitation plan. GAPs Usually GAPs refer to the food safety standards applied at the growing and harvesting level for all produce commodities. In addition to the general guidelines available from FDA, there are specific guidelines for several commodities that have been recognized as having greater risk factors. Those specific guidelines are Microbiological Safety Evaluations and Recommendations on Sprouted Seeds (U.S. FDA 1999), Commodity Specific Food Safety Guidelines for the Melon Supply Chain (U.S. FDA 2005), Commodity Specific Food Safety Guidelines for the Fresh Tomato Supply Chain (U.S. FDA 2006a), and the Commodity Specific Food Safety Guidelines for the Lettuce and Leafy Greens Supply Chain (U.S. FDA 2006b). The California Leafy Greens Council issued its latest compliance document in October 2007 entitled Commodity Specific Food Safety Guidelines for the Production and Harvest of Lettuce and Leafy Greens (California Leafy Green Products Handler Marketing Agreement 2007). It has more details regarding the food safety standards to be used in lettuce and leafy greens production. When conducting a risk assessment, one must consider the growing conditions for each commodity so that specific standards can be developed to address hazards that might be associated with incoming raw products. Using any of these guidelines for a similar commodity may be helpful, if stringent food safety practices are the goal. Raw Product Specifications Establishing raw product specifications can go a long way in addressing food safety. A front-line practice of inspecting each incoming load can address hazards that may be associated with the raw products. Some of the items on an inspection checklist might be to verify that there is a letter of guarantee on file, the truck trailer is clean and free of contamination, and the produce meets visual indicators of cleanliness. One recent trend in purchasing lettuce, spring mix, and other leafy greens is to have the product tested for microbiological pathogens in the field before harvest. Once the tests are confirmed negative, the field can be harvested and the product shipped to the processor. In addition to collecting test results or Certificates of Authenticity (COA) that are required for incoming products, inspection procedures for each load of raw product can record a lot of important information. Incoming temperatures, condition of the truck trailer and containers, and general quality of commodity are some of the characteristics that could be recorded for use in future analysis. COA are also appropriate for chemicals or other materials used for manufacturing and should be verified before the materials are accepted at the dock. Washing and Water Sanitation A washing step is critical in fresh-cut operations, and a whole day’s production could be discarded if the water system becomes contaminated. In a fresh-cut plant, water is used to move, chill, and wash products. All of these uses are critical in an efficient

312

Section IV. Produce Safety during Processing and Handling

system, and water is an integral tool that comes in close contact with the commodities. Sanitation of the water is imperative in preventing the creation of a “bacterial soup” that could spread contamination to the pooled products. A specific program is necessary in each facility to identify the proper treatment of the water, application of sanitation systems, and procedures to constantly monitor the system for performance. In fact, one of the challenges of the produce industry in the U.S. has been monitoring the effective sanitation of product wash water. Packinghouse and fresh-cut dump tanks, flume systems, and line sprays could all be sources of contamination if a sanitizer is not used. Since there has been no “silver bullet” that has proven to be consistently effective, there are several sanitizers that the industry has relied on for microbial control. The use of chlorine (in liquid, solid, and gas forms) is found in many packinghouses and fresh-cut operations because it is effective and economical. There are issues associated with the use of chlorine, including “off-gassing” of chlorine or acids into the processing environment when there is a high concentration of chemical sanitizer in the water. This can cause harm to the workers. In the case of an out-of-control metering pump, there is also the chance of overdosing the product with the chemical, resulting in off-flavors and odors that could jeopardize the product’s reputation in the marketplace. Regular testing of the sanitizer levels in the water is important when running wash water systems for many hours. In-line electronic continuous monitoring systems are becoming the norm, but a testing procedure to verify the system is also necessary. There are several test kits that can be used for chlorine and other sanitizers, so it is important to find the best one for the food-washing system. For instance, there are four wet-chemistry methods for chlorine: OT (Orthotolidine), Iodometric, DPD (N,Ndiethyl-p-phenolylenediamine), and FAS-DPD (ferrous ammonium sulfate DPD). Some of these tests are dependent on colorimetric methods so they would not be appropriate for wash water where vegetable juices can discolor the water (Gardner 2004). It will take some trial testing to determine which kits will work the best. Sometimes it is helpful to have two different measuring kits in case there are different kinds of washing systems. It is necessary to pay close attention to the equipment used to dose the water with the sanitizer. It could be critical in establishing effective monitoring, but it could also result in disadvantages such as frequent maintenance and increased calibration issues. Alternatives to chlorine include chlorine dioxide, ozone, hydrogen peroxide, and peroxyacetic acid. Each has its own dosing equipment, testing procedures, and limitations so it is important to get expert advice in this area. Finished Product Specifications Assuring the quality of the finished product and checking shipping standards are a final step in monitoring the products that have been produced in the plant. Management can work with customers to identify reasonable specifications for finished products and shipping characteristics to maintain proper control of the quality and safety of the products. A sample of product from each day’s production should be kept in a “retain storage area,” stored under proper temperatures for the appropriate shelf life, and then tested for quality. This step and the resulting records will help in an investigation, if there were a problem with products in the field.

Plant Sanitation and Good Manufacturing Practices

313

Finished products are usually stored at refrigerator temperatures between 2.2–5 °C (36–41 °F) to protect quality and to provide some protection against the growth of pathogenic bacteria. Every effort should be made to maintain the temperature throughout the supply chain to control quality and safety. In addition, cold storage rooms, packaging materials, and truck trailers should be inspected for cleanliness and to verify the absence of chemicals, condensate, or other substances that might contaminate the product. Inspections should be conducted for every load of product and noted in shipping records.

Facility and Equipment Sanitation It should be the goals of any sanitation plan in a food-processing environment to prevent the introduction of hazards (chemical, physical, and microbiological) into the environment and to control hazards during production and sanitation shifts. When designing a fresh-cut operation, sanitation should be a critical consideration at every step of the process. A checklist can be used in the planning phase so the entire project addresses sanitation and the prevention of contamination. The “Food Safety Guidelines for the Fresh-cut Produce Industry” (Gorny 2001) has an extensive 15page checklist that would be invaluable to anyone taking on the design of a new facility or the refurbishment of an existing building for the production of fresh-cut produce. Beyond the initial design, a facility needs a firm foundation for a comprehensive sanitation plan to be established that will assure continued control or removal of potential pathogens and other hazards during production. There are several parts to this foundation and they include a Master Sanitation Plan, internal inspections by management, Sanitation Standard Operating Procedures (SSOPs), a designated crew trained for sanitation procedures, a program to control the use and storage of chemicals, and a documented pest control program. If all of these parts are solid, a firm foundation is created to establish an effective sanitation program. Master Sanitation Plan (MSP) After the facility is up and running, a Master Sanitation Plan (MSP) needs to be developed that covers daily, weekly, monthly, and annual sanitation tasks so that all areas of the facility are cleaned on a regular basis. The wet environment of a fresh-cut plant can become host to microbiological contamination, if it is not regularly cleaned thoroughly and properly. The MSP serves as a reminder of what needs to be done and when to schedule that task. All cleaning tasks should be identified on a checklist. On this checklist, the person who does the cleaning should initial, date, and note anything out of place that needs to be addressed with corrective actions to improve sanitation. In addition to the checklists for the people who actually do the cleaning, management should conduct several inspections over time to make sure the facility is properly cleaned. Internal Inspections Inspections are important for a management team to continually find ways to improve sanitation. During a team inspection, an interactive discussion puts everyone on the

314

Section IV. Produce Safety during Processing and Handling

same page regarding a job to be done, the personnel assigned to do it, and what is needed to accomplish the task. The inspection records also serve for future planning and third-party audits to better understand the priorities. In addition, an operation supervisor needs to conduct a preoperational inspection every day before production begins to make sure the cleaning crew members did their job. This inspection can find gross negligence in cleaning or even a maintenance issue where something needs to be fixed before production begins. Sanitation Standard Operating Procedures (SSOPs) SSOPs are important for standardizing the cleaning and sanitation procedures for any food-processing plant. Standardization gives a plant manager the confidence that the people doing the cleaning will not miss any important step if they follow the SSOPs. Written procedures organize and streamline the cleaning process and reduce maintenance problems by listing the correct steps involved in equipment breakdown for cleaning. SSOPs must include the following key elements: • • • • • •

Equipment breakdown and reassembly instructions Cleaning instructions Approved chemicals and their proper use Frequency of cleaning Person responsible for the work Corrective actions for anything out of specification

As part of the prerequisites for an effective HACCP plan, SSOPs are essential to ensuring a strong sanitation program and should be written by someone in a management position so there is effective follow-up. Designated Cleaning Crew Another essential part of an effective sanitation program is a cleaning crew with specialized training. The sanitation crew will be responsible for proper cleaning techniques and chemical handling procedures. Cleaning crew members usually work when everyone else is gone for the night and need close supervision and frequent training to keep up with safety and regulatory rules. The cleaning and sanitizing chemical supplier could be a good resource for training and regulatory monitoring. Equipment manufacturers could also provide training on proper equipment cleaning and maintenance. Chemical Handling Procedures Approval of proper sanitation chemicals, training, labeling, and storage need close attention by management. Chemicals could become a safety hazard if not handled properly. There should be a locked storage area available for cleaning and sanitizing chemicals. Proper safety equipment such as gloves, face masks, and eye guards should be readily available to all sanitation workers. Training is important when handling chemicals to prevent accidents and unnecessary exposure to harmful chemicals. Proper labeling is essential in protecting workers and should be available in multiple lan-

Plant Sanitation and Good Manufacturing Practices

315

guages if some of the workers handling the chemicals speak and read languages other than English. Pest Control A good pest control program rounds out the foundation for an effective sanitation program in any food-processing facility. Preventing bacterial, insect, rodent, and other pest infestation is easier with a comprehensive approach. First, removing food and water sources can make the plant environment unwelcome to unwanted pests. In executing the sanitation program, it is important to make sure that the following are part of the everyday sanitation process: • • • • •

All food product debris is cleaned up and removed from the worksite promptly. Standing water is removed. Drains are properly cleaned and sanitized. Storage areas are inspected and cleaned. Garbage containers and waste storage areas are constantly monitored and cleaned.

Second, for pests and insects, a “catch and remove” approach is also needed because some of these pests may enter the building through holes, trucks, drains, doors and windows, boxes, totes, bins, or pallets. Traps are ideal for this phase of the program, but no pesticides should be used inside the building where food is produced. Mechanical traps or glue boards work for larger rodent pests. Glue boards and insect trap lights work well for the smaller pests. Regularly check for entry points such as cracks in the walls or door seals, open windows, and torn screens and correct them immediately. It takes a comprehensive approach to control unwanted pests in a food-processing environment. Most processors have found it works best to hire an outside pest control company that demonstrates an understanding of the unique needs of the food industry and follows updates to acceptable pest control standards for food-manufacturing facilities.

Monitoring When all of these food safety and sanitation programs are written and implemented, one more step is required to round out the overall program: monitoring and verification. In order to understand each part of this important step, it is easier to look at it in two parts. Monitoring comes before verification and must be implemented when the different parts of the food safety plan are established. Monitoring may be conducted by workers that record data at different checkpoints at different times, or by mechanical equipment that records or reports signals on electronic screens, paper records, or graph charts. Most of this information is analyzed or stored for reference in case of problems. Paper or electronic records are important for management to check to make sure equipment and systems are working properly. It is important to cover as many processing steps as possible during production to provide a record of what occurred and when.

316

Section IV. Produce Safety during Processing and Handling

Traceback and Product Recall When implemented and tested, a traceback and product recall plan can be one of the most important parts of a food safety program if contamination occurs. The produce industry is beginning to implement traceback monitoring steps in their labeling and production procedures at most levels of the industry. There are still many small or isolated companies that have not implemented optimal traceback monitoring procedures. When looking at new suppliers, make sure that traceback monitoring procedures are in place. This could set a company apart in a recall or withdrawal situation regarding how much product is lost and what kind of economic impact it would have on the system. In developing a traceback or product recall plan for a processing facility, consider implementing some key procedures. First, write a plan using models from the industry. Second, create a team that will be the go-to staff during a situation requiring use of the written plan. This team should reflect a cross-section of the staff that normally runs operations in the plant. Third, practice the plan at least twice a year to keep it updated and current. Finally, analyze the labeling system in use to make sure raw products and ingredients can be traced back to their origins and finished products can be located after they leave the shipping dock. A lot of money can be saved, and good will preserved, if every processor practices their plan with suppliers and customers alike. Record-Keeping Logs and Forms In an expanded view of the overall monitoring part of food safety, documentation and record keeping are required for food safety programs. Documentation is the physical step of recording data over time, whether that is by humans or equipment. Many processing steps can be monitored in terms of temperatures, sanitizer levels, incoming product checks, sanitation procedures, and more. Record keeping is the organization of this data into reports and the storage of the documents and data over time. Without documentation and record keeping, there is no proof that a food safety step has occurred and that could mean that the food may have been produced under unsafe conditions. There are many types of documentation that could be found in the food-processing industry, and listed below are only a few of the records, charts, logs, or forms that would be essential in monitoring a food safety program. The following list has been grouped according to work description within a typical processing facility. Food Safety Overview Documents • Quality/Safety Notebook with Tabs, Descriptions, and Forms • Company History • Description of the Products General Food Safety Documents • Risk Assessment • HACCP Plan • Glass and Brittle Plastic Policy • Traceback and Product Recall Plans

Plant Sanitation and Good Manufacturing Practices

• • • •

317

Customer Complaint Procedure and Forms Hold and Release Plan and Forms Supplier Approval Program Regulatory Inspection Procedures

Employee Documents • Employee Organizational Chart • Employee Training Topics, Procedures and Training Log • Employee Disciplinary Procedures and Incident Report Form • List of Employee GMPs • Blood Policy • Visitor Procedures and GMPs List Product Documents • Raw Ingredient Description and Specifications • Ingredient and Packaging Supply Letters of Guarantee • Finished Product Specifications • Label Samples and Description of Traceback Process • Food Packaging Printing and Material Specifications Processing Environment Documents • Product Flow Chart • Processing Standard Operating Procedures (SOPs) • Preoperational Inspection Form • Daily Production Records • Daily Quality Control Log • Pest Control Log • Receiving and Shipping Forms • Truck Inspections and Logs Sanitation and Maintenance Documents • Self-Inspection Forms • Preventive Maintenance Program and Forms • SSOPs • MSDS Chemical Files and Chemical Inventory Log

Verification Some of the steps involved in the verification of a food safety plan include third-party audits, supplier approval programs, and microbiological guidelines. These programs are primarily verification steps, but they do have a role as monitoring processes over time. For instance, the third-party audit scores and corrective actions could be monitored over time to show improvement. Supplier approval ratings could be useful when negotiations begin for a new season. There may be other verification steps for freshcut processors, but the three mentioned here are essential for management to use in rounding out the basic food safety program.

318

Section IV. Produce Safety during Processing and Handling

Third-Party Audits The first verification step includes a scheduled, annual third-party audit in which an auditing company sends a trained, independent auditor to the facility to look at all food safety procedures and verify that they are properly designed and implemented. This audit can take a day or more to complete. The auditor walks through the facility during production hours and monitors the different production steps to better understand the processes. During the walk-through, the auditor observes many processes including sanitation procedures and quality assurance monitoring steps. After observing production, the auditor then looks at many documents and written plans to determine whether adequate food safety procedures are implemented, monitored, and corrected on a daily basis. This is when record-keeping procedures become important as verification steps. The audit score is usually determined from answers generated on a written survey by the auditor and a complicated scoring system that summarizes the answers. The final score is sometimes put into numerical categories with rankings such as “superior” or “excellent” so the processors can determine where they stand with regard to all other processors undergoing the same audit procedure. Most buyers require an annual third-party audit at their suppliers’ locations to take the place of their own audit or to provide a trained external expert to look objectively at their procedures. Outside auditing firms can be better prepared to act in this capacity than a buyer because they can focus on this role with all their resources: • They can conduct uniform training of auditors. • They can use written survey questions that can be modified over time to better reflect updated food safety practices. • Their expertise allows them to cover many operations over a year ’s time, thus providing a reinforcing experience. • The third-party audit scores can provide a relatively unbiased or independent opinion that can be used in maintaining relationships or building new ones between the customers and the processors. Most third-party audits are preannounced, but processing companies are advised to be “inspection ready” at all times so that they can welcome an unannounced audit at any time. In fact, customers may require unannounced audits more in the future. Supplier Approval Programs A supplier approval procedure is another verification step that some customers want in a fresh-cut processing plant. Customers want to know that a processor is taking as much care in selecting ingredient suppliers as the customer is in selecting a processor. Processors need to determine what can be realistically measured in supplier food safety programs. Processors need to have accountability that the suppliers can provide a consistently safe product and work together in assuring the customer of the product’s safety. First, write out the realistic steps that can be taken to identify food safety procedures that are important for a supplier to have in place. For growers, that may be GAPs in place, a third-party audit, and a letter of guarantee that the produce was grown according to regulations covering agriculture. Second, visit the suppliers or call

Plant Sanitation and Good Manufacturing Practices

319

them to verify that the food safety steps are implemented correctly and ask for written proof. Third, have key documents sent to the processing facility regularly to ensure that food safety procedures are being followed properly. Keep records on all suppliers and make sure they are monitored over time. Microbiological Testing The third verification step covered here, microbiological testing, can serve to reflect on all the other steps a processor or supplier may take in assuring food safety. Microbiological testing for pathogenic organisms on the raw products can act to verify that a product entered the plant without gross levels of pathogens. Testing on finished products can verify they were not contaminated with gross levels of pathogens while being handled in the plant. At best, depending on the number of samples taken, testing can verify that a “lot” of predetermined size may be free of gross contamination of a pathogenic organism. Unless most of the product is tested, one probably will not find a “needle in the haystack” point of contamination. Therefore, random testing of raw product and finished product provides a verification step ensuring that there is no gross contamination that would pose health risks to the consumer. Microbiological testing of water, the environment, and equipment can help determine whether pathogens are present in the plant environment. Sanitation procedures are designed to control (even prevent) pathogens in the plant environment, so testing after cleanup and during production will provide ongoing status reports on achieving that goal. Using a sanitizer means that bacteria will be killed in the environment or water. If tests show the presence of high counts of bacteria, the sanitizer may not be working, and corrective action may need to be taken. Direct testing for pathogen indicators can determine whether the pathogens inhabit inaccessible areas of equipment that have not been cleaned well or in drains where moisture is a constant. Testing can be a strong verification step in sanitation, so this should be used regularly and monitored over time for training to achieve more effective cleaning procedures.

Summary No matter what kind of food-processing facility is being analyzed, certain standards apply for optimum prevention of microbiological contamination and consumer safety. HACCP, GMPs, GAPs, and SOPs may sound like an alphabet soup of regulatory acronyms, but they comprise comprehensive rules dedicated to ensuring food safety, and all managers should receive training in these programs. Employee training, traceback labeling, management oversight, record keeping, and verification steps are all critical implementation procedures for effective food safety programs. The U.S. freshcut produce industry has developed sound principles, practices, and procedures for ensuring food safety that, with diligence, vigilance, and efforts toward continuous improvement, will drive the industry forward with stronger, smarter food safety programs and dedication to ensuring the safety, health, and well-being of the consumer. These efforts are similar to The Global Food Safety Initiative (GFSI, see GFSI 2004) involving a retail-led network of food safety experts and their trade associations. Although the GFSI Guidance Document includes everything in the FDA’s GMPs, it also includes additional areas, and together they act in addition to any legal

320

Section IV. Produce Safety during Processing and Handling

requirements for food in countries of production and consumption to enhance food safety and consumer protection worldwide.

References California Leafy Green Products Handler Marketing Agreement. 2007. Commodity Specific Food Safety Guidelines for the Production and Harvest of Lettuce and Leafy Greens. Sacramento, CA. Available at http://www.caleafygreens.ca.gov/members/resources.asp, accessed September 30, 2008. Gardner M. 2004. Chlorine Testing Critical in Food Processing Applications. Industrial Waterworld, July/ August 2004. Available at http://www.pennnet.com/articles/print_toc.cfm?p=64&v=5&i=4, accessed September 30, 2008. GFSI. 2004. GFSI Guidance Document, 4th edition. Available at http://www.f4st-ec.org/site/pdf/Global_ Food_Safety_Initiative/GFSI_Guidance_Document_4th_edition.pdf, accessed September 30, 2008. Gorny J. (Editor). 2001. Food Safety Guidelines for the Fresh-cut Produce Industry, 4th ed. Washington, DC: United Fresh Produce Association. Available at www.unitedfresh.org, accessed September 30, 2008. National Advisory Committee on Microbiological Criteria for Foods (NACMCF). 1997. Hazard Analysis and Critical Control Point Principles and Application Guidelines, Washington, DC: U.S. Food & Drug Administration, U.S. Department of Agriculture and NACMCF. Available at http://www.cfsan.fda. gov/∼comm/nacmcfp.html, accessed September 30, 2008. Sivapalasingam, S. 2004. Fresh Produce: A Growing Cause of Outbreaks of Foodborne Illness in the United States, 1973 through 1997. Journal of Food Protection, Vol. 67, Issue 10, 2342–2353. U.S. Centers for Disease Control and Prevention—Outbreak Response and Surveillance Team. 2008. Outbreak Surveillance Data. Available at http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm, accessed September 30, 2008. U.S. FDA. 1998. Guide to Minimize Food Safety Hazards for Fresh Fruits and Vegetables. Available at http://www.foodsafety.gov/∼dms/prodguid.html, accessed September 30, 2008. U.S. FDA. 1999. Microbiological Safety Evaluations and Recommendations on Sprouted Seeds. Available at http://www.cfsan.fda.gov/∼mow/sprouts2.html, accessed September 30, 2008. U.S. FDA. 2005. Commodity Specific Food Safety Guidelines for the Melon Supply Chain. Available at http://www.cfsan.fda.gov/∼dms/melonsup.html, accessed September 30, 2008. U.S. FDA. 2006a. Commodity Specific Food Safety Guidelines for the Fresh Tomato Supply Chain. Available at http://www.cfsan.fda.gov/∼dms/tomatsup.html, accessed September 30, 2008. U.S. FDA. 2006b. Commodity Specific Food Safety Guidelines for the Lettuce and Leafy Greens Supply Chain. Available at http://www.cfsan.fda.gov/∼dms/lettsup.html, accessed September 30, 2008. U.S. FDA. 2007. Good Manufacturing Practices (GMPs). Code of Federal Regulations (CFR), Title 21, Section 110. Washington, DC: U.S. Government Printing Office. Available at http://www.cfsan.fda. gov/∼dms/reg-2.html, accessed November 2007. U.S. FDA. 2008. Total Diet Study. Available at http://www.cfsan.fda.gov/∼comm/tds-toc.html, accessed November 2007.

17 Third-Party Audit Programs for the Fresh-Produce Industry Kenneth S. Petersen

Introduction Over the past twenty years, the U.S. Centers for Disease Control and Prevention (CDC) have reported increased incidence of foodborne illnesses linked to fresh produce (U.S. CDC 2008); buyers of fresh produce are increasingly utilizing both second- and thirdparty audits to verify that their suppliers are taking effective precautionary steps to reduce the risk of microbial contamination of the produce. This chapter looks at the third-party audit, how it works, and what produce growers should expect when they are asked to supply a third-party audit to the purchasers of their products. History Since the 1960s when the Pillsbury Corporation developed the Hazard Analysis and Critical Control Point (HACCP) system for NASA to ensure food safety for manned space flights, third-party audits have been used by companies and the government to ensure that suppliers are following specific food safety practices. In the decades that followed, third-party audits became common in the processed foods and seafood industries, but the fresh produce industry generally did not adopt this practice. With the increased incidence of foodborne illnesses associated with fruits and vegetables, President Clinton announced a plan entitled Initiative to Ensure the Safety of Imported and Domestic Fruits and Vegetables. As part of this initiative, the U.S. Food and Drug Administration (FDA) partnered with the U.S. Department of Agriculture (USDA) to issue guidance on good agricultural practices (GAPs) and good manufacturing practices (GMPs) for fresh fruits and vegetables. In October 1998, the FDA issued the Guidance for Industry—Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables (U.S. Food and Drug Administration 1998). It is important to note that the recommendations in the Guide are voluntary, and they are not enforceable under FDA regulations. Since the release of the Guide, the FDA in partnership with the produce industry developed the Guide to Minimize Microbial Food Safety Hazards of Fresh-cut Fruits and Vegetables (U.S. Food and Drug Administration 2008). Additionally, the produce industry, in collaboration with the FDA, developed commodity-specific guidance documents for tomatoes and melons (U.S. Food and Drug Administration 2005, 2006). These guidelines, also voluntary, provide recommendations that reflect the current state of thinking of leading food safety experts and serve as the basis for most industryinitiated requirements for growers. 321

322

Section IV. Produce Safety during Processing and Handling

In 1999, Safeway Inc. initiated a program requiring third-party food safety audits of its suppliers of “high-risk” produce. High-risk was initially limited to leaf lettuce, but then its meaning was expanded to include other items, and eventually included all fruits and vegetables that were Safeway Inc. purchases. Shortly thereafter, Albertson’s Inc. requested its suppliers of fresh produce to verify safe production and packing practices. Specifically required were the development of safe production manuals and routine self- and third-party audits based on the sanitation standards provided in the FDA’s Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables. Within several years, other major retail and food-service chains instituted third-party audits as a means of ensuring adherence to good agricultural and manufacturing practices (Laborde 2000). In May 2006, the USDA Agricultural Marketing Service announced that all fruit and vegetable purchases made for feeding assistance programs such as the School Lunch Program would require third-party audits verifying that good agricultural and good handling practices were being followed (U.S. Department of Agriculture 2006).

Current Issues There is no single “national standard” that the produce industry is following, and individual entities requiring third-party verification of GAPs determine their own audit standards and auditors. It is not uncommon for the same operation to be audited twice, thrice, or even more to satisfy the different requirements for various producers/processors and buyers, leading growers to feel in a state of “audit fatigue” from the large number of different audits they go through during the year. Until a national standard that is accepted by the entire produce industry can be developed, or the entities requiring third-party audits accept audits from a wider range of auditing firms, this problem of “audit fatigue” appears likely to continue. Definitions The following terms are important to understand when discussing audits: Audit: A planned, systematic, independent, and documented examination and review to determine whether activities and related results comply with planned arrangements and whether these arrangements are implemented effectively and suitably to achieve objectives (Surak and Wilson 2007). Auditee: The individual or company who is being audited. This may or may not be the same entity that is the client. Auditing Body: A company or government agency that performs audits. These audits may be first-, second-, or third-party. Auditor: A member of the audit team performing the audit. Client: The individual or company that requests an audit. The client sets the parameters, or scope, of the audit and gives the authority to the auditor to perform the audit. External Audit (Second- or Third-Party Audit): An audit performed by an entity not employed by the auditee. A second-party audit is an audit that is conducted by the buyer of the auditee’s product.

Third-Party Audit Programs

323

Internal Audit (First-Party Audit): An audit performed by the auditee on its own system or process. Lead Auditor: The individual designated by the auditing body to lead the audit team. This individual is responsible for all aspects of the planning, preparation, and conducting of the audit. Third-Party Audit: A third-party audit is defined in the American Society for Quality Audit Handbook (Russell 2000) as an audit “that is performed by an audit organization independent of the customer-supplier relationship and is free of any conflict of interest.” The independence of the audit organization is critical in ensuring that the third-party audit is neutral and impartial in its assessment of the safety of the food handling system. Third-Party Audit Standard: Third-party audits are performed by auditing organizations to evaluate the auditee’s system or process and establish compliance with mutually agreed upon criteria. The criteria are usually either “client-developed standards” or an “industry standard” that is frequently utilized in that industry. The leafy greens industry relies on the Commodity Specific Food Safety Guidelines for the Production and Harvest of Lettuce and Leafy Greens (California Department of Food and Agriculture 2007) as the standard for audits carried out by the USDA and the California Department of Food and Agriculture in California and Arizona. Auditing organizations such as GlobalGAP and Safe Quality Food (SQF) have defined standards that are also widely used in the industry. Preparing for a Third-Party Audit It is important for operations to be prepared for third-party audits at any time by knowing the applicable audit standards and working vigilantly to maintain compliance. Because there are many different audit standards used in the produce industry, it would not be feasible to cover all the aspects of the different audit standards here. Instead, the next several sections will cover in a more general way the major components of preparing for third-party audits. The most critical element in being prepared for third-party audits is having a working food safety program. To be successful, this program requires careful planning and thorough implementation so that it is ingrained into the practices of every employee and every component of the operation. Development and Implementation of a Food Safety Plan Food safety plans outline the policies and procedures auditees follow to conform to audit standards and serve as the foundation of the working food safety program. Food safety plans are developed in one of two ways: hiring an outside consultant or having the auditee critically evaluate its own operation. Both approaches are widely used in the produce industry, but it is recommended that an auditee develop and implement its own food safety plan. The first step in developing a food safety plan is identifying specific risks within the operation, similar to the first step operations followed under a HACCP program. Once a hazard analysis has been conducted, policies and procedures can be used to develop Standard Operating Procedures (SOPs) to minimize the risks of microbial contamination occurring. The following excerpt helps define SOPs (U.S. Environmental Protection Agency 2007):

324

Section IV. Produce Safety during Processing and Handling

SOPs detail the regularly recurring work processes that are to be conducted or followed within an organization. They document the way activities are to be performed to facilitate consistent conformance to technical and quality system requirements and to support data quality. SOPs are initially developed using current local, state, or federal regulations and requirements outlined in the audit standard, and the SOPs often codify policies and procedures already practiced, but not formally documented previously. The SOP should be written so that the target audience finds it understandable and usable without requiring constant supervision. Copies of the SOPs should be available (both in hard copy and electronic format) and readily accessible for reference in the work areas of individuals performing the activity. SOPs can also be made available in foreign languages, if appropriate, to help prevent deviations from the SOP. Even the most wellwritten SOPs can fail, if they are not followed properly, and management needs to review and encourage the use of SOPs. The SOPs should outline how and what will be documented in a record (documentation is covered later in this chapter). For many operations, the initial implementation of a food safety plan and its associated SOPs may constitute a cultural change, but adapting to these changes is necessary to successfully meet audit requirements. Food safety plans do not need to be complex, and often simpler plans, if written correctly, work better. It is important to evaluate each policy directive and to make sure that each is obtainable and useful. As auditors observe an operation, they note any policy or procedure in the plan that is not being followed as “out of compliance” in the final audit report. If the maximum number of “out of compliance” issues allowed in an audit standard is exceeded, the auditee fails to meet the requirements of the audit standard. Care must be taken to avoid putting anything in the plan that is unobtainable or frivolous that could jeopardize meeting the requirements of the audit standard. An example of an unobtainable or frivolous policy in a food safety plan would be having an SOP requiring restroom facilities to be cleaned every hour. Suppose for the sake of this example that the usage of that restroom facility is such that cleaning it once daily suffices in maintaining a clean and hygienic facility. If the facilities are cleaned once daily and recorded in written documentation, this situation would not be in compliance with the auditee’s policy, even if the facilities are clean throughout the day. Compliance, in this case, requires that the restroom facilities must be cleaned every hour, even if they have not been used. Such a policy would be considered unobtainable or frivolous. The policy should be revised to something like “restroom facilities will be checked every hour for cleanliness, and cleaned if found to be dirty and, at a minimum, cleaned once per day.” In this case, the restroom facilities need only to be checked every hour, and cleaned only if dirty, at a minimum of once daily. The written documentation of the daily cleanings, as mentioned above, would suffice to ensure compliance with the auditee’s policy. Training It is important to understand the training requirements of the audit standard and the auditee’s food safety plan in order to successfully implement a food safety program. Once a food safety plan has been developed, management should arrange to train all those who are covered by the plan so that they understand and abide by it. Training

Third-Party Audit Programs

325

should be viewed as a continuous process, whether to educate new hires; to give employees a refresher on the plan; or to get feedback on what works, what does not, and what can be improved. An audit standard may require that someone in the operation will receive specific training in food safety or HACCP and that someone must attend meetings throughout the year in order to maintain certification to that standard. Documentation If it’s not documented, it didn’t happen. Properly documenting observations during an audit is as important as the actual observations themselves, because it provides the only source of information an auditor can review with regard to an auditee’s adherence to a food safety plan. These records are vital in demonstrating an auditee’s compliance with established food safety regulations, if an auditee is ever investigated by a local, state, or federal health department or agency. Most audit standards outline the documentation that is required and the documentation that the auditee identified as necessary during the development of its food safety program. Many local cooperative extension services and commercial audit companies have developed templates that can assist the auditee in preparing to properly document activities. In all cases, documentation should be legible, clear, concise, and to the point to provide the auditee and the auditor a clear picture of what happened. A good rule of thumb for documentation is to cover the “Five Ws and H” of investigative journalism; the who, what, where, when, why, and how. All documentation and records should cover these questions to facilitate the auditor ’s evaluation of the process. Figure 17.1 provides a sample training record that answers the Five Ws and H: Who was trained, What they were trained to do, Where and When they were trained, Why they were trained, and How they were trained.

Internal Audits and Third-Party Audits Internal audits and third-party audits both serve as a way to make continual improvements to the food safety plan. Continual improvements assist the auditee in making sure that the food safety program does not become out of date or obsolete. Internal Audit and Verification Once the food safety plan has been implemented, the auditee should perform internal audits of the operation to verify that the food safety plan is being followed properly and working out as it was intended. Internal audits are an excellent way for management to “shake down” the plan and make sure that it is working effectively. If something is found not to be working, it gives management an opportunity to address the issue and fix it before an outside auditor comes in. It may take several internal audits to work out the kinks in a newly implemented food safety program, but once the program has been established, internal audits may need to be performed only routinely on a yearly basis or whenever there is a major change in the operation that affects the identified risks in the food safety program. In some audit standards, internal audits are a required component of a food safety program. Internal audits should be documented, including the results and any corrective actions taken, if necessary.

326

Section IV. Produce Safety during Processing and Handling

Employee Training Form for Worker Health and Hygiene Ed Range Produce Co, Blairstown, NJ In accordance with ABC Produce Company Food Safety Plan, the following employees were trained on proper hand washing and personal hygiene using the Cornell University Training video Fruits, Vegetables, and Food Safety: Health and Hygiene on the Farm.

Trainer:

Location:

Date:

By signing below, I acknowledge that I have received and understand the training received on proper hand washing and personal hygiene. If I have questions about this training in the future, I will contact my immediate supervisor. Employee Name (please print)

Employee Signature:

1. Miguel Sonora 2. Ging Lee 3. Ivan Peppelski 4. Budwada Unifemi 5. John Doe

M Sonora Ging Lee Ivan Peppelski Budwawa Unifemi John Doe

Employee Training Form 01 (10/07)

Figure 17.1.

An example of a training record.

Scheduling a Third-Party Audit Once the auditee feels comfortable that its food safety program is in place and functioning well in terms of meeting the criteria for a third-party audit as outlined in the audit standard, the auditee should contact the appropriate auditing body. Most auditing bodies have an audit program coordinator that handles all audit requests and scheduling of audits and personnel. When speaking to the coordinator for the first time, be prepared to describe the operation, identify the applicable audit standard, and answer any other questions that he/she may have. The coordinator will collect the basic information about the auditee and forward it to the lead auditor, who serves as the contact between the auditing body, auditee, and client. The lead auditor contacts the auditee and client (often the same entity in the produce industry) to gather more specific information about the scope of the audit and schedule a date to perform the audit. The lead auditor develops an audit plan that includes such things as an estimate of the time needed to conduct the audit; the number of auditors required; and any other details needed to plan, prepare, and conduct the audit. Prior to the audit, the lead auditor may ask for a copy of the food safety plan and maps of the farm or operation for review, an escort or guide for the day of the audit, and access to a room for the audit team to

Third-Party Audit Programs

327

compile the audit report at the conclusion of the audit. If the auditee’s employees predominately speak a foreign language, the lead auditor may also ask for a translator to be provided. Audit Day All audits start with an opening meeting and finish with a closing meeting. The auditee (and/or client) management should be present at both meetings. At the opening meeting, the lead auditor introduces the audit team, outlines the audit plan, discusses timelines, requests documentation and records, and indicates that the audit team will be interviewing employees, as necessary. Auditors interview employees to verify that policies and procedures outlined in the food safety plan are being followed. It is generally beneficial for the auditee to notify employees that an audit is scheduled and that they may be interviewed by an auditor prior to the audit. After the opening meeting, auditors have the discretion to go anywhere in the operation at any time, but usually they request that the escort or guide bring them to the beginning of the process to follow the product through to the end of the process. For produce operations, the auditors usually start in the fields during harvest, and they finish in a packinghouse or storage facility, depending on the scope of the audit. The escort or guide should provide access to the facility, not lead the audit team through “a dog and pony show” that may raise suspicions and cause the team to want to dig deeper and find out what the auditee might be hiding. Once the fact-finding portion of the audit is completed, the audit team will meet internally to discuss individual observations, review interview notes, and prepare the final audit findings. Once that process is completed, a closing meeting is held. At the closing meeting, the lead auditor will discuss the findings of the audit and give the auditee management the opportunity to discuss, clarify, or dispute any findings, as appropriate. A copy of the audit report is given to the auditee, and, depending on the applicable audit standard, corrective action reports that require follow-up actions are issued. Corrective Action Reports The auditor issues corrective action reports for any observations or records that indicate noncompliance. Figure 17.2 shows a representative simple corrective action report using the restroom cleaning example used earlier in this chapter. Depending on the severity of the nonconformity, the corrective action report may require that the auditee address the issue in a certain period of time or require that the auditor perform a follow-up audit to verify that the auditee took appropriate corrective action to fix the problem. Corrective actions can be identified as either short-term or root-cause corrective actions. Short-term corrective actions are immediate steps to correct a specific issue. Root-cause corrective actions are long-term actions that look to solve habitual problems within a system. For example, an employee is found not to be wearing a hairnet, and there is a policy for all employees to wear hairnets. A shortterm corrective action would be removing that employee from the area to put on a hairnet. If there are multiple employees not wearing hairnets and the issue recurs and is documented over a significant period, a root-cause analysis would be done. This analysis might show that the employees were not trained properly with hairnets, did

328

Section IV. Produce Safety during Processing and Handling

NOTICE OF CORRECTIVE ACTION ABC AUDITNG COMPANY, Washington DC Name of Auditee: Ed Range Produce Co. Auditor Name: Ken Petersen

Date: January 15, 2008

OBSERVATION: During a records review of Ed Range Produce Co.’s restroom cleaning schedule, records indicate that the restrooms were only cleaned 1 time a day for the period December 15, 2007, to present. Records indicated that the maintenance supervisor performed this function and initialed the “Restroom Cleaning Log” when the task was completed. When the maintenance supervisor was interviewed, he indicated that he did only clean the bathrooms once daily during the time frame above. When auditor showed maintenance supervisor the restroom cleaning policy (page 7 of the Ed Range Produce Co SOP Manual) which states the restrooms are to be cleaned every hour, he stated that the policy had been changed and the SOP guide had not been updated. Auditor Signature: Kenneth S. Petersen

Date: January 15, 2008

CORRECTIVE ACTIONS TAKEN: (To be completed by Auditee) Based on the auditor’s observations listed above, the following corrective actions were taken: 1. The restroom cleaning policy had been changed December 15, 2007. Verbal instructions were given to maintenance supervisor on December 15th, however the SOP Guide had not been properly updated. The new restroom cleaning policy has been updated in the SOP guide and copies distributed to all maintenance personnel who perform that job function. 2. Management has been reminded that any verbal changes to company policy must be documented in the company’s food safety plan. All supervisors were also notified that if verbal instructions that supersede written company policy are not followed up with new written instructions within 7 days, the incidents are to be reported to the Vice President of Operations.

Signature: Ed Range

Date: January 17, 2007

Figure 17.2. An example of Notice of Correction Action.

not know they were required to wear hairnets, or that the auditee is not supplying enough hairnets for the number of employees in the field. The root-cause corrective action would have to examine the implementation of this policy and make sure that training is provided to employees; that employees are made aware of, understand, and abide by the policy; and that the supply of hairnets is sufficient to ensure compliance with the audit standard.

Conclusions Third-party audits are an important component of modern food safety policy. These voluntary, nonregulatory audits serve as tools for major retailers, food service providers, and government agencies to verify that the fruits and vegetables they purchase are grown under established good agricultural practices. They are not in themselves guarantors that produce is always safe, but they do give some assurances that produc-

Third-Party Audit Programs

329

ers and processors are doing everything they can to minimize the risks of microbial contamination. If the federal government institutes mandatory GAPs in the future, third-party audits will provide some assurance to consumers and buyers that those regulations are being followed.

References California Department of Food and Agriculture. 2007. California Leafy Green Products Handler Marketing Agreement, 2007. Commodity Specific Food Safety Guidelines for the Production and Harvest of Lettuce and Leafy Greens. Sacramento, CA. Available at http://www.caleafygreens.ca.gov/members/ resources.asp, accessed September 30, 2008. LaBorde L, 2000. Third Part Audits—New Demands for Assurances of Food Safety. Mushroom News 48(9):4–6, September, 2000. Available at http://foodsafety.cas.psu.edu/mush/sept_2000.pdf, accessed September 30, 2008. Russell, JP (Ed), 2000. The American Society for Quality Audit Handbook, 292 pp. ASQ Quality Press, Milwaukee, WI. Surak JG and Wilson S (eds.), 2007. The Certified HACCP Auditor Handbook, ASP Quality Press Milwaukee, Wisconsin, page 283. U.S. CDC (Centers for Disease Control and Prevention—Outbreak Response and Surveillance Team), 2008. Outbreak Surveillance Data. Available at http://www.cdc.gov/foodborneoutbreaks/outbreak_data.htm, accessed September 30, 2008. U.S. Department of Agriculture, 2006. Fresh Fruits and Vegetables: Required Good Agricultural Practices (GAP) Verification. Available at http://www.ams.usda.gov/AMSv1.0/getfile?dDocName=STELP RDC5054457, accessed September 30, 2008. U.S. Environmental Protection Agency, 2007. Guidance for Preparing Standard Operating Procedures (SOPs). EPA QA/G-6 April 2007. Available at http://www.epa.gov/QUALITY/qs-docs/g6-final.pdf, accessed September 30, 2008. U.S. Food and Drug Administration—Center for Food Safety and Applied Nutrition, 1998. Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables. Available at http://www.cfsan. fda.gov/∼dms/prodguid.html, accessed September 30, 2008. U.S. Food and Drug Administration—Center for Food Safety and Applied Nutrition, 2005. Commodity Specific Food Safety Guidelines for the Melon Supply Chain. Available at http://www.cfsan.fda. gov/∼dms/melonsup.html, accessed September 30, 2008. U.S. Food and Drug Administration—Center for Food Safety and Applied Nutrition, 2006. Commodity Specific Food Safety Guidelines for the Fresh Tomato Supply Chain. Available at http://www.cfsan.fda. gov/∼dms/tomatosup.html, accessed September 30, 2008. U.S. Food and Drug Administration—Center for Food Safety and Applied Nutrition, 2008. Guide to Minimize Microbial Food Safety Hazards of Fresh-cut Fruits and Vegetables. Available at http://www. cfsan.fda.gov/∼dms/prodgui4.html, accessed September 30, 2008.

18 Applications of Immunomagnetic Beads and Biosensors for Pathogen Detection in Produce Shu-I Tu, Joseph Uknalis, Andrew Gehring, and Peter Irwin

Introduction Recent outbreaks of pathogenic bacteria on produce have been widespread with severe consequences. Consuming hepatitis A–contaminated green onions sickened 555 persons and killed 3 from a restaurant in western Pennsylvania (CDC 2003). In October 2006, spinach contaminated with E. coli O157:H7 infected 199 people in 28 states, resulting in 141 hospitalizations, 31 cases of kidney failure, and 3 deaths (Surak 2007). Shortly thereafter, lettuce contaminated with E. coli O157:H7 associated with Taco Bell had 71 reported cases, with 53 reported hospitalizations and 8 cases of kidney failure (CDC 2006). Particularly susceptible to foodborne illness are the very young, the elderly, individuals with existing diseases, or immunocompromised individuals. In addition to the human toll, these outbreaks of foodborne illness have caused severe economic losses to food companies, consumers, and employers in general. Escherichia coli 0157:H7 has been implicated with increasing frequency from outbreaks associated with fresh produce, including bean sprouts, cantaloupes, apples, lettuce, spinach, tomatoes, etc. (Ackers and others 1998; Hillborn and others 1999). The mechanisms by which the pathogen is introduced are not fully understood; however, one hypothesis states that the plants may have been contaminated in fields by exposure to contaminated animal feces and/or improperly treated manure (Beuchat 1999). Current epidemiological data indicate that E. coli 0157:H7 may be present in up to 8.3% of dairy and beef cattle (Faith and others 1996) and that it is shed asymptomatically in the feces. Although current manure-handling guidelines recommend a composting period to reduce microbial pathogens in manure before its application as a field fertilizer (FDA 1998), research has demonstrated that manure can support the long-term survival of E. coli 0157:H7 in a variety of conditions (Kudva and others 1998; Wang and others 1996). A second vehicle by which E. coli 0157:H7 may be introduced is flood irrigation with water contaminated with cattle feces or contact with contaminated surface runoff (Ackers and others 1998; Hillborn and others 1999). A number of recent E. coli 0157:H7 outbreaks have been linked to contaminated water (CDC 1999); furthermore, studies have demonstrated the ability of the pathogen to survive for extended periods in water (Chalmers and others 2000; Wang and Doyle 1998). The dose for E. coli 0157:H7 to cause human illness is very low and may be as few as 10 organisms (FDA 1998). Some victims, particularly the very young, have developed the hemolytic uremic syndrome (HUS), which is characterized by renal 331

332

Section IV. Produce Safety during Processing and Handling

failure and hemolytic anemia. From 0 to 15% of hemorrhagic colitis victims may develop HUS, which can lead to the permanent loss of kidney function. In the elderly, HUS, plus two other symptoms, fever and neurological symptoms, constitutes thrombotic thrombocytopenic purpura (TTP). The mortality rate of this condition in the elderly can be as high as 50%. Thus, there is a need of rapid and sensitive methods to detect this pathogen in foods. Traditional bacterial pathogen detection methods can take several days to confirm a positive sample. In the case of Campylobacter, culturing and plating takes 14–16 days for a positive result (Brooks and others 2004). Different selective media are used for the detection of particular bacteria species. They can contain inhibitors (in order to stop or delay the growth of nontargeted strains) or particular substrates that only the targeted bacteria can degrade or that confers a particular color to the growing colonies, such as rainbow agar for Salmonella detection. (Fratamico 2003). Although these methods are inexpensive they are very time consuming. To assure the safety of our foods and a rapid response to help those afflicted, clearly more rapid detection methods are needed.

Biosensors for Pathogen Detection Efforts to develop alternative food pathogen detection methods in general include the development of biosensor technologies for pathogen detection. Biosensors use biological receptor compounds (e.g., antibody, enzyme, nucleic acid, etc.) and the transduction of the molecular interaction through changes in physical and/or physicochemical properties in real time to detect the presence of the entity specific to the bioreceptor (Leonard and others 2003) Principally, there are four types of biosensors that measure signal transduction through changes in optical, mass, electrochemical, and thermal properties (Goepel 1991; Seyhi 1994; Goepel and Heiduschka 1995). Some of the general principles and applications of biosensors are briefly described as follows. Optical biosensors based on the evanescent wave (EW) use the technique of attenuated total reflection (ATR) spectroscopy and surface plasmon resonance (SPR) to measure real-time interaction between biomolecules. The basis of ATR is the reflection of light inside the core of a waveguide when the angle of incidence is greater than the critical angle. Waveguides can be slab guides, planar integrated optics, or optical fibers. Light waves are propagated along waveguides by the law of total internal reflection (TIR). Even though the light is totally internally reflected, the intensity does not abruptly fall to zero at the interface, resulting in generation of evanescent wave (EW), which penetrates exponentially into the medium of lower refractive index (Squillante 1998). The wavelength of light, ratio of the refractive indices, and angle of the light at the interface determine the penetration depth (Anderson and others 1993), which are typically 50 to 1000 nm; thus the EW is able to interact with many monolayers at the surface of waveguides (Lave and others 1991). Reactions occurring very close to the interface perturb the evanescent wave, and the changes in signals can be related to the amount of binding between the target and immobilized ligand at the interface. When metal surfaces are used to immobilize the receptors, the change in EW induced by the surface binding may also change the plasmon resonance of the surface

Applications of Immunomagnetic Beads and Biosensors

333

metal layer (SPR) (Tubb and others 1997). Thus, the main difference between ATR and SPR is that the former measures the changes of EW at the interphase directly and the latter measures the induced changes in the resonant excitation of the free electrons of the metal layer providing the anchoring sites for the specific receptor. Both ATR (Geng and others 2006) and SPR (Fratamico 1998) have been applied to measure food pathogens. Acoustic wave biosensors are based on the decrease of oscillating frequency of bioreceptor-coated piezoelectric crystals upon the binding of target analyte. The change in frequency is governed by the ratio of the mass of analyte and the piezoelectric crystal (Griffiths and Hall 1993). Recently, this type of approach has been applied to measure E. coli O157:H7 (Campbell and Mutharasan 2007). The sensitivity of this type of sensor is superior. However, the fabrication and treatment of the crystal require considerable technical training and expertise (Invitski and others 1999). The electrochemical sensor involves the use of a receptor-coated electrode that expresses a change in electro-properties upon the binding of target analyte. The best known electrochemical devise for measuring specific analyte is the glass pH-electrode that expresses potential change upon the binding of protons on the glass surface. In the medical field, the most widely used electrochemical sensor is the glucose monitoring sensor (D’Costa and others 1986). This approach has also been applied to measure food pathogens using alkaline phosphatase labeled antibodies to link to the bacteria that were captured by the antibodies immobilized on the electrode (Gehring and others 1996). The enzyme then was used to convert phenolic phosphate to phenolic compound that could be characterized by its specific redox potential. The magnitude of the redox current could relate to the number of pathogen captured. Thermometric biosensors exploit the fundamental property of biological reactions, i.e. absorption or evolution of heat (Spink and Wadsö 1976). This is reflected as a change in the temperature within the reaction medium. Its exploitation in biosensors led to the development of thermometric devices (Mosbach and Danielsson 1974). These predominantly measure the changes in temperature of the circulating fluid following the reaction of a suitable substrate with the immobilized enzyme molecules. The most basic version of such a device is a thermometer, routinely used for measurement of body or ambient temperature. Based on similar principles, in thermometric devices the heat is measured using sensitive thermistors. Such a device is popularly referred to as an enzyme thermistor, ET (Danielsson and Mosbach 1988). Several instruments were designed in the past 2 decades and they combined the principles of calorimetry, enzyme catalysis, immobilization on suitable matrices, and flow injection analysis for small metabolite detections. Because of its low sensitivity, the application of thermal sensor for pathogen detection has not yet been attempted. The detection of microorganisms by DNA amplification has been extensively applied. Using polymerase chain reaction (PCR) target nucleic segments of defined length and sequence are amplified by repetitive cycles of strand denaturation, annealing, and extension of oligonucleotide primers by the thermostable DNA polymerase, Thermus aquaticus (Taq) DNA polymerase (Bsat and others 1994). PCR has distinct advantages over culturing and other methods for the detection of microbial pathogens and offers the advantages of specificity, sensitivity, rapidity, accuracy, and capacity to

334

Section IV. Produce Safety during Processing and Handling

detect small amounts of target nucleic acid in a sample (Toze 1999). PCR has been shown to accurately detect low numbers of microbes such as viruses (Schwab and others 1996) and bacteria (Jensen and others 1993; Fach and Popoff 1997). Multiple primers can be used to detect different pathogens from one multiplex reaction. However, this technique can be limited by problems such as the sensitivity of the polymerase enzyme to environmental contaminants, difficulties in quantification, generation of false positives through the detection of naked nucleic acids, nonviable microorganisms, or contamination of samples in the laboratory (Toze 1999). Nucleic acid-based assays can indicate only the genetic potential of a microorganism to produce toxin or to express virulence and do not provide any information on toxins in foods or environmental samples. From a practical point of view, the routine detection of microbes using PCR can be expensive and complicated, requiring highly skilled workers to carry out the tests. All of these methods require samples of small volume and they may have the sensitivity to detect infectious dosage of pathogens in the sample (e.g., only a few cells of E. coli O157:H7). However, the low levels of pathogen may not be uniformly distributed, but highly localized in a food matrix. Assuring that the samples used for detection contain the target pathogen is a demanding challenge. To circumvent this problem, culture enrichment of multiple samples is often needed to increase cell concentration and thus enhance pathogen detection. Alternatively, immunomagnetic beads (IMB) can be used to rapidly and effectively separate and concentrate targeted pathogens, and the method has attracted increased interest (Molday and others 1977; Sinclair 1998). IMB have been applied in several rapid methods to capture pathogens prior to analysis (Fratamico and others 1992; Olsvik and others 1994). IMB also have been used to increase the signal intensity by concentrating captured pathogens into smaller detection volumes (Gehring and others 1996; Yu and Bruno 1996). In the past, we developed detection processes that involved first capturing targeted pathogens in foods from briefly enriched cultures by the use of specific IMB (Tu and others 2001a). The captured pathogens were further conjugated with second antibodies labeled with signal generating tags. The sandwiched complexes involving IMB, targeted pathogens, and labeled antibodies could be processed by the use of suitable magnetic devices. The captured pathogens were then revealed by different optical and electrical approaches. With this general approach, we were able to detect ∼1 CFU/g of target pathogens in meat samples in a standard 8-h shift (Tu and others 2001b). Some of those developed approaches were also applied to produce systems. In the following sections, we briefly summarize our experience of detecting pathogens in produce by IMB and biosensors.

Biosensor Processes Involving the Use of IMB for Pathogen Detection For these approaches, micron-sized iron-containing beads coated with antibodies specific to antigens of target organisms form IMB that are used to capture those targets. The captured pathogens and excess IMB can be easily separated from other solution components and conveniently transferred to a desired detection environment by the use of high-strength neodymium boron iron magnets associated with an automated

Applications of Immunomagnetic Beads and Biosensors

335

and programmable instrument (e.g., KingFisher apparatus of ThermoFisher Scientific, Waltham, MA). Several companies produce a variety of IMB, such as Dynal beads and Quantum Dots from Invitrogen Corp. (Carlsbad, CA), BioMag beads from Polysciences Inc. (Warrington, PA), ProMag beads from Bangs Laboratories (Fishers, IN), MACS Microbeads from Miltenyi Biotech Inc. (Auburn, CA), and MagSpheres from Luminex Corporation (Austin, TX). Our previous work has shown that based on hydrodynamic considerations, larger and denser beads have greater capturing efficiencies than lighter and smaller beads (Tu and others 2003a). These beads are used to capture bacteria or toxins from briefly enriched food samples, washed and then incubated with another antibody conjugated with a signaling tag that might generate easily detectable signals (e.g., absorption and fluorescence changes) through either enzymatic or chemical reactions (Tu and others 2001b; Li and others 2004). Detection of Pathogens on Cantaloupe Cantaloupes grow in contact with the earth, which increases their potential for contact with soilborne bacteria, fungi, insects, and animals. The possibility of a product becoming infected is compounded by contaminated irrigation water, improperly applied fertilizers, ineffective washing techniques, and poor hygiene practices of field workers. It has been shown that E. coli O157:H7 can survive up to 100 days in soil (Ingham and others 2004). Several outbreaks and recalls of cantaloupe have occurred, in particular a multistate outbreak strain of Salmonella on cantaloupes from Mexico that caused 133 cases from 2000–2002 (CDC 2002). In November 2006, more than 62,000 cases of cantaloupes from the western U.S. were recalled by Rio Vista, Ltd. of Rio Rico, Arizona, because routine sampling by the FDA tested positive for Salmonella (FDA 2006). Other microbes have been cited in outbreaks on cantaloupe, including Campylobacter and Norovirus (Bowen and others 2006). Because of its rough surface and porous veins, bacteria can attach to cantaloupe surfaces rather tightly, as evidenced by the difficulty in removing the bacteria through simple aqueous washings (Ukuku and others 2001). The bacteria may become incorporated into biofilms with existing microflora, which can further shield from the effects of washing or chemical treatments (Annous and others 2005). This noncompetitive relationship has been demonstrated by inoculating the surface with phytopathogenic mold, which does not inhibit the growth of subsequently inoculated Salmonella (Richards and Beuchat 2005). The waxy surface of the fruit can repel aqueous sanitizers (Beuchat and Ryu 1997). This strong attachment and hydrophobicity may add further complications to the detection, quantification, and reduction of suspected pathogens on the surfaces of cantaloupes. Sanitizing methods have included the use of hydrogen peroxide, chlorine, 94 °C water, (Ukuku 2006) as well as ozone, peroxyacetic acid, and chlorinated trisodium phosphate (Rogers and others 2004). Bacteria can be transferred to the flesh of the melons by cutting, which provides a surface that supports the growth of pathogenic bacteria (Del Rosario and Beuchat 1995). The possibility of crosscontamination from food service workers or other foods led to the proposal of a Hazard Analysis and Critical Control Point (HACCP) plan for handling of fresh produce (Beuchat 1995). Twenty cases of E. coli O157:H7 cross-contamination of cantaloupe were documented in Oregon in 1993 (Jackson and others 2000).

336

Section IV. Produce Safety during Processing and Handling

Because E. coli 0157:H7 bacteria can be a hazard on cantaloupe from the field as well as in postharvest handling, rapid and sensitive detection methods are needed. To determine the levels of E. coli 0157:H7 on cantaloupe, we tested the applicability of two different detection methods (Fig. 18.1) involving the use of immunomagnetic heads to first capture and concentrate E. coli 0157:H7 from cantaloupe samples. The captured bacteria (shown as B in the figure) were than detected either by the bioluminescence of cellular NAD(P)H or by a chemiluminescent sandwich assay. The results showed that the methods developed were capable of detecting relatively low levels of the E. coli 0157:H7 spiked on the surfaces of cantaloupes within 3.3 h (Tu and others 2004). NAD(P)H Method The NAD(P)H method mentioned above involves the measurement of cellular NAD(P) H via an externally added electron transfer system that uses membrane permeable menadione to oxidize internal NAD(P)H to NAD(P)+. Menadione reduces molecular oxygen to hydrogen peroxide (H2O2) that generates chemiluminescent luminol by the action of horseradish peroxidase (Fig. 18.2). This reaction system utilized the cellular NAD(P)H and membrane-bound electron transfer process to produce luminolsupported and peroxidase-catalyzed chemiluminescence. As described in our previous report (Tu and others 2004), the NAD(P)H method was useful to measure the presence of viable cells. Figure 18.2 shows the method

NAD (P)H Method

IMB

Sandwich Method

IMB

B

B

NAD (P)H in cell Catalyses reaction

Peroxidase labeled antibody

menadione

menadiol H2O2 + Luminols

Luminols

+

–

O2

Peroxidase

O2

Luminescence, hn + oxidized products

Figure 18.1. Two luminescent assays for bacterial detection. The bacteria (B) either provide the cellular reducing power in the forms of NAD(P)H or are sandwiched with peroxidase labeled antibody.

Applications of Immunomagnetic Beads and Biosensors

337

Luminescence Intensity

10,000 8,000 6,000 4,000 2,000 0 Background

100000

1000000

10000000

CFU/mL

Luminescence intensity

Figure 18.2.

Sensitivity of NAD(P)H method.

9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 0

3.1x10e4

3.1x10e5

3.1x10e6

CFU/mL Figure 18.3.

Detection levels of bacteria inoculated onto cantaloupe.

could measure 106 CFU/ml of E. coli O157:H7. The solid line in the figure represented twice the value of background and was arbitrarily chosen as the limit of detection. We spiked cantaloupe surfaces with different levels of E. coli O157:H7 in circular areas of 2 cm diameter. After drying for 1 h to promote adhesion, the spiked area was removed as discs and incubated in nutrient broth for 3.3 h. As illustrated in Figure 18.3, the NAD(P)H method was able to detect the presence of ∼3 × 106 CFU of the

338

Section IV. Produce Safety during Processing and Handling

bacteria in an area of ∼3 cm2 of cantaloupe surface. The background signal is higher in cantaloupe assay due to presence of other materials that increase the false positive (rind, pulp, endogenous bacteria). The detection after 3.3 h is noteworthy because the discs were added to 100 ml nutrient broth and only 200 μl were assayed.

Luminescence Intensity

Sandwich Method This assay uses bacteria captured by beads and then incubated with peroxidase-labeled antibody against the target antigen. The beads are eluted into a luminol-based cocktail (BioFX Corp., Owings Mills, MD). The assay measures the activity of the peroxidase conjugated to anti–E. coli 0157 antibodies. The amount of activity is proportional to the concentration of bacteria captured by the beads. Unlike the NAD(P)H method, the sandwich method detects intact (viable and injured) and fragmented target cells, because the sandwich detection is relying on interactions between applied antibodies and antigens on bacterial surfaces. As shown in Figure 18.4, the sensitivity of the sandwich method in detecting cultured E. coli O157:H7 is more than 10 times better than that of the NAD(P)H method. The horizontal bar represents a value twice the background reading and is arbitrarily set as the limit of detection. Thus 100,000 CFU/ml is approximately the detection limit using this method. This sandwich method was applied to detect E. coli O157:H7 spiked on cantaloupe as described for Figure 18.3. As shown in Figure 18.5, the sandwich method could detect ∼1000 CFU/cm2 of cantaloupe surface after an enrichment of 3.3 h (Tu and others 2004). As with the NAD(P)H method, the background of cantaloupe samples was higher than that of the pure cultures cells but not as relatively high as the NAD(P)H method. Thus, a lower number of cells can be detected after the incubation period using the sandwich method. The lower background and greater sensitivity make the sandwich test more desirable for detecting E. coli O157:H7 in cantaloupe samples.

450,000 400,000 350,000 300,000 250,000 200,000 150,000 100,000 50,000 0 Background

10000

100000 CFU/mL

Figure 18.4.

Sensitivity of the sandwich method.

1000000

Luminescence Intensity

Applications of Immunomagnetic Beads and Biosensors

339

80,000 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 0

3.1x10e2

3.1x10e3

CFU/mL Figure 18.5. cantaloupe.

Sandwich method detection levels of bacteria inoculated onto

Detection of Pathogens Associated with Alfalfa Sprouts The use of uncooked sprouts as a salad ingredient has gained considerable acceptance by American consumers. However, outbreaks of E. coli 0157:H7 and Salmonella spp. associated with the consumption of raw sprouts have become a concern (Breuer and others 2001; Brooks and others 2001; CDC 2001; Ferguson and others 2005). Pathogenic bacteria on the seeds, if any, can rapidly grow to 106–108 CFU/g of product under the warm, moist, nutrient rich sprouting conditions (NACMCF 1999; Stewart and others 2001a). The internalization of the pathogenic bacteria into the edible parts of the sprouts, the cotyledons and hypocotyls, makes them difficult to disinfect after sprouting (Gandhi and others 2001; Hara-Kudo and others 1997; Ito and others 1998). In 1999, the FDA recommended that sprout growers decontaminate seeds to reduce the microbial hazards of sprouts (FDA 1999). While currently approved treatments of seeds may reduce 99–99.9% of microbial populations, they do not guarantee a pathogen-free sprouted product (Beuchat and Ryu 1997; Brooks and others 2001; Jaquette and others 1996; NACMCF 1999; Pandrangi and others 2003; Proctor and others 2001; Stewart and others 2001b; Taormina and others 1999; Weissinger and Beuchat 2000). Testing seeds for contamination is problematic due to the low levels of contamination and nonuniform distribution of pathogens (Splittstoesser and others 1983). The FDA has recommended testing spent irrigation water from sprout production for the presence of E. coli 0157:H7 and Salmonella spp. (FDA 1999). Testing spent irrigation water has many advantages over testing sprouts. In order to test the sprouts, multiple samples must be taken from various areas of the sprouting drum to ensure that the sampling is representative of the microflora present. Also, pummeling of the sprouts to break them open prior to testing, as recommended by the FDA (FDA 1999), may release phytoalexins inhibitory to the growth of some pathogens during enrichment or isolation (Jaquette and others 1996). The only disadvantage of testing spent irrigation water is that the level of microorganisms recovered is generally 1 log lower

340

Section IV. Produce Safety during Processing and Handling

than the level in the sprouts and low levels of pathogens may be missed (Fu and others 2001). However, if the testing of spent irrigation water is conducted at 48 h after the commencement of sprouting, as recommended by the FDA, the level of pathogens present in the irrigation water will be at a maximum level (Fu and others 2001; Splittstoesser and others 1983; Stewart 2001a,b) and may be more readily detectable. For testing irrigation water, the FDA (1999) has recommended using VIP EHEC (Biocontrol Systems, Bellview, WA) or Reveal E. coli 0157:H7 tests (Neogen Corp., Lansing, MI) for the detection of E. coli 0157:H7, and Assurance Gold Salmonella EIA or Visual Immunoprecipitate (VIP) assay for Salmonella (both from Biocontrol Systems, Inc., Bellview, WA) for the detection of Salmonella spp. However, the E. coli tests require an overnight incubation in modified buffered peptone water with three added antibiotics, and the Salmonella methodology requires preenrichment and enrichment for approximately 48–50 h before testing. Thus, there is a need to develop sensitive and specific alternatives that can be completed in shorter time periods. Time-Resolved Fluorescence of Lanthanide Cations The involvement of 4f orbitals in the electronic structure of lanthanide (La) cations such as europium permits a transfer of excitation energy from ligands to central La cations prior to the emission of ion fluorescence that is characteristic by a relatively long fluorescence half-life (∼50–1000 msec) and a considerable Stoke’s shift (>200 nm) between the absorption and emission maxima. In contrast, the fluorescence half-life and Stoke’s shift of common organic and biochemical compounds are in the range of 1–1000 μsec and 20–100 nm, respectively. Thus, with a pulsed excitation, the fluorescence of La may be easily separated from the interference fluorescence and scattered excitation light by delaying the emission measurement (time-resolved fluorescence, TRF). In addition, the quantum yield of La-chelates is usually quite high, e.g., 0.18 for Eu-(4,4,4-trifluoro-[Z-thienyl-l,3-butanedionato]) at 614 nm (Halverson and others 1964). A combination of time-delayed fluorescence and the unique properties of Lachelates has led to the development of a new technique called dissociation-enhanced lanthanide fluoroimmunoassay (DELFIA). In this technique, antibodies are modified to contain binding groups capable of forming very low fluorescence La-complex. The modified antibodies are used to capture target species and the antibody-bound La cations are then extracted out by an “enhancement solution” that contains chelates capable of forming strongly fluorescent products (Tu and others 2001b). Time-Resolved Fluorescence Approach (TRF) We have applied TRF measurement in combination with immunomagnetic capture to develop a sensitive and rapid method for pathogen detection (IMB-TRF). This approach has been demonstrated to detect E. coli and Salmonella in ground meats at a 1 CFU/g level and to show that nontarget microorganisms do not interfere with the detection (Tu and others 2002). In the experiment, europium (Eu) or samarium (Sm) labeled antibodies to the bacteria were incubated with IMB captured target organisms. Excess labeled antibody was washed away and the remaining beads were eluted into an “enhancement buffer,” which contained chelators that extracted Eu or Sm to form strongly fluorescent products. The samples were then read in a Victor 1420 Multilabel

Applications of Immunomagnetic Beads and Biosensors

341

1.0E+07 100000

TRF Signal (log)

1.0E+06

10000 1.0E+05

1000

1.0E+04

1.0E+03 H7 F4546

NM 98A06026

E. coli O157

Bredeney 3VIPHE

Muenchen HERV2C

Salmonella

Figure 18.6. The sensitivity of IMB-TRF method for detecting different strains of E. coli and Salmonella.

Counter that provided pulsed excitation and delayed emission measurements (PerkinElmer Wallac, Turku, Finland). Figure 18.6 shows the sensitivity of the IMB-TRF method for detecting different strains of E. coli and Salmonella. Bacterial samples with indicated concentrations were first treated with proper IMBs. The captured bacteria were then reacted with either Eu-labeled (for E. coli) or Sm-labeled (for Salmonella) antibodies prior to TRF measurements (Tu and others 2002). As shown, detection limits were thus 750 CFU/ml for E. coli and 250 CFU/ml for Salmonella. Similar trends were found with Salmonella strains Stanley HO558, Anatum 4317, Infantis F4319, and Newport H1275 (data not shown). To test the feasibility of using the developed IMB-TRF method to detect the pathogens in the alfalfa sprouts, experiments shown in Figure 18.7 were performed (Tu and others 2003b). Commercially obtained alfalfa sprouts were inoculated with the pathogens at indicated levels and then “stomachered” and incubated for 4.5 h at 37 °C. Data shown indicated that with crushed sprouts, the IMB-TRF could easily detect the presence of E. coli O157:H7 and E. coli O157:NM. However, the approach failed to detect Salmonella under the experimental conditions. The basis for this has yet to be determined but it has been reported by Castro-Rosas and Escartin (2000) that Vibrio cholerae O1 and Salmonella typhi showed no growth when inoculated onto alfalfa sprouts 24 h after germination. They attributed this observation to the abundance of competing background microflora at 24 h into the germination process. Detection of Pathogens in Laboratory-Cultivated Sprouts Grown from Inoculated Seeds Results of Figure 18.6 indicated that using whole alfalfa sprouts for pathogen detection by IMB-TRF might have its drawbacks. Thus, we decided to germinate contaminated

342

Section IV. Produce Safety during Processing and Handling

TRF signal (log)

1.0E+06

1.0E+05 1 10 100

1.0E+04

1.0E+03 H7 E. coli O157:H7

NM

Bredeney Salmonella

Muenchen

CFU/mL Figure 18.7.

Detection of pathogens on crushed sprouts.

alfalfa seeds (4 CFU/g) in our laboratory using sterile Mason jars and sterile tap water for irrigation. Thus, an analysis of spent irrigation water should indicate whether the sprouts, and therefore the seeds, are contaminated by pathogens. Experimentally, alfalfa seeds artificially contaminated with E. coli O157 or Salmonella were used to produce sprouts (Tu and others 2002). As shown in Figure 18.8, water sample analyses were applicable for the detection of both E. coli O157 and Salmonella. Detection of Pathogens in Irrigation Water and Sprouts Although spent irrigation water testing is recommended by the FDA (1999), it may give false negative results because microbial counts in the irrigation water are, on the average, 1 log lower than those detected in sprout samples (FDA 1999; Fu and others 2001). For this reason, we chose to apply the developed detection method to both the water and sprouts simultaneously. We did not utilize the Seward Stomacher for pummeling sprouts germinated from lab-inoculated seeds because of complications described in Figure 18.6. Instead, whole sprouts were aseptically transferred to the proper culture medium for the enrichment. With this experimental design, both the sprouts and the spent irrigation water equally showed the presence of the pathogens as depicted in Figure 18.9. Unlike the results described in Figure 18.7, the use of sprouts germinated from contaminated seeds, under applied laboratory conditions, showed positive detection for both E. coli O157 and Salmonella. Apparently, the

TRF Signal (log)

1.0E+07

1.0E+06

1.0E+05

1.0E+04

Figure 18.8.

Muenchen

Anatum

Infantis

Bredeney Salmonella

NM

H7 E. coli O157

1.0E+03

Detection of pathogens in irrigation water.

TRF Signal (log)

1.0E+07

1.0E+06

1.0E+05 Sprouts Water

1.0E+04

1.0E+03 NM E. coli O157

H7

Infantis Salmonella

Muenchen

Figure 18.9. Detection of pathogens in sprouts germinated from contaminated seeds.

343

344

Section IV. Produce Safety during Processing and Handling

competitive exclusion of pathogens by background microflora is minimized by the use of pathogen-inoculated seeds, and this methodology more closely approximates naturally occurring contamination of seeds.

Conclusions The number of outbreaks linked to fresh produce reported to the CDC has increased in the last years (Bean and Griffin 1990; CDC 2000). This increase may be due in part to improved surveillance, but other factors may also come into play. Proposed reasons include the significant increase in the consumption of fresh produce in the United States due to the growing awareness of fruits and vegetables as a part of a healthy diet. Also, greater volumes of minimally processed produce are being shipped from central locations and distributed over much larger geographical areas to meet the increased popularity of salad bars. This, coupled with increased global trade, significantly increases human exposure to a wide variety of foodborne pathogens and also increases the chances of outbreaks (Harris and others 2003). To minimize the possibility of outbreaks, producers such as the International Sprout Growers Association (ISGA) have taken positive steps to address this problem by pursuing the use of 2% calcium hypochlorite for soaking alfalfa seeds prior to germination and growth. This intervention method has the potential to substantially reduce, but not necessarily eliminate, pathogenic microbial contamination of seeds that can be passed on to the consumer through ingestion of raw sprouts. Thus, the development of effective technologies that can be applied to detect pathogenic bacteria in produce is desirable. In this chapter, we provide evidence demonstrating that a combination of IMB to capture and biosensors to detect (a method that was originally developed for detecting pathogens in meats) has the potential to detect low levels of pathogenic bacteria in produce, specifically cantaloupes and alfalfa seeds and sprouts. The sensitivity of developed IMB-TRF processes allows rapid detection, within an 8-h shift, of select pathogens even in the presence of high levels of the background microflora. In principle, the technology may be applied to the detection of pathogens in other produce that are of outbreak concerns. The availability of magnetic bead manipulator and biosensor detector in 96-well formats will certainly increase the feasibility of highthroughput screening of pathogens in meats and produce.

Disclaimer Mention of brand or firm names does not constitute an endorsement by the U.S. Department of Agriculture or any other federal agency or governmental entity over others of similar nature not mentioned.

References Ackers ML, Mahon BE, Leahy E, Goode B, Damrow T, Hayes PS, Bibb WF, Rice DH, Barrett TJ, Hutwagner L, Griffin PM, and Slutsker L. 1998. An outbreak of Escherichia coli 0157:H7 infections associated with leaf lettuce consumption. Journal of Infectious Disease 177:1588–1593. Anderson GP, Golden JP, and Ligler FS. 1993. A fiber optic biosensor—combination tapered fibers designed for improved signal acquisition. Biosensors and Bioelectronics 8:249–256.

Applications of Immunomagnetic Beads and Biosensors

345

Annous BA, Solomon EB, Cooke PH, and Burke A. 2005. Biofilm formation by Salmonella spp. on cantaloupe melons. Journal of Food Safety 25:276–287. Bean NH, and Griffin PM. 1990. Foodborne disease outbreaks in the United States, 1973–1987: pathogens, vehicles, and trends. Journal of Food Protection 53(9):804–817. Beuchat LR. 1999. Survival of enterohemorrhagic Escherichia coli 0157:H7 in bovine feces applied to lettuce and the effectiveness of chlorinated water as a disinfectant. Journal of Food Protection 62:845–849. Beuchat LR. 1995. Pathogenic microorganisms associated with fresh produce. Journal of Food Protection 59(2):204–216. Beuchat LR, and Ryu J. 1997a. Produce handling and processing practices. Emerging and Infectious Diseases 3:159–465. Beuchat LR. 1997b. Comparison of chemical treatments to kill Salmonella on alfalfa seeds destined for sprout production. International Journal of Food Microbiology 34:329–333. Bowen A, Fry A, Richards G, and Beauchat L. 2006. Infections associated with cantaloupe consumption: a public health concern. Epidemiology and Infection 134:667–674. Breuer T, Benkel DH, Shapiro RL, Hall WN, Winnett MM, Linn MJ, Neimann J, Barrett TJ, Dietrich S, Downes FP, Toney DM, Pearson JL, Rolka H, Slutsker L, and Griffin PM. 2001. A multistate outbreak of Escherichia coli 0157:H7 infections linked to alfalfa sprouts grown from contaminated seeds. Emerging and Infectious Diseases 7:977–982. Brooks BW, Devenish J, Lutze-Wallace JCL, Milnes D, Robertson RH, and Berlie-Surujballi G. 2004. Microbial immunity in the reproductive tract. Vet Microbiol 103:77–84. Brooks JT, Rowe SY, Shillam P, Heltzel DM, Hunter SB, Slutsker L, Hoekstra RM, and Luby SP. 2001. Salmonella typhimurium infections transmitted by chlorine-pretreated clover sprout seeds. American Journal of Epidemiology 154:1020–1028. Bsat N, Weidmenn M, Czajka J, Barany F, Piani M, and Batt CA. 1994. Food safety applications of nucleic acid-based assays. Food Technology 48:142–145. Campbell GA, and Mutharasan R. 2007. A method of measuring Escherichia coli O157:H7 at 1 cell/mL in 1 liter sample using antibody functionalized piezoelectric-excited millimeter-sized cantilever sensor. Environmental Science and Technology 41:1668–1674. Castro-Rosas J, and Escartin EF. 2000. Survival and growth of Vibrio cholera 01, Salmonella typhi, and E. coli in alfalfa sprouts. Journal of Food Science 65:162–165. CDC (Centers for Disease Control and Prevention). 1999. Outbreak of Escherichia coli 0157:H7 and Campylobacter among attendees of the Washington county fair—New York, 1999. Morbidity and Mortality Weekly Report 48:803–804. ———. 2000. CDC surveillance summaries; March 17, 2000. Morbidity and Mortality Weekly Report 49:1–5. ———. 2001. Outbreaks of Salmonella serotype Kottbus infection associated with eating alfalfa sprouts— Arizona, California, Colorado, and New Mexico, February–April 2001. Morbidity and Mortality Weekly Report 51:7–9. ———. 2002. Multistate Outbreaks of Salmonella Serotype Poona Infections Associated with Eating Cantaloupe from Mexico—United States and Canada, 2000—2002. Morbidity and Mortality Weekly Report 51:1044–1047. ———. 2003. Hepatitis A Outbreak Associated with Green Onions at a Restaurant—Monaca, Pennsylvania. http://www.cdc.gov/od/oc/media/mmwrnews/n031128.htm#mmwr4 ———. 2006. Multistate Outbreak of E. coli O157 Infections, November–December 2006. http://www. cdc.gov/ecoli/2006/december/121406.htm. Chalmers RM, Aird H, and Bolton FJ. 2000. Waterborne Escherichia coli 0157. Journal of Applied Microbiology 88:1245–1325. Danielsson B, and Mosbach K. 1988. Enzyme thermistors. Methods in Enzymology 137:181–197. D’Costa EJ, Higgins IJ, and Turner APF. 1986. Quinoprotein glucose dehydrogenase and its application in an amperometric glucose sensor. Biosensors 2:71–87. Del Rosario BA, and Beuchat LR. 1995. Survival and growth of enterohemorrhagic Escherichia coli 0157:H7 in cantaloupe and watermelon. Journal of Food Protection 58:105–107. Fach P, Popoff MR. 1997. Detection of enterotoxigenic Clostridium perfringens in food and fecal samples with duplex PCR and slide latex agglutination test. Applied and Environmental Microbiology 63:4232–4236.

346

Section IV. Produce Safety during Processing and Handling

Faith NG, Shere JA, Brosch R, Arnold KW, Ansay SE, Lee MS, Luchansky JB, and Kaspar CW. 1996. Prevalence and clonal nature of Escherichia coli 0157:H7 on dairy farms in Wisconsin. Applied and Environmental Microbiology 62:1519–1525. Ferguson D, Scheftel J, Cronquist A, Smith K, Woo-Ming A, Anderson E, Knutsen J, De AK, and Gershman K. 2005. Temporally distinct Escherichia coli O157 outbreaks associated with alfalfa sprouts linked to a common seed source—Colorado and Minnesota, 2003. Epidemiology and Infection 133:439–447. FDA (Food and Drug Administration). 1998. Guide to minimize microbial food safety hazards for fresh fruits and vegetables. Center for Food Safety and Applied Nutrition, Washington, D.C. ———. 1999. Guidance for industry: reducing microbial food safety hazards for sprouted seeds and guidance for industry: sampling and microbial testing of spent irrigation water during sprout production. Fed Reg 64:57893–57902. ———. 2006. Rio Vista, Ltd. Recalls Cantaloupes Nationwide Because of Possible Health Risk http:// www.fda.gov/oc/po/firmrecalls/riovista11_06.html. Fratamico P. 1998. Detection of Escherichia coli O157:H7 using a surface plasmon resonance biosensor. Biotechnology and Technology 12:571–576. Fratamico PM. 2003. Comparison of culture, polymerase chain reaction (PCR), TaqMan Salmonella, and Transia Card Salmonella assays for detection of Salmonella spp. in naturally contaminated ground chicken, ground turkey, and ground beef. Molecular and Cellular Probes 17:215–221. Fratamico PM, Schultz FJ, and Buchanan RL. 1992. Rapid isolation of Escherichia coli 0157:H7 from enrichment cultures of foods using an immunomagnetic separation method. Food Microbiology 9:105–113. Fu T, Stewart D, Reineke K, Ulaszek J, Schlesser J, and Tortorello M. 2001. Use of spent irrigation water for microbial analysis of alfalfa sprouts. Journal of Food Protection 64:802–806. Gandhi M, Golding S, Yaron S, and Matthews KR. 2001. Use of green fluorescent protein expressing Salmonella Stanley to investigate survival, spatial location, and control on alfalfa sprouts. Journal of Food Protection 64:1891–1898. Gehring AG, Crawford CG, Mazenko R, Van Houten LJ, and Brewster JD. 1996. Enzyme-linked immunomagnetic electrochemical detection of Salmonella typhimurium. Journal of Immunological Methods 195:15–25. Geng T, Uknalis J, Tu S, and Bhunia AK. 2006. Fiber-optic biosensor employing alexa-fluor conjugated antibody for detection of Escherichia coli O157:H7 from ground beef in four hours. Sensor 6:796–807. Goepel W. 1991. Chemical sensing, molecular electronics and nanotechnology: interface technologies down to the molecular scale. Sensors and Actuators B 4:7–21. Goepel W, Heiduschka P. 1995. Interface analysis in biosensor design. Biosensors and Bioelectronics 10:853–883. Griffiths D, and Hall G. 1993. Biosensors—what real progress is being made? TIBTECH 11:122–130. Halverson F, Brinen FJ, and Leto JR. 1964. Photoluminescence of Lanthanide complexes. II. Enhancement by an insulating sheath. Journal of Chemical Physics 41:157–163. Hara-Kudo Y, Konuma H, Iwaki M, Kasuga F, Sugita-Konishi Y, Ito Y, and Kumagai SS. 1997. Potential hazard of radish sprouts as a vehicle of Escherichia coli 0157:H7. Journal of Food Protection 60:1125–1127. Harris LJ, Farber JN, Beuchat LR, Parish ME, Suslow TV, Garrett EH, and Busta FF. 2003. Outbreaks associated with fresh produce: incidence, growth, and survival of pathogens in fresh and fresh-cut produce. Comprehensive Reviews in Food Science and Food Safety 2:78–141. Hillborn ED, Mermin JH, Mshar PA, Hadler JL, Voetsch A, Wojtkunski C, Swartz M, Mshar R, LambertFair MA, Farrar JA, Glynn MK, and Slutsker L. 1999. A multistate outbreak of Escherichia coli 0157:H7 infections associated with consumption of mesclun lettuce. Archives of Internal Medicine 159:1758–1764. Ingham S, Losinski J, Andrews MP, Breuer JE, Breuer JR, Wood TM and Wright TH. 2004. Escherichia coli contamination of vegetables grown in soils fertilized with noncomposted bovine manure: garden scale studies. Applied and Environmental Microbiology 70:6420–6427. Invitski D, Abdel-Hamid I, Atanasov P, and Wilkins E. 1999. Biosensors for the detection of pathogenic bacteria. Biosensors and Bioelectronics 14:599–624.

Applications of Immunomagnetic Beads and Biosensors

347

Ito Y, Sugita-Konishi Y, Kasuga F, Iwaki M, Hara-Kudo Y, Saito N, Noguchi Y, Honuma H, and Sumagai S. 1998. Enterohemorrhagic Escherichia coli 0157:H7 present in radish sprouts. Applied and Environmental Microbiology 64:1532–1535. Jackson LA, Keene WE, McAnulty JM, Alexander ER, Diermayer M, Davis MA, Hedberg K, Boase J, Barrett TJ, Samadpour M, and Flemimg DW. 2000. Where’s the beef? The role of cross contamination in 4 chain restaurant associated outbreaks of E. coli O157:H7 in the Pacific Northwest. Archives of Internal Medicine 160(15):2380–2385. Jaquette CB, Beuchat, LR, and Mahon BE. 1996. Efficacy of chlorine and heat treatment in killing Salmonella stanley inoculated onto alfalfa seeds and growth and survival of the pathogen during sprouting and storage. Applied and Environmental Microbiology 62:2212–2215. Jensen M, Webster JA, and Straus N. 1993. Rapid identification of bacteria on the basis of polymerase chain reaction-amplified ribosomal DNA spacer polymorphisms. Applied and Environmental Microbiology 59:945–952. Kudva LT, Blanch K, and Hovde CJ. 1998. Analysis of Escherichia coli 0157:H7 survival in ovine or bovine manure and manure slurry. Applied and Environmental Microbiology 64:3166–3174. Lave WF, Button LJ, and Slovacek RE. 1991. In: Biosensors with Fiberoptics; Wise DL and Wingard LB. Ed.; Humana Press: Clifton, NJ, 139–180. Leonard P, Hearty S, Brennan J, Dunne L, Quinn J, Chakraborty T, and O’Kennedy R. 2003. Advances in biosensors for detection of pathogens in food and water. Enzyme and Microbiological Technology 32:3–13. Li Y, Dick WA, Tuovinen OH. 2004. Fluorescence microscopy for visualization of soil microorganisms—a review. Biology of Fertility Soils 5:301–311. Molday RS, Yen SPS, and Rembaum A. 1977. Application of magnetic microspheres in labeling and separation of cells. Nature 268:437–438. Mosbach K, and Danielsson B. 1974. An enzyme thermistor. Biochimica Biophysica Acta 364:140– 145. NACMCF (National Advisory Committee on Microbiological Criteria for Foods). 1999. Microbiological safety evaluations and recommendations on sprouted seeds. International Journal of Food Microbiology 52:123–153. Olsvik O, Popovic T, Skjerve E, Cudjoe KS, Bornes E, Ugelstad J, and Uhlen M. 1994. Magnetic separation techniques in diagnostic microbiology. Clinical Microbiology Reviews 7:43–54. Pandrangi S, Elwell MW, Anahmeswaran RC, and Laborde LF. 2003. Efficacy of sulfuric acid scarification and disinfectant treatments in eliminating Escherechia coli 0157:H7 from alfalfa seeds prior to sprouting. Journal of Food Science 68:613–618. Proctor ME, Hamacher M, Tortorello ML, Archer JA, and Davis JP. 2001. Multistate outbreak of Salmonella serovar Muenchen infections associated with alfalfa sprouts grown from seeds pretreated with calcium hypochlorite. Journal of Clinical Microbiology 39:3461–3465. Richards GM, and Beuchat LR. 2005. Metabolic association of molds and Salmonella poona on intact and wounded cantaloupe rind. International Journal of Food Microbiology 97:327–339. Rogers SL, Cash JN, Siddiq M, and Ryser ET. 2004. A comparison of different chemical sanitizers for inactivating Escherichia coli O157:H7 and Listeria monocytogenes in solution and on apples, lettuce, strawberries, and cantaloupe. Journal of Food Protection 67:721–731. Schwab KJ, De Leon R, and Sobsey MD. 1996. Immunoaffinity concentration and purification of waterborne enteric viruses for detection by reverse transcriptase PCR. Applied and Environmental Microbiology 63:4401–4407. Seyhi RS. 1994. Transducer aspects of biosensors. Biosensors and Bioelectronics 9:243–264. Sinclair B. 1998. To bead or not to bead: Applications of magnetic bead technology. The Scientist 12:17–24. Spink C, and Wadsö I. 1976. Calorimetry as an analytical tool in biochemistry and biology. Methods in Biochemical Analysis 23:1–159. Splittstoesser DF, Queale DT, and Andaloro BW. 1983. The microbiology of vegetable sprouts during commercial production. Journal of Food Safety 5:79–86. Squillante E. 1998. Applications of fiber-optic evanescent wave spectroscopy. Drug Development and Industrial Pharmacy 24:1163–1175.

348

Section IV. Produce Safety during Processing and Handling

Stewart DS, Reineke KF, Ulaszek JM, and Totorello ML. 2001a. Growth of Escherichia coli O157:H7 during sprouting of alfalfa seeds. Letters in Applied Microbiology 33:95–99. ———. 2001b. Growth of SalmonelIa dungri sprouting of alfalfa seeds associated with salmonellosis outbreaks. Journal of Food Protection 64:618–622. Surak JG. 2007. A Recipe for Safe Food: ISO 22000 and HACCP. Quality Progress October:21–27. Taormina PJ, Beuchat LR, and Slutsker L. 1999. Infections associated with eating seed sprouts: an internationa1 concern. Emerging and Infectious Diseases 5:626–634. Toze S. 1999. PCR and the detection of microbial pathogens in water and wastewater. Water Research 33:3545–3556. Tu S, Golden M, Andreotti P, and Irwin P. 2002. The use of time resolved fluoroimmunoassay to simultaneously detect Escherichia coli 0157:H7, Salmonella enterica serovar typhimurium and Salmonella enterica serovar enteriditis in foods. Journal of Rapid Methods and Automation Microbiology 10:7–48. Tu S, Golden M, Andreotti P, Yu LSL, and Irwin PL. 2001b. Applications of time-resolved fluoroimmunoassay to detect magnetic seed captured Escherichia coli O157:H7. Journal of Rapid Methods and Automation in Microbiology 9:71–84. Tu S, Golden MH, Fett WF, Gehring AG, and Irwin PL. 2003b. Rapid detection of outbreak Escherichia coli O157 and Salmonella on alfalfa sprouts by immunomagnetic capture and time resolved fluorescence. Journal of Food Science 23:75–89. Tu S, Patterson DL, Briggs CE, Irwin PL, and Yu LSL. 2001a. Detection of immunomagnetically captured Escherichia coli O157:H7 by antibody-conjugated alkaline phosphatase. Journal of Industrial Microbiology and Biotechnology 26:345–349. Tu S, Uknalis J, and Gehring A. 2004. Optical methods for detecting Escherichia coli O157:H7 spiked on cantaloupes Nondestructive Sensing for Food Safety, Quality, and Natural Resources Proc. of SPIE, Soc. of Photo-Optical Instrumentation Engineer Bellingham, WA. 5587:183–189. Tu S, Uknalis J, Gore M, Irwin PL, and Feder I. 2003a. Factors affecting the bacterial capture efficiency of immuno beads: a comparison between beads with different size and density. Journal of Rapid Methods and Automation in Microbiology 11:35–46. Tubb AJC, Payne FP, Millington RB, and Lowe CR. 1997. Single mode optical fibre surface plasmon wave chemical sensor. Sensors and Actuators B 4:71–79. Ukuku DO. 2006. Effect of sanitizing treatments on removal of bacteria from cantaloupe surface, and recontamination with Salmonella. Food Microbiology 23:289–293. Ukuku DO, Pilizota V, and Sapers GM. 2001. Bioluminescence ATP assay for estimating total plate counts of surface micro-flora of whole cantaloupe and determining efficacy of washing treatment. Journal of Food Protection 64:813–819. Wang G, and Doyle MP. 1998. Survival of enterohemorrhagic Escherichia coli 0157:H7 in water. Journal of Food Protection 61:662–667. Wang G, Zhao T, and Doyle MP. 1996. Fate of enterohemorrhagic Escherichia coli 0157:H7 in bovine feces. Applied and Environmental Microbiology 62:2567–2570. Weissinger WR, and Beuchat LR. 2000. Comparison of aqueous chemical treatments to eliminate Salmonella on alfalfa sprouts. Journal of Food Protection 63:1475–1482. Yu H, and Bruno JG. 1996. Immunomagnetic-electrochemiluminescent detection of Escherichia coli 0157:H7 and Salmonella typhimurium in foods and environmental water samples. Applied and Environmental Microbiology 62:587–592.

Section V Public, Legal, and Economic Perspectives

19 Public Response to the 2006 Recall of Contaminated Spinach William K. Hallman, Cara L. Cuite, Jocilyn E. Dellava, Mary L. Nucci, and Sarah C. Condry

Introduction On September 14, 2006, the U.S. Food and Drug Administration (FDA) issued an advisory to consumers not to eat bagged fresh spinach because of suspected contamination by E. coli O157:H7 (U.S. Food and Drug Administration 2006a). This advisory was based on information provided by the Centers for Disease Control and Prevention (CDC) regarding a multistate foodborne illness outbreak that, by that time, had already caused 50 cases of illness, resulting in 8 cases of kidney failure and 1 death, and which was thought to be associated with the consumption of fresh spinach. In its advisory, the FDA described the symptoms of illnesses resulting from ingestion of E. coli O157:H7 as causing diarrhea, often with bloody stools (they mentioned no other symptoms). They urged those who believed that they were experiencing symptoms of illness after consuming bagged spinach to contact their health care providers. The following day, the FDA’s advisory was expanded to include all fresh spinach and products containing fresh spinach. In part, this extension was in response to reports that retailers sometimes opened bagged spinach and sold it as loose spinach (Brackett 2006), making it available on salad bars and in produce sections of markets. At the same time, a series of voluntary recalls of fresh spinach began nationwide, beginning with products from Natural Selection Foods, LLC, of San Juan Bautista, California, with “Best if Used by Dates” of August 17, 2006, through October 1, 2006 (U.S. Food and Drug Administration 2006b). As one of the largest packers of fresh spinach in the nation, Natural Selection Foods produced bagged spinach under 30 brand names, all of which were part of the recall. Within a week, 5 additional firms issued their own voluntary recalls after discovering that they had included spinach processed by Natural Selections Foods as components of their mixed salad greens or as ingredients in pizzas and some other prepared foods. In response to the FDA’s initial advisory and the subsequent recalls, stores and restaurants quickly removed fresh spinach from their shelves and menus. Harvesting, marketing, and bulk sales of fresh spinach also came to a halt (Calvin 2007). News coverage of the advisory and the subsequent recalls was extensive, including both the evening and morning shows of the three broadcast television networks: ABC, CBS, and NBC (Nucci 2008) and national and regional newspapers (Pleasant 2008). By the time the recalls were issued, contaminated product had already been consumed. Most of those who would eventually become sick from the contaminated spinach had already become ill between August 19 and September 5, 2006 (U.S. Centers for Disease Control and Prevention 2006c). 351

352

Section V. Public, Legal, and Economic Perspectives

On September 20, the FDA issued an updated press release, advising consumers to continue to avoid consuming fresh spinach or products containing fresh spinach. They added, however, that it was safe to eat frozen spinach, canned spinach, and spinach included in premade meals manufactured by food companies. In addition, DNA fingerprinting had linked a sample from a package of Dole Baby Spinach with the outbreak strain of E. coli O157:H7 (U.S. Food and Drug Administration 2006c). The following day, the FDA issued a statement that they were working closely with the CDC and the State of California and had determined that the spinach implicated in the outbreak had been grown in Monterey, San Benito, and Santa Clara counties in California. The FDA stated that produce other than spinach grown in these counties had not been implicated in the outbreak; however, they were equally clear that the advisory against eating spinach was still in effect (U.S. Food and Drug Administration 2006d). On September 22, the FDA issued a statement that, “The public can be confident that spinach grown in the nonimplicated areas can be consumed.” They added that “industry is working to get spinach from areas not implicated in the current E. coli O157:H7 outbreak back on the market” (U.S. Food and Drug Administration 2006e), suggesting that the incident was over. At a press conference a little more than a week later, on September 30, FDA official David Acheson attempted to provide additional closure to the event by saying, “Based on where we are at this point in the investigation, spinach is as safe as it was before this event” (Shin 2006). However, as late as October 6, the FDA continued to proceed with caution, issuing press releases reminding retailers, food-service operators and consumers that they should not sell or consume raw spinach or blends that might contain spinach that was “the subject of the earlier recalls” (U.S. Food and Drug Administration 2006f). In total, 205 people in 26 states were reported to the CDC as having been infected with the outbreak strain of E. coli O157:H7 (U.S. Food and Drug Administration 2007b). As a result of eating the contaminated spinach, more than 100 people were hospitalized, 31 developed hemolytic uremic syndrome, which can result in kidney failure, and 3 people died (U.S. Food and Drug Administration 2006f). There remains considerable ambiguity concerning the vector responsible for the presence of the E. coli on the contaminated spinach. Investigators were able to identify 13 bags of baby spinach manufactured by Natural Selections Foods for the Dole™ Brand. Product codes were found for 11 of these bags, all bearing the product codes beginning with “P227A,” indicating that they had been packed on August 15, 2006 (California Food Emergency Response Team 2007). A genetic match for the particular strain of E. coli O157:H7 responsible for making some people sick was found in samples taken from a stream and from feces of cattle and wild pigs present on ranches implicated in the outbreak (Brackett 2006). Investigators were also able to match environmental samples of E. coli O157:H7 from one specific field to the strain that had caused the outbreak (California Food Emergency Response Team 2007). However, in releasing its final report on the 2006 outbreak, the FDA concluded “the precise means by which the bacteria spread to the spinach remains unknown” (U.S. Food and Drug Administration 2007b).

Public Response to the 2006 Recall of Contaminated Spinach

353

Due to the nature, scope, and significance of this contamination incident, and the potential lessons that might be learned from it, the Food Policy Institute (FPI) of the New Jersey Agricultural Experiment Station at Rutgers, the State University of New Jersey undertook an analysis of the information that key actors attempted to deliver as events unfolded; the media coverage of those messages and events; and the information that consumers received, remembered, and acted upon. This report focuses on the third portion of this analysis; that is, what did consumers know, where did they get that information, and what did they do (and continue to do) in response to the advisories issued by the FDA warning them not to eat fresh spinach?

Methods Sample To explore these questions, a nationally representative sample of 1,200 Americans from all 50 states was interviewed by telephone during November 8–29, 2006. Computer Assisted Telephone Interviews (CATI) were conducted with noninstitutionalized adults aged 18 or over. Proportional random digit dialing was used to select survey participant households, and the CATI system was programmed to provide prompts to select the appropriate proportions of male and female participants. Working nonbusiness numbers were called a minimum of 12 times distributed over days of the week and time of day in attempts to obtain interviews. The cooperation rate was 48%, with a resulting sampling error of ±2.8%. The resulting data were weighted by gender, age, race, ethnicity, and education to approximate U.S. Census figures. Survey Instrument While the initial response of the FDA was to issue an “alert” to consumers advising them not to eat fresh spinach, for convenience in referring to the period of time and the events associated with the contamination of fresh spinach with E. coli O157:H7 and the subsequent foodborne illness outbreak, they used the term “spinach recall” in the survey instrument, and we adopted the same convention for this chapter. This terminology is consistent with that used in much of the media coverage that occurred during the period of interest. Before fielding the survey, a search of the coverage of nine newspapers between September 15 and September 22, 2006, revealed that the word “recall” was used 107 times in conjunction with “spinach,” while “advisory” was only used 30 times (Cuite 2008). As the results below suggest, the term “recall” was familiar to most of our respondents. To help prevent response biases, the flow of the questionnaire and specific questions were tailored depending on whether respondents had heard about the spinach recall. For example, respondents who had heard about the spinach recall were asked. “Did you eat spinach before the recall?” and consumers who were unaware of the recall were simply asked, “Do you eat spinach?” Similarly, some questions that referred specifically to knowledge or behaviors during the recall were not asked of respondents who had not heard of the recall or who did not eat spinach before the recall.

354

Section V. Public, Legal, and Economic Perspectives

Results To more effectively communicate the findings of the survey, the results are described in three sections, corresponding to what Americans said they knew and did before, during, and after the recall. However, it should be noted that the order in which the results are described does not necessarily reflect the order in which specific questions were asked. Before the Recall Data from the U.S. Department of Agriculture (USDA) shows that before the recall, consumption of fresh spinach had been increasing dramatically, due largely to sales of packages of “triple-washed” spinach, one of the fastest growing segments of the packaged salad industry. Although consumption of frozen spinach had been relatively stable from 1980–2005 and consumption of canned spinach had been declining, consumption of fresh spinach increased from less than 0.5 lb/yr per person in 1980 to 2.5 lb/yr in 2005 (U.S. Department of Agriculture, Economic Research Service 2007a). Consistent with this, the survey found that nearly half of Americans (48%) said they ate fresh spinach before the recall. Most also reported that they ate it relatively frequently. More than one-in-five (22%) of those who said they ate fresh spinach reported that they had done so “a few times a week,” 18% “once a week,” 29% “a few times a month,” and 14% reported eating spinach “once a month.” The remainder said they ate fresh spinach “a few times a year” (14%), “less than a few times a year” (2%), or “only a few times in their life” (1%). Thus, most (83%) of those who ate spinach before the recall did so at least once a month, and 40% did so at least once a week. A logistic regression analysis indicates that Americans with more education (odds ratio OR = 1.49; 95% confidence interval CI = [1.33, 1.67]) or higher incomes (OR = 1.11; CI = [1.05, 1.19]) were more likely to report having eaten fresh spinach before the recall than those with less education or lower incomes. However, there were no significant differences in age, race, or gender between those who reported that they ate spinach before the recall and those who did not. During the Recall Awareness and Interest At the time of their interviews, the majority of Americans (87%) reported they had been aware of the spinach recall. More than half (56%) knew that there had been a recent food recall and were able to volunteer that it had been a recall involving spinach. An additional one-third (31%) reported being aware of the spinach recall when asked specifically if they had heard about it. Only 13% of American consumers said they had been unaware of the recall, 19% of those who did not eat spinach before the recall and 9% of prerecall spinach consumers. As shown in Figure 19.1, more than half (52%) reported having heard “a lot” or “a great deal” about the spinach recall and 86% report having heard at least “a little” about it. Those who had been aware of the recall were asked what questions (if any) had come to mind when they had first heard about it. The responses to this open-ended question were then categorized based on content (see Table 19.1). The results show that the majority of the questions raised by the 446 participants who responded focused

Public Response to the 2006 Recall of Contaminated Spinach

355

50 45 40 35 30

26%

26%

25 19% 20 15%

14%

15 10 5 0 A Great Deal

A Lot

Some

A Little

Nothing

Figure 19.1. How much Americans heard or read about the spinach recall. Note: N = 1,200. Those who had heard “nothing” include those who reported being unaware of the recall.

on how the contamination happened, what products had been affected, and when the problem might end. Figure 19.2 shows that the majority of respondents who were aware of the spinach recall reported that they had first learned about it through television broadcasts (71%). One-twelfth (8%) reported that they had first heard about the recall from another person. The remainder said that they first learned about the recall from the radio or through the newspaper or Internet. Although only a small portion of the population reported that they had first heard about the recall from someone else, 84% of the respondents who had been aware of the recall said they had talked with others about it at some point. Nearly one-third (31%) reported they had engaged in conversations about the recall “frequently” or “occasionally”; one-quarter (25%) reported they had discussed the recall “a few times” and 29% said they did so “once or twice.” Thus, only 16% reported having never discussed the spinach recall with someone else. Overall, most of the respondents (59%) indicated that they had been interested in stories about the spinach recall. Moreover, 44% agreed that they had “closely followed news stories about the spinach recall,” 23% agreed that they had “watched the news specifically to hear about the recall,” and 12% reported that they had “searched on the Internet to find more information about the spinach recall.”

356

Section V. Public, Legal, and Economic Perspectives

Table 19.1. Questions Americans asked when they first heard about the spinach recall Topic of Question

% of All Questions

What caused the problem? What caused the contamination? Where did the contamination originate? Other source of contamination questions

39% 26% 12% 1%

What was affected? Was frozen spinach contaminated? Was canned spinach contaminated? Was fresh spinach contaminated? Was packaged spinach contaminated? Was organic spinach contaminated? Which brands/effective dates did it affect? What other foods are affected? Where was the spinach being sold? Was the spinach that I purchased affected? Was the spinach that I had eaten affected? Other product(s) affected questions

38% 3% 2% 1% 2% <1% 10% 2% 9% 5% 3% 1%

Health/safety questions What are the symptoms of illness? Does washing eliminate contamination? Does cooking eliminate contamination? A/O health/safety mentions

8% 3% <1% <1% 4%

When will the problem be over? When will spinach be safe to eat?

5% 5%

Other questions Can this be prevented from happening again? Why didn’t we receive more timely information? Other miscellaneous questions

3% 3% 5%

Note: 586 total responses; 446 respondents gave one or more responses.

As might be expected, significantly more of those who ate spinach before the recall (73%) reported that they had been interested in news stories about the recall than those who did not eat spinach before the recall (44%; χ2 (1, N = 1034) = 25.72, p < 0.001). Similarly, more than half (52%) of those who ate fresh spinach before the recall said that they had closely followed the news stories about the recall; this was true of only one-third (36%) of those who had not eaten spinach (χ2 (1, N = 1019) = 92.30, p < 0.001). In addition, compared to those who were not spinach consumers before the recall, significantly more of those who ate spinach reported that they had searched the Internet to find information about the recall (9% vs. 14%, respectively; χ2 (1, N = 1037) = 7.07, p < 0.01). There were no significant differences in the percentage of people who reported having watched the news to specifically hear about the recall. Knowledge about the Details of the Recall Most Americans reported having been aware of the recall, having heard a fair amount about it, having been interested in news stories about it, and having talked about it

Public Response to the 2006 Recall of Contaminated Spinach

8%

357

3%

4%

TV

9%

Newspaper Radio Internet

5%

Person Elsewhere 71%

Figure 19.2.

Where Americans first heard about the spinach recall. Note: N = 1,027.

Table 19.2. Americans’ knowledge of types of spinach recalled Type of Spinach Recalled: Bagged fresh Loose fresh Not Recalled: Frozen Canned

“True” (was recalled)

“False” (was not recalled)

“Don’t know”

95% 68%

1% 16%

4% 16%

22% 16%

57% 70%

21% 14%

Note: N = 1,029.

with others. However, their knowledge of many of the important details of the recall was significantly incomplete. For example, one of the key messages during the spinach recall was that consumers should not eat any fresh spinach, whether sold loose or in a bag. However, because neither frozen nor canned spinach was suspected of having been contaminated, the FDA made a point in some of its press releases to say that both frozen and canned spinach were safe for consumers to eat. To test the extent to which consumers paid attention to these specific messages, respondents who were aware of the recall were asked a series of true/false questions about whether each of four types of spinach had been recalled (Table 19.2). Impressively, nearly all (95%) of the respondents correctly reported that it was true that “bagged fresh spinach” had been recalled. However, only two-thirds (68%) knew that “loose fresh spinach” had been recalled. By comparison, only 70% knew that

358

Section V. Public, Legal, and Economic Perspectives

canned spinach and 57% knew that frozen spinach had not been recalled. This suggests that there was a considerable amount of confusion about whether spinach products other than those that had been bagged were safe to eat. There also appears to have been some confusion about where the contaminated spinach had been grown. This is important because this information was critical to knowing which fresh spinach products were part of the ongoing recall and which were safe to eat. When asked where the contaminated spinach was grown, only slightly more than half (52%) of the respondents who were aware of the spinach recall correctly reported that the affected spinach had been grown in California; more than four in ten (41%) reported that they didn’t know. About 5% reported that the contaminated spinach had been grown in states other than California, or in both California and other states, and 2% percent provided other responses. Knowledge about E. coli and Symptoms of Infection Every press release from both the FDA and the CDC, as well as most news stories concerning the spinach contamination and the resulting illnesses specifically named E. coli as the contaminant involved. Even with this widespread distribution of information, only about half (52%) of the respondents who said they were aware of the recall were able to correctly volunteer that E. coli was the contaminant that had caused people to become ill. One-third (31%) said that they did not know what contaminant had been involved, and a small number of incorrect responses were volunteered, such as Salmonella (4%), animal waste (4%), or other general sources of exposure such as “a bacteria” (2%) or “a chemical” (1%). Most Americans were also confused about the symptoms associated with infection by E. coli O157:H7 (Table 19.3). Beginning with its September 14, 2006, press release, and in subsequent releases, the FDA described the symptoms of E. coli O157:H7 illness in the same way (U.S. Food and Drug Administration 2006a): E. coli O157:H7 causes diarrhea, often with bloody stools. Although most healthy adults can recover completely within a week, some people can develop a form of kidney failure called Hemolytic Uremic Syndrome (HUS). HUS is most likely to occur in young children and the elderly. The condition can lead to serious kidney damage and even death. In its own Health Alert on September 14, 2006, the CDC specifically noted that “the E. coli O157:H7 bacterium causes diarrhea that is often bloody and accompanied by abdominal cramps, but fever is absent or mild” (U.S. Centers for Disease Control and Prevention 2006a). When asked about a list of symptoms, the majority (87%) of Americans correctly recognized that cramping (or abdominal cramps) is a common symptom of E. coli infection. Although the CDC identifies bloody diarrhea as the distinguishing characteristic of E. coli O157:H7 infections, only about two-thirds (64%) of Americans correctly recognized this as a symptom. Instead, they are more likely to incorrectly associate the symptoms of nausea (88%) and vomiting (87%) with an E. coli O157:H7 infection. Moreover, though fever is not generally associated with E. coli O157:H7 infections, more than three-quarters (77%) of Americans identified fever as a symptom, and nearly one-quarter (22%) indicated that rashes were a symptom of E. coli infections despite the fact that rashes are not commonly associated with any foodborne illness.

Public Response to the 2006 Recall of Contaminated Spinach

359

Table 19.3. Percentage identifying symptoms as resulting from E. coli infection Symptom Nausea Vomiting Cramping Fever Bloody diarrhea Rash

Percentage 88% 87% 87% 77% 64% 22%

Note: Shaded boxes indicate common symptoms identified by the CDC.

1200 Americans Interviewed 48% 575 ate spinach more than a few times before the recall 91%

9%

522 were aware of the recall

70 ate fresh spinach during the recall

52 knew about the recall when they ate fresh spinach

81%

507 were aware of the recall

19%

118 were not aware of the recall

87%

13%

74%

53 were not aware of the recall

625 did not eat spinach more than a few times before the recall

452 did not eat fresh spinach during the recall

26%

18 did not know about the recall when they ate fresh spinach

Figure 19.3. Classification of sample by awareness of recall and eating spinach prior to and during the recall.

Some Americans Ate Fresh Spinach during Recall Of particular concern from a public health perspective is that more than one in eight Americans (13%) who were aware of the recall and who ate spinach prior to it reported eating fresh spinach during the recall. Nearly three-quarters of these (74%) said that they knew about the recall of fresh spinach when they consumed it (see Fig. 19.3). Nearly one-third (30%) of those aware of the recall (and who were spinach consumers before it) reported that they had fresh spinach in their homes when they first

360

Section V. Public, Legal, and Economic Perspectives

learned about the recall. Although more than three-quarters (77%) reported discarding the spinach once they learned about the recall, more than one-quarter (27%) said they consumed some or all of the spinach they had at home. Of these, nearly three-quarters (72%) said they knew about the recall at the time they ate it. Many Thought Washing Produce Would Make It Safe to Eat In their September 15, 2006 update on the outbreak of E. coli O157:H7 infections from fresh spinach, the CDC specifically noted under the heading “Advice for Consumers about this Outbreak,” that “bacteria stick to produce even when it is washed, and sometimes the bacteria are inside the produce.” In the same press release, under the heading “General Advice for Consumers,” the CDC also advised consumers that they should “wash produce with clean cool running water just before eating and cut away damaged areas.” (U.S. Centers for Disease Control and Prevention 2006b). In addition, Robert Brackett, Director of the FDA’s Center for Food Safety and Nutrition (CFSAN), was widely quoted advising consumers to discard any spinach they had already purchased, noting that simply washing the spinach would not make it safe to eat. Perhaps it was because of this apparently contradictory advice or because washing food is so often a recommended action for food safety (U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition 2005) that there was some confusion about the role of washing in eliminating possible contaminants. Nearly two-thirds (64%) correctly recognized that the statement, “Bagged spinach marked as ‘Triple washed’ is certain not to have any E. coli” is untrue. Yet, whether they had heard of the recall or not, 44% of Americans thought it true that properly washing contaminated food makes food safe to eat, and nearly half (48%) reported that the spinach recall caused them to wash their food more thoroughly. Concerns of Spinach Safety Affect Other Produce Americans appear to have generalized their concerns about spinach to other, similar produce. Regardless of whether they ate spinach before the recall or not, nearly onefifth (18%) of those aware of the recall said they stopped buying other bagged produce because of the spinach recall. As a result, these data suggest that the recall affected the sales of spinach and of other produce as well. After the Recall Many Americans Unsure When Recall Was Over By the time the survey was conducted, more than 6 weeks had elapsed since the FDA had issued its September 22nd press release advising consumers that they could be confident in eating spinach grown outside the three counties in California that had been implicated in the E. coli contamination (U.S. Food and Drug Administration 2006e), and nearly a month had passed since David Acheson of the FDA had been quoted as saying that “… spinach is as safe as it was before this event” (Shin 2006). Yet, at the time they were interviewed, almost half of those who were aware of the spinach recall (45%) were not confident that the recall had ended. More than one-tenth (13%) of the respondents thought it was true that “the spinach recall is still in effect” and nearly one-fifth (18%) said they were not sure. About half (55%) said that it was

Public Response to the 2006 Recall of Contaminated Spinach

361

definitely “false” that the spinach recall was still in effect and 14% said that it was “likely false.” Many Think Spinach Is Safer Now Than before the Recall To examine perceptions of the threat of contracting a foodborne illness, respondents were asked to rate the likelihood of becoming infected from eating “uncooked fresh spinach” and “a food other than spinach” using a semantic differential scale where a rating of 0 represented “not at all likely” and a rating of 10 represented “extremely likely.” As represented in Figure 19.4, respondents who were aware of the recall reported their likelihood of infection from eating fresh spinach before the recall to be relatively low (M = 2.86, SD = 2.94). Not surprisingly, respondents reported their likelihood of infection from eating fresh spinach during the recall (M = 5.09, SD = 3.24) to be significantly higher than before the recall (t(954) = − 21.71, p < 0.001). Respondents reported their likelihood of becoming infected at the time of the interview (i.e., after the recall) (M = 1.92, SD = 2.59) as significantly lower on average than during the recall (t(965) = 28.32, p < 0.001). Their estimates of the likelihood of infection resulting from eating fresh spinach at the time of the interview were also significantly lower than their estimates of likelihood of infection prior to the recall (t(953) = 8.88, p < 0.001). In comparison, the participants also estimated their likelihood of infection from eating a food other than spinach at the time of the interview to be both low (M = 2.95, SD = 2.81) and nearly equivalent to their estimates of the likelihood of infection from eating spinach prior to the recall. However, they estimated their current risks of becoming infected as the result of eating spinach as significantly lower than the current risk posed by eating foods other than spinach (t(956) = − 9.64, p < 0.001). Thus, they viewed consumption of spinach after the recall as safer than eating other types of foods. Most Americans Have or Will Eat Fresh Spinach Again At the time of the interview, 44% of those who had heard about the recall and who ate spinach prior to it reported that they had eaten spinach since the recall ended (see

Extremely likely

Not at all likely

Figure 19.4.

10 9 8 7 6 5 4 3 2 1 0 Before

During

Right Now

Other Food Now

Perceived likelihood of getting sick from consuming spinach.

362

Section V. Public, Legal, and Economic Perspectives

Table 19.4. Likelihood of eating spinach as of November, 2006 Have already eaten spinach Definitely will eat spinach in the future Likely to eat spinach in future As likely as not to eat spinach in future Unlikely to eat spinach in future Definitely will not eat spinach in future Don’t know if or when will eat spinach in the future

Percentage 44% 20% 13% 10% 5% 5% 3%

Note: N = 494.

Table 19.4). These respondents reported that it took a relatively short time (approximately 2 weeks) after the recall ended for them to resume eating spinach (M = 14.50 days, SD = 12.01; Mdn = 14.00). Those who had not yet resumed eating spinach at the time of the interview said it would take an average of about 2 months for them to start eating fresh spinach again (M = 56.98 days, SD = 74.81; Mdn = 30.00), with estimates ranging from 1 day to 1 year. Only 5% of those who had heard about the recall and who ate spinach prior to it said that they would never eat fresh spinach again. Demographics Related to Those Eating Spinach after the Recall There were important demographic differences among those people who had already begun eating spinach again as of November and those who had not. As shown in Table 19.5, among those who were aware of the recall and ate spinach prior to it, older people (χ2 (4, N = 502) = 14.99; p < .005) and those with lower incomes (χ2 (3, N = 430) = 20.53; p < .001) were less likely to have eaten spinach since the recall ended. Conversely, whites were more likely to have eaten spinach since the end of the recall (χ2 (2, N = 487) = 8.65; p < .05). Education and gender were not related to eating spinach after the recall. In addition to the demographic predictors, two other variables are strongly related to eating spinach after the recall. Those who ate spinach more frequently prior to the recall were more likely to report having eaten spinach since the recall ended (χ2 (12, N = 525) = 33.90; p < .001), as shown in Table 19.6). Not surprisingly, those Americans who were able to identify the recall as having ended were significantly more likely to report having resumed eating spinach (χ2(2, N = 513) = 39.6, p < .0001).

Discussion and Conclusions The results of the survey show that the main public health goal of the recall was met. Most Americans heard FDA’s message to consumers warning that bagged fresh spinach had been contaminated and should not be eaten. Moreover, the data clearly indicate that the majority of consumers did stop eating spinach during the recall. In part, this was because the recall was effective at the retail level, so there was no fresh spinach available for consumers to purchase or to eat. The survey also shows that the recall was effective at the level of the individual consumer in that most discarded any spinach that they had already purchased.

Table 19.5. Whether respondents have eaten spinach at time of survey by demographic variables As of November, 2006 Have Eaten Spinach

Have Not Eaten Spinach

Age* 18–34 35–44 45–54 55–64 65 & over

44% 46% 55% 44% 27%

56% 54% 45% 56% 73%

Education High school or less Some college 4-year college degree Graduate school

37% 34% 45% 42%

63% 66% 55% 58%

Ethnicity* White Black Other

47% 25% 33%

53% 75% 67%

Income* Under $35,000 $35,001–$50,000 $50,001–$75,000 $75,001 & over

27% 43% 51% 54%

73% 57% 49% 46%

Sex Women Men

43% 45%

58% 55%

Note: Includes only participants who ate spinach prior to the recall and had heard of the recall. * Represents significant differences on that demographic variable between those who have and have not eaten spinach.

Table 19.6. Whether respondents have eaten spinach at time of survey by frequency of eating spinach prior to the recall and awareness that the recall has ended As of November, 2006 Have Eaten Spinach

Have Not Eaten Spinach

Frequency of eating spinach prior* A few times a week Once a week A few times a month Once a month A few times a year Less than a few times a year

53% 55% 49% 34% 20% 0%

47% 45% 51% 66% 80% 100%

Awareness of recall having ended* Believe that recall ended Believe that recall has not ended Don’t know if recall has ended

51% 18% 19%

49% 82% 81%

Note: Includes only participants who ate spinach prior to the recall and had heard of the recall. * Represents significant differences on that variable between those who have and have not eaten spinach.

363

364

Section V. Public, Legal, and Economic Perspectives

Many Americans were unaware of important details related to the recall. Many were confused about the types of spinach products affected, where it had been grown, the organism that caused the contamination, the symptoms of the resulting illness, and, perhaps most significantly, whether or not the recall had ended. These uncertainties are problematic because without knowing the correct information, it would have been difficult for consumers to follow the FDA’s advice or to evaluate what products were safe to eat, when they were safe to eat, and under what conditions. Although the recall was largely successful in achieving its primary aims, the data also suggest that there were unintended consequences as well. For example, most consumers stopped eating spinach as a result of the recall, but the data show that many stopped buying other bagged produce, too. This was reflected in an immediate decline in sales of spinach and in other produce reported by the industry (Schmit 2007). In part, this may have been in reaction to a violation of consumers’ expectations regarding foods such as spinach and lettuce that have typically been promoted as “healthy” foods. In addition, because they are marketed as “triple washed” and “ready to eat” in a raw form, consumers might have reasonably expected that such products require no additional actions to make them safe to eat. Moreover, thoroughly washing produce—the one postpurchase action consumers are often advised to take to improve food safety—was dismissed for its limitation to removing some surface pathogens and deemed ineffective in the case of E. coli contaminated spinach. As a result, consumers may have adopted a strategy of trying to completely eliminate the associated risk by halting their purchases of spinach. They may have also avoided purchasing other produce (like bagged lettuce), fearing that whatever problems caused the contamination in spinach might also be occurring in similar products or might do so in the future. The recall of contaminated spinach in Fall 2006 was unusual. Most food product recalls are limited in scope and are normally issued to recover the products of a single manufacturer or distributor, and they are often restricted to the food manufactured or processed at a single location, to specific lot numbers, and distributed within a circumscribed area (U.S. Government Accountability Office 2004). Thus, the broad nature of the recall, suggesting that all fresh spinach across the country should be considered as potentially contaminated and therefore unsafe to eat, combined with a message that no amount of washing would make it safe, distinguished it from the more routine advisories and notice of recalls typically issued by the FDA. The unusual nature of the spinach recall—which suggested that anyone who ate fresh spinach was vulnerable to becoming ill, that there was little that consumers could do to avoid getting sick other than to stop eating it, and that there were potentially serious consequences of being infected with E. coli O157:H7—likely led to both the broad media coverage it received and to the large number of conversations Americans reported having had about it. The fact that a definitive cause of the contamination still has not been identified, and therefore not clearly remediated, may help to explain the reluctance of some consumers to resume their patterns of eating spinach and other produce grown in the same way or in the same geographic areas as the contaminated spinach. This may have been reinforced by the lack of a clear statement by the government indicating that spinach was now “safe” to eat, giving closure to the recall.

Public Response to the 2006 Recall of Contaminated Spinach

365

Instead, the FDA issued a press release on September 22, 2006, indicating that “the public can be confident that spinach grown in the nonimplicated areas can be consumed” (U.S. Food and Drug Administration 2006e). Unfortunately, this statement, intended to let consumers know that they could resume eating spinach, generated much less press coverage than the original press releases warning consumers that they should not eat any fresh spinach (Nucci 2008; Pleasant 2008). Whether due to a lack of a definitive statement, lack of press coverage, or lack of attention by consumers, it is clear that many Americans did not get or believe the message that spinach was now safe to eat. As of November 2006, nearly half of those who had heard about the spinach recall were not completely confident that it had ended. In addition, only a little more than half thought it definitely true that authorities had declared that fresh spinach available in supermarkets was “safe to eat.” A year later, spinach sales had yet to recover to their prerecall levels. According to data from the Perishables group, spinach sales in July of 2007 were almost $9,000,000 lower than in August of 2006 (Weise and Schmit 2007). This is consistent with data from the USDA’s Economic Research Service that showed declines in both the supply and consumption of fresh spinach during 2007. In 2005, slightly more than 758 million pounds of fresh market spinach were produced domestically and nearly 28 million pounds were imported. In 2007, domestic production was estimated to be only 680 million pounds, with accompanying imports declining to 20 million pounds. Similarly, per capita use of fresh spinach fell from 2.5 pounds in 2005 to an estimated 2.2 pounds in 2007, a decline of 12% (U.S. Department of Agriculture, Economic Research Service 2007a). These declines in the market for fresh spinach are consistent with the results of our survey wherein 5% of consumers who had heard about the recall and who ate spinach prior to it said they would not eat spinach again, 5% said they were unlikely to eat spinach again, and 3% said they didn’t know if, or when, they would do so. We consider our data likely to underestimate the full effect of the recall on produce sales. The ambiguity regarding the end of the recall and lack of closure to the incident may explain why many respondents said that they would wait an average of 2 months before they would resume their consumption of spinach. In part, this waiting period would likely be used by consumers to make sure that the contamination problem was truly over. All of the respondents to the survey were interviewed by November 29, 2006. Soon after, on December 6, 2006, the FDA announced that it was investigating E. coli O157:H7 infections associated with multiple Taco Bell restaurants in four states (U.S. Food and Drug Administration 2006g). This outbreak sickened 71 people, resulting in the hospitalization of 53, and in 8 cases of HUS by the time it was considered over on December 14, 2006 (U.S. Food and Drug Administration 2006i). Green onions contaminated with E. coli were originally suspected as the cause of the outbreak, and were voluntarily recalled from Taco Bell restaurants (U.S. Food and Drug Administration 2006h); however, the FDA subsequently identified shredded iceberg lettuce served at the restaurants as the source of the contamination (U.S. Food and Drug Administration 2006i). That E. coli O157:H7 infections associated with

366

Section V. Public, Legal, and Economic Perspectives

Taco Bell restaurants occurred during this “wait and see” period likely reinforced some consumers’ beliefs that contamination problems involving produce had not yet been resolved. The result of two serious, widely publicized E. coli contamination incidents occurring in rapid succession likely heightened consumer awareness and amplified consumer concerns about the safety of eating fresh produce. These were quickly followed by a series of other high-profile recalls and incidents involving food safety and foodborne illnesses. These included the recall of peanut butter contaminated with Salmonella (U.S. Food and Drug Administration 2007a) and several recalls of ground beef contaminated with E. coli O157:H7 (U.S. Department of Agriculture Food Safety and Inspection Service 2007b), including what would become one of the largest recalls of ground beef in history. These were followed by the recall of Salmonellacontaminated vegetable snacks (U.S. Food and Drug Administration 2007e), canned hot dog chili contaminated with botulinum toxin (U.S. Food and Drug Administration 2007g), canned cut green beans contaminated with botulinum toxin (U.S. Food and Drug Administration 2007h), and chicken and turkey pot pies contaminated with Salmonella (U.S. Department of Agriculture Food Safety and Inspection Service 2007c). These recalls of domestically produced foods were interspersed with a series of well-publicized problems and concerns with imported food and other products from China. These included the deaths of cats and dogs as the result of eating melaminecontaminated pet food (U.S. Food and Drug Administration 2007c), the discovery that toothpaste imported from China had been contaminated by diethylene glycol (U.S. Food and Drug Administration 2007d), and the detention of farm-raised seafood contaminated by chemical residues (U.S. Food and Drug Administration 2007f). Consequently, the recall of E. coli O157:H7–contaminated spinach in the autumn of 2006 was the first in a series of food safety problems that received national attention. It served as a “signal event,” focusing attention initially on potential contamination problems in the produce industry. Subsequent recalls raised and reinforced broader questions about the safety of the American food system. As a result, the contamination of spinach in Fall 2006 had several measurable impacts. It caused the deaths of 3 people and made more than 200 others ill. It significantly reduced the market for, and consumption of, fresh spinach, resulting in economic losses for the produce and retail industries. It also had an impact not just on how Americans perceived the wholesomeness of eating raw spinach, but on how they viewed the safety of the entire food supply. What is remarkable, is that these consequences were the result of contamination originating from a single field of spinach.

Acknowledgment The research described herein is based on Rutgers Food Policy Institute (FPI) working paper, RR-0107-013 by C. L. Cuite, S. C. Condry, M. L. Nucci, and W. K. Hallman (2007), which was supported by a grant provided to Rutgers FPI by the Cooperative State Research, Education, and Extension Service (CSREES) of the United States Department of Agriculture (USDA) under the National Integrated Food Safety Initiative (NIFSI) grant #2005-51110-02335 “Food Biosecurity: Modeling the Health,

Public Response to the 2006 Recall of Contaminated Spinach

367

Economic Social, and Psychological Consequences of Intentional and Unintentional Food Contamination,” Dr. William K. Hallman, Principal Investigator. The opinions expressed in the article are those of the authors and do not necessarily reflect official positions or policies of the USDA; the Food Policy Institute; the New Jersey Agricultural Experiment Station; Rutgers, the State University of New Jersey; or any other state or federal government agency.

References Brackett RE. 2006. Statement of Robert E. Brackett, Ph.D. Director, Center for Food Safety and Applied Nutrition, Food and Drug Administration, before the Committee on Health, Education, Labor and Pensions, United States Senate. (November 15, 2006). Available at http://www.fda.gov/ola/2006/ foodsafety1115.html, accessed Oct 6, 2008. California Food Emergency Response Team. 2007. Investigation of an Escherichia coli O157:H7 Outbreak Associated with Dole Pre-packaged Spinach. Available at http://www.dhs.ca.gov/fdb/local/PDF/2006% 20Spinach%20Report%20Final%20redacted.PDF, accessed Apr 10, 2008. Calvin L. 2007. Outbreak Linked to Spinach Forces Reassessment of Food Safety Practices. Amber Waves 5(3):24–31. Available at http://www.ers.usda.gov/AmberWaves/June07/Features/Spinach.htm, accessed Apr 10, 2008. Cuite CL. 2008. Americans’ Responses to the Spinach Recall of 2006: Part of Examining the 2006 Spinach Crisis from Multiple Perspectives. Presented to U.S. Food and Drug Administration, Center for Food Safety and Nutrition, College Park, MD. Available at http://foodpolicyinstitute.rutgers.edu/news/docs/ FPI_2006_Spinach_Crisis.pdf, accessed Oct 5, 2008. Cuite CL, Condry SC, Nucci ML, Hallman WK. 2007. Public Response to the Contaminated Spinach Recall of 2006. (Publication number RR-0107-013). New Brunswick, New Jersey: Rutgers, the State University of New Jersey, Food Policy Institute. Available at http://www.foodpolicyinstitute.org/publications/ default.asp?id=114, accessed Oct 5, 2008. Nucci ML. 2008. Communicating Food Safety: Broadcast News Coverage of the Spinach Recall of 2006. Part of Examining the 2006 Spinach Crisis from Multiple Perspectives. Presented to U.S. Food and Drug Administration, Center for Food Safety and Nutrition, College Park, MD. Available at http:// foodpolicyinstitute.rutgers.edu/news/docs/FPI_2006_Spinach_Crisis.pdf, accessed Oct 5, 2008. Pleasant AF. 2008. Scared Off Spinach? An Analysis of Selected Print Media Coverage of the Spinach/ E. coli Incident in the United States, 2006. Presented to U S Food and Drug Administration, Center for Food Safety and Nutrition, College Park, MD. Available at http://foodpolicyinstitute.rutgers.edu/news/ docs/FPI_2006_Spinach_Crisis.pdf, accessed Oct 5, 2008. Schmit J. 2007. E. coli’s long gone, but spinach sales are still hurting. USA Today Money 1B (January 30). Shin A. 2006. Fresh spinach declared safe to eat. Washington Post D01 (September 30). U.S. Centers for Disease Control and Prevention. 2006a, September 14. Multiple States Investigating a Large Outbreak of E. coli O157:H7 Infections. [Press Release]. Available at http://www2a.cdc.gov/ HAN/ArchiveSys/ViewMsgV.asp?AlertNum=00249, accessed Oct 5, 2008. ———. 2006b, September 15. Update on Multi-State Outbreak of E. coli O157:H7 Infections from Fresh Spinach. [Press Release]. Available at http://www.cdc.gov/ecoli/2006/september/updates/091506.htm, accessed Oct 5, 2008. ———. 2006c, October 6. Update on Multi-State Outbreak of E. coli O157:H7 Infections from Fresh Spinach. [Press Release]. Available at http://www.cdc.gov/ecoli/2006/september/updates/100606.htm, accessed Oct 5, 2008. U.S. Department of Agriculture, Economic Research Service. 2007a. Vegetable and Melons Yearbook. Available at http://www.ers.usda.gov/Publications/VGS/2007/07JulYearbook/VGS2007.pdf, accessed Oct 5, 2008. U.S. Department of Agriculture, Food Safety and Inspection Service. 2007b, June 3. California Firm Recalls Ground Beef For Possible E. coli O157:H7 Contamination. [Press Release]. Available at http://www. fsis.usda.gov/News_&_Events/Recall_025_2007_Release/index.asp, accessed Oct 6, 2008.

368

Section V. Public, Legal, and Economic Perspectives

———. 2007c, October 9. FSIS Issues Public Health Alert For Frozen Chicken and Turkey Pot Pies. [Press Release]. Available at http://www.fsis.usda.gov/news/NR_100907_01/index.asp, accessed Oct 6, 2008. U.S. Food and Drug Administration. 2006a, September 14. FDA Warning on Serious Foodborne E.coli O157:H7 Outbreak: One Death and Multiple Hospitalizations in Several States. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/spinach.html, accessed Feb 1, 2007. ———. 2006b, September 15. FDA Statement on Foodborne E. coli O157:H7 Outbreak in Spinach. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/spinach.html, accessed Feb 1, 2007. ———. 2006c, September 20. FDA Statement on Foodborne E. coli O157:H7 Outbreak in Spinach. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/spinach.html, accessed Feb 1, 2007. ———. 2006d, September 21. FDA Statement on Foodborne E. coli O157:H7 Outbreak in Spinach. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/spinach.html, accessed Feb 1, 2007. ———. 2006e, September 22. FDA Statement on Foodborne E. coli O157:H7 Outbreak in Spinach. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/spinach.html, accessed Feb 1, 2007. ———. 2006f, October 6. FDA Statement on Foodborne E. coli O157:H7 Outbreak in Spinach. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/spinach.html, accessed Feb 1, 2007. ———. 2006g, December 6. FDA Investigating E. coli O157 Infections Associated with Taco Bell Restaurants in Northeast. [Press Release]. Available at http://www.fda.gov/bbs/topics/NEWS/2006/ NEW01517.html, accessed Oct 5, 2008. ———. 2006h, December 8. UPDATE: FDA Investigates E. Coli O157 Cases Associated with Taco Bell Restaurants. [Press Release]. Available at http://www.fda.gov/bbs/topics/NEWS/2006/NEW01518.html, accessed Oct 5, 2008. ———. 2006i, December 14. UPDATE: E. coli O157:H7 Outbreak at Taco Bell Restaurants Likely Over; FDA Traceback Investigation Continues. [Press Release]. Available at http://www.fda.gov/bbs/topics/ NEWS/2006/NEW01527.html, accessed Oct 5, 2008. ———. 2007a, February 14. FDA Warns Consumers Not to Eat Certain Jars of Peter Pan Peanut Butter and Great Value Peanut Butter. [Press Release]. Available at http://www.fda.gov/bbs/topics/NEWS/2007/ NEW01563.html, accessed Oct 5, 2008. ———. 2007b, March 23. FDA Finalizes Report on 2006 Spinach Outbreak. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/spinach.html, accessed Feb 1, 2007. ———. 2007c, April 5. FDA Update and Synopsis on the Pet Food Outbreak. [Press Release]. Available at http://www.fda.gov/oc/opacom/hottopics/petfood.html, accessed Oct 1, 2008. ———. 2007d, June 1. FDA Advises Consumers to Avoid Toothpaste From China Containing Harmful Chemical. [Press Release]. Available at http://www.fda.gov/bbs/topics/NEWS/2007/NEW01646.html, accessed Oct 1, 2008. ———. 2007e, June 28a. Veggie Booty Snack Food Identified in Product Recall. [Press Release]. Available at http://www.fda.gov/oc/po/firmrecalls/roberts206_07.html, accessed Oct 6, 2008. ———. 2007f, June 28b. FDA Detains Imports of Farm-Raised Chinese Seafood. [Press Release]. Available at http://www.fda.gov/bbs/topics/NEWS/2007/NEW01660.html, accessed Oct 6, 2008. ———. 2007g, July 18. FDA Warns Consumers about Risk of Botulism Poisoning from Hot Dog Chili Sauce Marketed Under a Variety of Brand Names. [Press Release]. Available at http://www.fda.gov/bbs/ topics/NEWS/2007/NEW01669.html, accessed Oct 6, 2008. ———. 2007h, August 2. Lakeside Foods Issues Nationwide Recall of French Style Green Beans. [Press Release]. Available at http://www.fda.gov/bbs/topics/NEWS/2007/NEW01676.html, accessed Oct 6, 2008. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. 2005. Preparation tips for fresh produce. Available at http://www.cfsan.fda.gov/∼dms/prodsafe.html#prep, accessed Oct 7, 2008. U.S. Government Accountability Office. 2004. Food Safety: USDA and FDA Need to Better Ensure Prompt and Complete Recalls of Potentially Unsafe Food. GAO-05-511-83. Available at http://www.gao.gov/ new.items/d0551.pdf, accessed Oct 7, 2008. Weise E, Schmit J. 2007. Spinach recall: 5 faces. 5 agonizing deaths. 1 year later. USA Today Money, 1B (September 21).

20 Produce in Public: Spinach, Safety, and Public Policy Douglas A. Powell, Casey J. Jacob, and Benjamin Chapman

Introduction In October 1996, a 16-month-old Denver girl drank Smoothie juice manufactured by Odwalla Inc. of Half Moon Bay, California. She died several weeks later. Sixty-four others became ill in several western U.S. states and British Columbia after drinking the same juices, which contained unpasteurized apple cider contaminated with Escherichia coli O157:H7. Investigators believed that some of the apples used to make the cider might have been insufficiently washed after falling to the ground and coming into contact with deer feces (Leiss and Powell 1997). Almost 10 years later, on Sept. 14, 2006, the U.S. Food and Drug Administration (FDA) announced that an outbreak of E. coli O157:H7 had killed a 77-year-old woman and sickened 49 others (U.S. Food and Drug Administration 2006a). The FDA learned from the Centers for Disease Control (CDC) and Wisconsin health officials that the outbreak may have been linked to the consumption of produce, and bagged fresh spinach was identified as a possible cause (Bridges 2006a). In the decade between these two watershed outbreaks, almost 500 outbreaks of foodborne illness involving fresh produce have been documented, publicized, and led to some changes within the industry. Yet note what author Malcolm Gladwell (2000) would call a tipping point (“a point at which a slow gradual change becomes irreversible and then proceeds with gathering pace”). Public awareness about produce-associated risks did not reach a tipping point until the spinach E. coli O157:H7 outbreak in the fall of 2006. At what point did sufficient evidence exist to compel the fresh produce industry to embrace the kind of change that the sector has heralded since 2007? And at what point will future evidence be deemed sufficient to initiate future changes in any industry?

Research on North American Outbreaks Fresh fruits and vegetables were identified as the source of several outbreaks of foodborne illness in the early 1990s, particularly leafy greens (Table 20.1). Although poor employee hygiene was responsible for over 40% of sourceidentified produce-related outbreaks (Bean and Griffin 1990), produce can also be contaminated due to internalization of pathogens both through the root system and flesh or stem scars. Evidence of infiltration of bacteria into vegetables has been reported (Bartz 1982; Bartz and Showalter 1981; Burnett and others 2000; Seo and Frank 1999; Zhuang and others 1995), and substantial evidence exists to conclude that pathogens can become incorporated into fresh produce. Previous research, for example, suggests 369

370

Section V. Public, Legal, and Economic Perspectives

Table 20.1. Outbreaks of foodborne illness related to leafy greens, 1992–1996 Date Apr-92 Jan-93 Jul-93 Aug-93 Jul-93 Sep-94 Jul-95 Sep-95 Sep-95 Oct-95 May-96 Jun-96

Product Lettuce Lettuce Lettuce Salad Salad Salad Lettuce Lettuce Salad Lettuce Lettuce Lettuce

Pathogen Salmonella enteriditis Salmonella Heidelberg Norovirus E. coli O157:H7 E. coli O157:H7 E. coli O157:H7 E. coli O153:H48 E. coli O153:H47 E. coli O157:H7 E. coli O153:H46 E. coli O157:H10 E. coli O153:H49

Cases 12 18 285 53 10 26 74 30 20 11 61 7

Setting/Dish Salad Restaurant Restaurant Salad bar Unknown School Lettuce Scout camp Salad Salad Mesclun mix Mesclun mix

State VT MN IL WA WA TX MT ME ID OH CT/IL NY

that pathogens can enter lettuce plants through its roots and end up in the edible leaves. Small gaps in growing roots through which plant pathogens infect tissue may also allow E. coli entry (Solomon and others 2002b; Warriner and others 2003a; Warriner and others 2003b). This research has been well known and publicized in mass media. The 1993 outbreak of E. coli O157:H7 associated with undercooked hamburgers at the Jack-in-the-Box fast-food chain propelled microbial food safety to the forefront of public awareness, at least in the U.S. (Leiss and Powell 1997). In 1996, following extensive public and political discussions about microbial food safety in meat, the focus shifted to fresh fruits and vegetables following an outbreak of Cyclospora cayetanesis linked to Guatemalan raspberries that sickened 1,465 in 21 U.S. states and two Canadian provinces (CDC 1997). That same year, Beuchat (1996) published a review on pathogenic microorganisms in fresh fruits and vegetables and identified numerous pathways of contamination. By 1997, researchers at CDC were stating that pathogens could contaminate at any point along the fresh produce food chain, at the farm, processing plant, transportation vehicle, retail store or foodservice operation, and the home, and that by understanding where potential problems existed, it was possible to develop strategies to reduce the risks of contamination (Tauxe and others 1997). Researchers also reported that the use of pathogen-free water for washing would minimize risks of contamination (Suslow 1997; Beuchat 1998). Beuchat and Ryu (1997) reported in a review that sources of pathogenic microorganisms for produce included the following: Preharvest • • • • • • •

Feces Soil Irrigation water Water used to apply fungicides, insecticides Green or inadequately composted manure Air (dust) Wild and domestic animals (including fowl and reptiles)

Produce in Public: Spinach, Safety, and Public Policy

371

• Insects • Human handling Postharvest • • • • • • • • • • • • • • •

Feces Human handling (workers, consumers) Harvesting equipment Transport containers (field to packing shed) Wild and domestic animals (including fowl and reptiles) Insects Air (dust) Wash and rinse water Sorting, packing, cutting, and further processing equipment Ice Transport vehicles Improper storage (temperature, physical environment) Improper packaging (including new packaging technologies) Cross-contamination (other foods in storage, preparation, and display areas) Improper display temperature

In response to outbreaks and academic publications, the International Fresh-cut Produce Association developed and published industry guidelines on produce food safety (International Fresh-cut Produce Association 1997). In 1998, 2,288 became ill associated with 32 separate fresh produce outbreaks. In January 1998, the New York Times ran a front-page article, along with several additional articles, highlighting produce safety and noting the importance of on-farm strategies to reduce risk. Beuchat (1998) noted that risk management strategies for fresh produce were difficult because potential pathogen sources in fresh fruit and vegetable production were numerous, and irrigation water containing raw sewage or improperly treated effluents from sewage treatment plants may contain hepatitis A, Norwalk viruses, or enteroviruses in addition to bacterial pathogens such as E.coli O157:H7, Salmonella spp., and Shigella spp. In 1999, several more outbreaks of Shiga-toxin producing E. coli (STEC) were linked to leafy greens (Table 20.2), and the U.S. group, the United Fresh Fruit and Vegetable Association, developed and published food safety guidelines based on Hazard Analysis and Critical Control Point (HACCP) concepts for the industry (United Fresh Fruit and Vegetable Association 1999). Rafferty and others (2000) demonstrated that E. coli could spread in plant production cuttings from one contaminated source, and magnify an outbreak to a whole farm. A 2001 outbreak of Shigella flexneri (886 ill) in tomatoes (Reller and others 2006) further focused public and scientific attention on fresh produce. Solomon and others (2002a) discovered that the transmission of E. coli O157:H7 to lettuce was possible through both spray and drip irrigation. They also found that the pathogen persisted on the plants for 20 days following inoculation and submerging the lettuce in a solution of 200 ppm chlorine did not eliminate all viable E.coli

372

Section V. Public, Legal, and Economic Perspectives

Table 20.2. 1999 U.S. outbreaks of Shiga-toxin producing E. coli linked to leafy greens Date Feb-99 Jun-99 Sep-99 Oct-99 Oct-99 Oct-99

Product Lettuce Salad Lettuce Lettuce Lettuce Salad

Pathogen E. coli O157:H9 E. coli O111:H8 E. coli O157:H11 E. coli O157:H7 E. coli O157:H7 E. coli O157:H7

Cases 65 58 6 40 47 5

Setting/Dish Restaurant Texas camp Iceberg Nursing home Restaurant Restaurant

State NE TX WA PA OH OR

Table 20.3. Leafy green outbreaks of Shiga-toxin producing E. coli, 2000–2002 Date Oct-00 Nov-01 Jul-02 Nov-02 Dec-02

Product Salad Lettuce Lettuce Lettuce Lettuce

Pathogen E. coli O157:H7 E. coli O157:H7 E. coli O157:H8 E. coli O157:H7 E. coli O157:H7

Cases 6 20 55 13 3

Setting Deli Restaurant Bagged tossed Restaurant Restaurant

State IN TX WA IL MN

Table 20.4. Leafy green Shiga-toxin producing E. coli outbreaks, 2003–2005 Date Sep-03 Nov-03 Nov-04 Sep-05

Product Lettuce Spinach Lettuce Lettuce

Pathogen E. coli O157:H7 E. coli O157:H7 E. coli O157:H7 E. coli O157:H7

Cases 51 16 6 11

Setting Restaurant Nursing Home Restaurant Dole bagged

State CA CA NJ multiple

O157:H7 cells, suggesting that irrigation water of unknown microbial quality should be avoided in lettuce production (Solomon and others 2002a). In a follow-up experiment, Solomon and others (2002b) explored the transmission of E. coli O157:H7 from manure-contaminated soil and irrigation water to lettuce plants. The researchers recovered viable cells from the inner tissues of the lettuce plants and found that the cells migrated to internal locations in plant tissue and were thus protected from the action of sanitizing agents. These experiments demonstrated that E. coli O157:H7 could enter the lettuce plant through the root system and migrate throughout the edible portion of the plant (Solomon and others 2002b). Such results were widely reported in the general media. During this time, several outbreaks of E. coli were again linked to lettuce and salad (Table 20.3). In 2003, according to Mexican growers, the market impact of an outbreak of hepatitis A traced to exported green onions lasted up to 4 months while prices fell 72% (Calvin and others 2004). Roma tomatoes were identified as the source of a salmonellosis outbreak that resulted in over 560 cases in both Canada and the U.S. (CDC 2005). During 2003–2005, several additional outbreaks of E. coli O157:H7 were linked to fresh leafy greens, including one multistate outbreak involving Dole bagged lettuce (Table 20.4).

Produce in Public: Spinach, Safety, and Public Policy

373

During 2005–2006, four large multistate outbreaks of Salmonella infections associated with eating raw tomatoes at restaurants occurred in the U.S., resulting in 459 culture-confirmed cases of salmonellosis in 21 states. Investigations determined that the tomatoes had been supplied to restaurants either whole or precut from tomato fields in Florida, Ohio, and Virginia (CDC 2007). Allwood and colleagues (2004) examined 40 items of fresh produce taken from a retail setting in the U.S. and that had been preprocessed (including cut, shredded, chopped, or peeled) at or before the point of purchase. They found fecal contamination indicators (E. coli, F-specific coliphages, and noroviruses) were present in 48% of samples. Researchers in Minnesota conducted a small-scale comparative study of organic versus conventionally grown produce. They found that although all samples were virtually free of pathogens, E. coli was 19 times more prevalent on produce acquired from the organic farms (Mukherjee and others 2004). They estimated that this was due to the common use of manure aged for less than a year. The use of cattle manure was found to be of higher risk because E. coli was found 2.4 times more often on farms using it rather than other animal manures (Mukherjee and others 2004).

Industry Efforts/Regulation The current state of risk-based food safety systems suggests that food producers follow HACCP-based programs, employing the strategy to its limits where applicable (U.S. National Advisory Committee on Microbiological Criteria for Foods 1999). There are a variety of generic and specific guidelines for safe fresh fruit and vegetable production in North America. These programs are generally based on HACCP and many are also based on the U.S. FDA’s Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables published in 1998. It has been suggested that produce farmers follow risk-based guidelines. The U.S. National Advisory Committee on Microbiological Criteria for Foods (1999) suggests that that although HACCP should be used, not enough is known about the vectors of contamination. Programs need to be flexible, but still based on what is known. They also suggest that a formal HACCP system is too rigid for the farm, but the principles can still be applied to reduce risk. Translating HACCP-based strategies to the farm has resulted in a set of generic guidelines described as good agricultural practices (GAPs) and include the following (U.S. Food and Drug Administration 1998; Beuchat and Ryu 1997; Powell and others 2002; Luedtke and others 2003): • • • • • • • • •

Equipment maintenance program Sanitation program within facilities/packing areas End-of-season cleaning Washroom facilities Employee training Pest control program Storage maintenance program Transportation program Microbiological sampling

374

Section V. Public, Legal, and Economic Perspectives

In response to increasing outbreak investigation results pointing to lettuce and tomatoes as sources of foodborne illness, Dr. Robert Brackett of U.S. FDA/CFSAN issued a letter in February 2004 to the entire lettuce and tomato supply chains expressing “serious concern” and expectations of the industry to enhance the safety of their products (U.S. Food and Drug Administration 2004), stating: Because fresh vegetables such as lettuce and tomatoes are commonly consumed in their raw state without processing to reduce or eliminate pathogens, the manner in which they are grown, harvested, sorted, packed, and distributed is crucial to ensuring that the potential for microbial contamination is minimized, thereby reducing the risk of illness to consumers. In 1998, the FDA issued a Guide to Minimize Microbial Food Safety Hazards for Fruits and Vegetables (available at http://www.foodsafety.gov/∼dms/ prodguid.html), which discusses recommended good agricultural practices (GAPs) and good manufacturing practices (GMPs) that growers, packers and shippers can undertake to address common risk factors in their operations and thereby, minimize food safety hazards potentially associated with fresh produce. We have worked in partnership with your industries in the U.S. and abroad since that time to promote our recommendations and to advance the scientific knowledge applicable to enhancing the safety of fresh fruits and vegetables. In addition, in 2001, FDA made available a report prepared for the agency under contract by the Institute for Food Technologists, “Analysis and Evaluation of Preventive Control Measures for the Control and Reduction/Elimination of Microbial Hazards on Fresh and Fresh-Cut Produce,” available at http://www. cfsan.fda.gov/∼comm/ift3-toc.html. This report summarizes the current scientific research relating to the various methods of eliminating or reducing pathogens, while maintaining fresh attributes, on whole and fresh-cut produce. In view of continuing outbreaks associated with fresh lettuce and fresh tomatoes, we strongly encourage firms in your industries to review their current operations in light of the agency’s guidance for minimizing microbial food safety hazards in fresh lettuce and fresh tomatoes, as well as other available information regarding pathogen reduction or elimination on fresh produce. We further encourage these firms to consider modifying their operations accordingly, to ensure that they are taking the appropriate measures to provide a safe product to the consumer. Since the available information concerning some of the recent outbreaks does not definitively identify the point of origin of the contamination, we recommend that firms from the farm level through the distribution level undertake these steps.” In Nov. 4, 2005, Dr. Brackett wrote a second letter to California lettuce producers, packers and shippers, urging them to reexamine and modify operations from the farm through to distributors to ensure that consumers were provided with a safe product (U.S. Food and Drug Administration 2005). Dr. Brackett’s November letter noted that the FDA was aware of 18 outbreaks of foodborne illness since 1995 caused by E. coli O157:H7 in which fresh or fresh-cut lettuce was implicated as the outbreak vehicle. In one additional case, fresh-cut spinach was implicated. These 19 outbreaks accounted for 409 reported cases of illness and two deaths:

Produce in Public: Spinach, Safety, and Public Policy

375

In view of continuing outbreaks associated with fresh and fresh-cut lettuce and other leafy greens, particularly from California, we are issuing this second letter to reiterate our concerns and to strongly encourage firms in your industry to review their current operations in light of the agency’s guidance for minimizing microbial food safety hazards in fresh fruits and vegetables, as well as other available information regarding the reduction or elimination of pathogens on fresh produce. We encourage firms to consider modifying their operations accordingly to ensure that they are taking the appropriate measures to provide a safe product to the consumer. We recommend that firms from the farm level through the distribution level undertake these steps. and Foodborne illness investigations rarely pinpoint the point of origin of the contamination. However, claims that “we cannot take action until we know the cause” are unacceptable. We believe that there are actions that can and should be undertaken immediately to address this issue. For example, at least some outbreaks may be related to contamination that may have occurred in the production environment. In June 2004, the California Department of Health Services, Food and Drug Branch (CDHS-FDB) initiated multi-agency, collaborative research aimed at identifying the environmental reservoirs for E. coli O157:H7, and understanding how lettuce may become contaminated. In a preliminary report presented at the August 2005 annual meeting of the International Association for Food Protection, E. coli O157:H7 was isolated from sediment in an irrigation canal bordering a ranch that had been identified in three separate outbreaks. The ranch is bowl-shaped; it sits upon a drained lake and is highly susceptible to localized flooding. Expanded sampling in the Santa Rita Creek and the Salinas Valley area indicate that creeks and rivers in the Salinas watershed are contaminated periodically with E. coli O157:H7. The specific source of contamination that led to the outbreaks was not identified. However, several possible sources of contamination were identified, both on the ranch initially studied and upstream. Although it is unlikely that contamination in all 19 outbreaks was caused by flooding from agricultural water sources, we would like to take this opportunity to clarify that FDA considers ready to eat crops (such as lettuce) that have been in contact with flood waters to be adulterated due to potential exposure to sewage, animal waste, heavy metals, pathogenic microorganisms, or other contaminants. FDA is not aware of any method of reconditioning these crops that will provide a reasonable assurance of safety for human food use or otherwise bring them into compliance with the law. Therefore, FDA recommends that such crops be excluded from the human food supply and disposed of in a manner that ensures they do not contaminate unaffected crops during harvesting, storage, or distribution. Adulterated food may be subject to seizure under the Federal Food, Drug, and Cosmetic Act, and those responsible for its introduction or delivery for introduction into interstate commerce may be enjoined from continuing to do so or prosecuted for having done so. For retail and foodservice establishments, the U.S. FDA 2005 Model Food Code Section 3-302.15 specifies: “Raw fruits and vegetables shall be thoroughly washed in

376

Section V. Public, Legal, and Economic Perspectives

water to remove soil and other contaminants before being cut, combined with other ingredients, cooked, served, or offered for human consumption in ready-to-eat form.” Packaged produce labeled “ready-to-eat,” “washed,” or “triple washed” need not be washed. In April 2006, Commodity Specific Food Safety Guidelines for the Lettuce and Leafy Greens Supply Chain produced by International Fresh-cut Produce Association (2006), the Produce Marketing Association, United Fresh Fruit and Vegetable Growers, and Western Growers were published, as follows: Whatever the preferred production and processing method may be for a single producer, the lettuce/leafy greens industry recognizes the following basic principles that serve as the foundation for all food safety programs found within the industry: • The lettuce/leafy greens industry recognizes that once lettuce/leafy greens are contaminated, removing or killing pathogens is difficult. Therefore, prevention of microbial contamination at all steps from production to distribution is strongly favored over treatments to eliminate contamination after it has occurred. • The lettuce/leafy greens industry supports implementation and documentation of food safety programs that utilize risk assessment techniques that identify significant risks and use a preventive approach to ensure safe food products. • The lettuce/leafy greens industry also supports and encourages routine and regularly scheduled food safety awareness training for all persons who grow, handle, distribute, process, prepare and/or serve lettuce/leafy greens products. • The human pathogens most often associated with produce (Salmonella and E. coli O157:H7) cause infection and illness by the fecal-oral route of food contamination. Therefore, lettuce/leafy greens food safety programs should pay special attention to controlling, reducing and eliminating potential fecal contamination from people and domestic and wild animals through the most likely conduits, that being human hands, water and soil. What was absent in this decade of outbreaks, letters from regulators, and plans from industry associations and media accounts, was verification that farmers and others in the farm-to-fork food safety system were seriously adapting to the messages about risk and the numbers of sick people, and then translating such information into behavioral changes that enhanced front-line food safety practices. The 1996 outbreaks and the 1998 FDA guide did help initiate efforts to implement on-farm food safety programs. The very nature of produce that makes it healthy (fresh and consumed raw) is what makes fresh produce a high-risk food for transmitting microbial contamination. Without the microbiological kill step provided by cooking, produce is vulnerable to contamination from the farm-to-fork. Pathogens can contaminate at any point along the food chain—at the farm, in the packing shed, in the processing plant, in the transportation vehicle, at the retail store or foodservice operation, and in the home. By understanding where potential problems exist, it is possible to develop strategies to reduce risks of contamination (Tauxe and others 1997).

Produce in Public: Spinach, Safety, and Public Policy

377

Consequently, methods of growing, handling, processing, packaging, and distributing fresh produce have received increased attention in terms of identifying and minimizing microbiological hazards. HACCP is a system of food safety control based on a systematic approach to the identification and assessment of hazards associated with food operations and the definition of means for their control. This approach focuses on prevention and control and is advocated at every stage in the food chain, from primary producers through to the final consumer (California Strawberry Commission 2005; International Fresh-cut Produce Association 1997; United Fresh Fruit and Vegetable Association 1999). The produce industry has focused on developing and implementing programs aimed at reducing foodborne disease and illness. Complete HACCP systems cannot be implemented in fresh produce operations, as there is no definite kill step, such as pasteurization. Instead, these HACCP-based systems help to identify and reduce the potential for microbial contamination along the entire production and distribution process. A successful program helps avoid recall campaigns, adverse publicity, and loss of sales, and serves to enhance public health. There is value in applying the steps of HACCP to fruit and vegetable production, using available scientific information as part of the framework, to reduce the risk of foodborne pathogens. All of this was known a decade before the “tipping point” outbreak of E. coli O157:H7 in spinach in the fall of 2006.

Spinach On Sept. 14, 2006, the FDA issued a public statement warning against the consumption of bagged fresh spinach. “Given the severity of this illness and the seriousness of the outbreak,” stated Dr. Robert Brackett, Director of the FDA’s Center for Food Safety and Applied Nutrition (CFSAN), “FDA believes that a warning to consumers is needed.” (U.S. Food and Drug Administration 2006i). That evening, Dr. David Acheson, head of the FDA/CFSAN, told the public, “The FDA is advising consumers not to eat bagged fresh spinach at this time and that any individuals who believe they may have experienced symptoms of illness associated with E. coli contact their health care provider.” (Pal 2006). Dr. Brackett assured the public, “We are working closely with the U.S. Centers for Disease Control and Prevention (CDC) and state and local agencies to determine the cause and scope of the problem.” (Harris 2006). According to the New York Times, when asked whether consumers should also avoid bagged salads, Dr. Acheson answered, “At this point, there is nothing to implicate bagged salad.” (Harris 2006). Bill Marler, a Seattle lawyer specializing in foodborne illnesses, filed a lawsuit that day on behalf of Gwyn Wellborn, who became seriously ill after eating a bag of Dole baby spinach. “We are not pointing at a single source unambiguously,” said Marler. “Dole is one of the companies on the radar.” (Lynn 2006). A spokesman for Washington State’s Department of Health was still saying, “Nobody wants to point fingers yet until they know they are pointing in the right direction.” (Lin 2006). Dr. William E. Keene, a senior epidemiologist for the Oregon Public Health Division, told the Oregon Statesman-Journal, “We’re sure that packaged spinach is the source of the outbreak. What we’re not sure about is the brand.” (Lynn 2006).

378

Section V. Public, Legal, and Economic Perspectives

By Sept. 15, 2006, state and local agencies issued advisories. “We’re telling people if they have bagged produce and they feel like it’s a risk, throw it out,” said T.J. Bucholz, the spokesman of the Michigan Department of Community Health, “If they feel like they have to eat it, wash it first in warm water.” (Pal 2006). The Canadian Food Inspection Agency warned against fresh spinach imported from the U.S., describing the outbreak as sickening nearly 100 people, though no cases had been reported in Canada. “Washing the spinach won’t make it safe,” announced CBC News (2006), “because the E. coli sticks to the leaves tightly.” This statement referred to Dr. Brackett’s message to the Associated Press (AP) that day when he said, “If you wash it, it is not going to get rid of it.” The AP story also stated that federal and state health officials were trying to pinpoint a source of contamination in California, where the spinach was believed to have been grown (Bridges 2006b). In another AP report, Dr. Acheson said health officials did not know of any link to a specific growing region, grower, brand, or supplier. Amy Philpott, a spokeswoman for the United Fresh Produce Association, said, “Our industry is very concerned. We’re taking this very seriously (Bridges 2006a).” David Brown (2006) of the Washington Post speculated, “Crops such as spinach could conceivably be contaminated by liquid fertilizer sprayed on fields,” because of the enteric nature of the bacterium. Natural Selection Foods, LLC (2006), initiated a recall of all their products containing spinach with “Best if Used by Dates” of August 17th through October 1st after consulting with the FDA and the California Department of Health Services. “While neither the FDA nor the CDHS have yet determined the source of the E. coli problem, we believe that recalling all spinach product packed in our facilities is the right thing to do,” said Charles Sweat (2006) in a separate statement. “The FDA has said that they are looking at the entire industry and we will continue to do our part in their investigation.” Tom Stenzel, President and CEO of the United Fresh Produce Association (2006), issued the statement, “The fresh produce industry commends the company for taking this pro-active action to ensure consumer health. While we understand that no definitive evidence has yet linked E. coli O157:H7 to a specific spinach sample, we applaud the company for voluntarily recalling product to ensure the utmost caution in protecting health.” Stenzel also said on behalf of the Association, “We commend the FDA and industry for working together to first protect public health, and then isolate the cause of the outbreak in order to help restore confidence in the overall spinach industry as quickly as possible.” Later in the day, Natural Selection Foods was linked to the outbreak through an epidemiological study wherein victims recalled eating spinach sold by the company under several different brands. “We are very, very upset about this,” said Natural Selection Foods’ spokeswoman Samantha Cabaluna. “What we do is produce food that we want to be healthy and safe for consumers, so this is a tragedy for us.” The company offered refunds for spinach tossed out due to the recall or coupons to buy new packages (Bridges 2006c). Natural Selection Foods LLC halted shipments of all fresh spinach products and said in a statement that it was cooperating with federal and state health officials to identify the source of the contamination (Bridges 2006d).

Produce in Public: Spinach, Safety, and Public Policy

379

Kenter Canyon Farms, Inc. initiated a voluntary recall of repackaged spinach that was not realized by the FDA until recall audits were conducted in October (U.S. Food and Drug Administration 2006g). On Sept. 16, 2006, Dole Food Company, Inc. (2006) announced its support of the recall by Natural Selection Food LLC in a statement on Saturday, though Dole holds no economic interest in the firm. The statement also said, “Dole is committed to assisting the FDA, the California Department of Health Services and other regulatory agencies in their investigations and this recall by Natural Selection Foods.” Earthbound Farm, the packaged brand of Natural Selection Foods produce, was still working with officials to determine the source of the contamination. “We’re not even thinking about the cost right now,” Earthbound spokeswoman Samantha Cabaluna said. “We’re trying to do the right thing, to protect public health and get to the bottom of this.” She also mentioned that it was unclear whether the organic or conventionally grown spinach was the cause of the outbreak (Robertson 2006). By Sunday, Sept. 17, 2006, Earthbound Farm (2006) assured customers that none of their organic spinach had been directly connected with the illnesses at this point in the investigation. This statement came after concern arose that organic farming practices may have lead to contamination. River Ranch, another California company, also recalled its spring mix containing spinach because it had received bulk spring mix containing spinach from Natural Selections (U.S. Food and Drug Administration 2006j). Their spring mix was also sold under Hy-Vee, Fresh N’ Easy, and Farmers Market labels (Norton 2006). The FDA decided to expand the Lettuce Safety Initiative to include spinach. “The primary goals of the initiative are to reduce public health risks by focusing on the product, agents and areas of greatest concern and to alert consumers early and respond rapidly in the event of an outbreak,” said an FDA update on the spinach outbreak (U.S. Food and Drug Administration 2006b). By that time, a total of 109 cases had been linked to the outbreak, including 16 cases of Hemolytic Uremic Syndrome (HUS), and more were continually being reported to the CDC (U.S. Food and Drug Administration 2006b). Columbus, Ohio, news station WBNS reported the death of 23-month-old Olivia Perkins from complications related to E. coli contraction (Black 2006). By Monday, Sept. 18, 2006, Caroline Smith DeWaal, director of food safety for the Center for Science in the Public Interest, was quoted as saying, “We think this incident shows the FDA is suffering from the same weak-kneed approach that they had before they were given more power to regulate beef in the 1990s. … “No one is really in charge of food safety on the farm, and the FDA has come in with fairly weak guidelines there that they can only suggest but not enforce.” (Wood 2006). On Wednesday, Sept. 20, 2006, the New Mexico Department of Health (2006) confirmed the presence of the outbreak strain of E. coli O157:H7 in a bag of spinach belonging to a patient in the investigation using “DNA fingerprinting.” The spinach had been produced and packed conventionally at an Earthbound Farm facility. “This news confirms our decision to go out early with our voluntary recall,” said an Earthbound Farm media release (2006). “Natural Selection Foods is a company built

380

Section V. Public, Legal, and Economic Perspectives

on doing the right thing. In this situation, protecting the public health even before we had much information was the right thing to do.” On Sept. 24, 2006, the Utah Department of Health and the Salt Lake Valley Health Department had positively identified the outbreak strain in a bag of Dole baby spinach purchased in Utah (U.S. Food and Drug Administration 2006c). By Sept. 26, 2008, the FDA stated, “FDA has determined that the spinach implicated in the outbreak was grown in three California counties: Monterey, San Benito, and Santa Clara. Spinach grown in the rest of the United States has not been implicated in the current E. coli O157:H7 outbreak. The public can be confident that spinach grown in the nonimplicated areas can be consumed.” (U.S. Food and Drug Administration 2006d). Additionally, the Pennsylvania Department of Health confirmed the presence of the outbreak strain in a bag of Dole baby spinach. By Sept. 29, 2006, the FDA announced that “all spinach implicated in the current outbreak has traced back to Natural Selection Foods LLC of San Juan Buatista, California,” according to epidemiological investigations coordinated by the CDC at multiple state laboratories (U.S. Food and Drug Administration 2006a). The Grower Shipper Association of Central California, the Produce Marketing Association, the United Fresh Produce Association, and the Western Growers Association, said, “We are committed to working together as one industry to learn everything we can from this tragedy, and will redouble our efforts to do everything in our power to reduce the potential risk of foodborne illness. As we have in the past, we will work aggressively with the Food and Drug Administration and state regulatory authorities to ensure the industry’s growing and processing practices continue to be based on the very best scientific information available, and that we are doing everything possible to provide the nation with safe and healthy produce.” The FDA commented, “Implementation of these plans will be voluntary, but FDA and the State of California are not excluding the possibility of regulatory requirements in the future.” Also at this time, the Colorado Department of Public Health and Environment, the Ohio Department of Health, and the Nevada Department of Health and Human Services confirmed the presence of the outbreak in a sample of Dole spinach, and the Wisconsin Department of Health and Family Services confirmed its presence in two bags of Dole baby spinach. The Pennsylvania Department of Health found a second bag of Dole spinach containing the outbreak strain. On Oct. 4, 2006, the Arizona Department of Health Services found the outbreak strain in a bag of Dole spinach (U.S. Food Drug Administration 2006e). U.S. Attorney Kevin V. Ryan simultaneously issued search warrants on Growers Express and Natural Selection Foods (U.S. Food and Drug Administration 2006f). An Oct. 5, 2006, letter from Earthbound Farm cofounders, Drew and Myra Goodman stated, “While our food safety systems have always been at the top of the industry, this outbreak has demonstrated the immediate need for improved industry protocols.” The letter went on to describe the new safety measures now in place at the company’s growing and processing facilities (Goodman 2006). On October 12, 2006, FDA and the state of California announced that samples of cattle feces had tested positive for the outbreak strain of E. coli. Infected cattle feces were found on one of four fields implicated by a traceback investigation (U.S. Food and Drug Administration 2006h). “This is a significant finding because it is

Produce in Public: Spinach, Safety, and Public Policy

381

the first time we linked a spinach or lettuce E. coli O157:H7 outbreak to test results from a specific ranch in the Salinas Valley,” said California State Public Health Officer Dr. Mark Horton (California Department of Health Services Media Release 2006).

Summary Following the September 2006 E. coli O157:H7 outbreak linked to Natural Selections Foods, the California Food Emergency Response Team (CalFERT) conducted an investigation of the spinach supply chain (including growing conditions, harvesting, washing, and packaging). Investigators observed conditions and collected finished product and environmental samples. CalFERT investigators found no E. coli O157:H7 at the processing site and reported that: “no obvious sources for introduction of the pathogen were identified at the processing facility. However, a number of conditions were observed that may have provided opportunities for the spread of pathogens, if pathogens arrived on incoming spinach.” CalFERT investigators reported that E. coli O157:H7 was discovered in environmental samples taken from near each of the four fields linked to the outbreak-linked product codes. CalFERT investigators also reported that only one of the sampled fields produced E. coli O157:H7 isolates that matched the outbreak strain. These matching samples were taken from river water, cattle feces, and wild pig feces. Investigators reportedly found evidence of wild pigs among cattle pastures as well as in the spinach production fields. It was also found that there was a lack of fencing to keep wild animals such as pigs out of fields and that a well used for irrigation had a damaged casing. CalFERT investigators reported that a small amount of land on this site was leased to Mission Organics for a transitional readyto-eat organic spinach product. The product was being grown to organic standards but was being sold as conventional during a 3-year transition period. Ultimately, investigators showed that the E. coli O157:H7 was found on a transitional organic spinach field and was the same serotype as that found in a neighboring grass-fed cow-calf operation. These findings, coupled with the public outcry linked to the outbreak and the media coverage, sparked a myriad of changes and initiatives by the industry, government, and others. One question remains, which may never be answered: Why this outbreak at this time? A decade of evidence existed highlighting problems with fresh produce and warning letters were written, and yet little seemed to have been accomplished. The real challenge for food safety professionals, is to garner support for safe food practices in the absence of an outbreak, to create a culture that values microbiologically safe food from farm-to-fork at all times, and not just in response to the glare of the media spotlight.

References Allwood, P.B., Malik, Y.S., Maherchandani, S., Vought, K., Johnson, L., Braymen, C., Hedberg, C.W. and Goyal, S.M. 2004. Occurrence of Escherichia coli, noroviruses, and F-specific coliphages in fresh market-ready produce. Journal of Food Protection 67:2387–2390. Bartz, J.A. 1982. Infiltration of tomatoes immersed at different temperatures to different depths in suspensions of Erwinia carotovora subsp. carotovora. Plant Disease 66:302–305.

382

Section V. Public, Legal, and Economic Perspectives

Bartz, J.A., and Showalter, R.K. 1981. Infiltration of tomatoes by aqueous bacterial suspensions. Phytopathology 71:515–518. Bean, N.H., and Griffin, P.M. 1990. Foodborne disease outbreaks in the United States, 1973–1987: pathogens, vehicles, and trends. Journal of Food Protection 53:804–817. Beuchat, L.R. 1996. Pathogenic microorganisms associated with fresh produce. Journal of Food Protection 59:204–216. ———. 1998. Surface decontamination of fruits and vegetables eaten raw: a review. WHO/FSF/FOS/ Publication 98.2. World Health Organization, Geneva. 49 pp. Beuchat, L.R., and Ryu, J. 1997. Produce handling and processing practices. Emerging Infectious Diseases 3:459–465. Black, James. 2006. Toddler dies from E. coli bacteria. WBNS 10 17 Sept. 2006. Bridges, Andrew. 2006a. E. coli cases traced to bagged spinach. Associated Press. 15 Sept. 2006. ———. 2006b. E. coli outbreak spreads to ninth state. Associated Press. 15 Sept. 2006. ———. 2006c. Officials hunt for other E. coli sources. Associated Press. 16 Sept. 2006. ———. 2006d. Tainted spinach traced to California. Associated Press. 16 Sept. 2006. Brown, David. 2006. E. coli blamed on spinach: FDA warns about bagged vegetable. Washington Post. 15 Sept. 2006. Burnett, S.L., Chen, J., and Beuchat, L.R. 2000. Attachment of Escherichia coli O157:H7 to the surfaces and internal structures of apples as detected by confocal scanning laser microscopy. Applied and Environmental Microbiology 66:4679–4687. California Department of Health Services Media Release. 2006. State Health Department announces test results match genetic fingerprints to E. coli outbreak. 12 Oct. 2006. California Food Emergency Response Team. 2007. Investigation of an Escherichia coli O157:H7 Outbreak Associated with Dole Pre-packaged Spinach. Available at: http://www.dhs.ca.gov/ps/fdb/local/ PDF/2006%20Spinach%20Report%20Final%20redacted.PDF, accessed 05 May 2008. California Strawberry Commission. 2005. California Strawberry Commission safety program. Available at: http://www.calstrawberry.com/fileData/docs/FSP_English.pdf. Calvin, L., Avendaño, B., and Schwentesius, R. 2004. The economics of food safety: The case of green onions and hepatitis A outbreaks. Electronic Outlook Report from the Economic Research Service, United States Department of Agriculture. Report No. (VGS30501). CBC News. 2006. Don’t eat fresh spinach imported from U.S. CBC News (Health) 15 Sept. 2006. Available at: http://www.cbc.ca/story/health/national/2006/09/15/spinach-ecoli.html, accessed 16 Sept. 2006. CDC. 1997. Update: Outbreaks of Cyclosporiasis—United States and Canada, 1997. Morbidity and Mortality Weekly Report 46:521–523. ———. 2005. Outbreaks of Salmonella infections associated with eating Roma tomatoes—United States and Canada, 2004. Morbidity and Mortality Weekly 54:325–328. ———. 2007. Multistate outbreaks of Salmonella infections associated with raw tomatoes eaten in restaurants—United States, 2005—2006. Morbidity and Mortality Weekly Report 56:909–911. Dole Food Company, Inc. 2006. Dole Food Company statement on FDA investigation of packaged fresh spinach. 16 Sept. 2006. Earthbound Farm. 2006. Consumer update from Earthbound Farm RE: E. coli outbreak. 17 Sept. 2006. Available at: http://www.ebfarm.com/, accessed 17 Sept. 2006. Gladwell, Malcolm. 2000. The tipping point: how little things can make a big difference. USA: Malcolm Gladwell. Goodman, Drew & Myra. 2006. A letter from Earthbound Farm. Earthbound Farm 05 Oct. 2006. Available at: http://www.ebfarm.com/Press/SpinachUpdates/FromTheFounders.aspx, accessed 05 Oct. 2006. Harris, Gardiner, with files from Denise Grady. 2006. F.D.A. warns of outbreak and not to eat bag spinach. New York Times. 14 Sept. 2006. Available at: http://www.nytimes.com/2006/09/15/us/15spinach.html, accessed Sept 15, 2006. International Fresh-cut Produce Association. 1997. Voluntary Food Safety Guidelines for Fresh Produce. Alexandria, VA. 32 pp. International Fresh-cut Produce Association, the Produce Marketing Association, United Fresh Fruit and Vegetable Growers, and Western Growers. 2006. Commodity specific food safety guidelines for the lettuce and leafy greens supply chain. Available at: http://www.cfsan.fda.gov/∼acrobat/lettsup.pdf, accessed 12 Feb. 2008.

Produce in Public: Spinach, Safety, and Public Policy

383

Leiss, W., and Powell, D.A. 1997. Mad cows and mother ’s milk: the perils of poor risk communication. Montreal & Kingston: McGill-Queen’s University Press. Lin, Rong-Gong II. 2006. FDA issues spinach warning: The bagged product is suspected in an E. coli outbreak that has killed one and sickened 48. Los Angeles Times. 15 Sept. 2006. Luedtke, A., Chapman, B.J., Powell, D.A. 2003. Implementation and analysis of an on farm food safety program for the production of greenhouse vegetables. Journal of Food Protection 66:485–489. Lynn, C. (with files from Timothy Alex Akimoff). 2006. Officials warn of E. coli in bagged spinach: 5 confirmed cases in Oregon include a Salem woman. Statesman-Journal. 15 Sept. 2006. Available at: http://159.54.226.83/apps/pbcs.dll/article?AID=/20060915/NEWS/609150321/1001, accessed 16 Sept. 2006. Mukherjee, A., Speh, D., Dyck, E., and Diez-Gonzalez, F. 2004. Preharvest evaluation of coliforms, Escherichia coli, Salmonella, and Escherichia coli O157:H7 in organic and conventional produce grown by Minnesota farmers. Journal of Food Protection 67:894–900. Natural Selection Foods, LLC. 2006. FDA investigation; Packaged fresh spinach recall. 15 Sept. 2006. New Mexico Department of Health. 2006. State links first spinach sample to national E. coli outbreak. 20 Sept. 2006. Norton, Justin M. 2006. Tainted spinach. Associated Press. 18 Sept. 2006. Pal, Jyoti. 2006. E. coli outbreak linked to spinach. The Money Times. 15 Sept. 2006. Available at: http:// www.themoneytimes.com/articles/20060915/e_coli_outbreak_linked_to_spinach-id-101577.html, accessed 15 Sept. 2006. Powell, D.A., Bobadilla-Ruiz, M., Whitfield, A., Griffiths, M.W., and Luedtke, A. 2002. Development, implementation, and analysis of an on-farm food safety program for the production of greenhouse vegetables. Journal of Food Protection 65:918–923. Rafferty, S.M., Williams, S., Falkiner, F.R., and Cassells, A.C. 2000. Persistence in in vitro cultures of cabbage (Brassica oleracea var capitata l.) of human food poisoning pathogens: Escherichia coli and Serratia marcescens. ISHS Acta Horticulturae 530: International Symposium on Methods and Markers for Quality Assurance in Micropropagation. Cork, Ireland. Reller, M.E., Nelson, J.M., Mølbak, K., Ackman, D.M., Schoonmaker-Bopp, D.J., Root, T.P., and Mintz E.D. 2006. A large, multiple-restaurant outbreak of infection with Shigella flexneri serotype 2a traced to tomatoes. Clinical Infectious Diseases 42:163–169. Robertson, Jordan. 2006. Earthbound Farm scrutinized over spinach. Associated Press. 16 Sept. 2006. Seo, K.H., and J.F. Frank. 1999. Attachment of Escherichia coli O157:H7 to lettuce leaf surface and bacterial viability in response to chlorine treatment as demonstrated by using confocal scanning laser microscopy. Journal of Food Protection 62:3–9. Solomon, E.B., Potenski, C.J., and Matthews, K.R. 2002a. Effect of irrigation method on transmission to and persistence of Escherichia coli O157:H7 on lettuce. Journal of Food Protection 65:673–676. Solomon, E.B., Yaron, S., and Matthews, K.R. 2002b. Transmission of Escherichia coli O157:H7 from contaminated manure and irrigation water to lettuce plant tissue and its subsequent internalization. Applied Environmental Microbiology 68:397–400. Suslow, T. 1997. Postharvest chlorination: Basic properties and key points for effective disinfection. University of California. Publication 8003. 4 pp. Sweat, Charles. 2006. A statement from Charles Sweat, COO Natural Selection Foods (Spinach, E. coli). 15 Sept. 2006. Tauxe, R., Kruse, H., Hedberg, C., Potter, M., Madden, J., and Wachsmuth, K. 1997. Microbial hazards and emerging issues associated with produce, a preliminary report to the National Advisory Committee on Microbiologic Criteria for Foods. Journal of Food Protection 60:1400–1408. United Fresh Fruit and Vegetable Association. 1999. Industry-wide guidance to minimize microbiological food safety risks for produce. 16 pp. United Fresh Produce Association. 2006. United Fresh Produce Association applauds voluntary spinach recall. 16 Sept. 2006. U.S. Food and Drug Administration. 1998. Guide to minimize microbial food safety hazards for fresh fruits and vegetables. U.S. Department of Health and Human Services, Food and Drug Administration and Center for Food Safety and Applied Nutrition. Washington, D.C. 49 pp.

384

Section V. Public, Legal, and Economic Perspectives

———. 2004. Letter to firms that grow, pack, or ship fresh lettuce and fresh tomatoes. Center for Food Safety and Applied Nutrition, Office of Plant and Dairy Foods. 05 Feb. 2004. Available at: http://www. cfsan.fda.gov/∼dms/prodltr.html, accessed 24 Mar. 2008. ———. 2005. Letter to California firms that grow, pack, process, or ship fresh and fresh-cut lettuce. Center for Food Safety and Applied Nutrition, Office of Plant and Dairy Foods. 04 Nov. 2005. Available at: http://www.cfsan.fda.gov/∼dms/prodltr2.html, accessed 24 Mar. 2008. ———. 2006a. FDA announces findings from investigation of foodborne E. coli O157:H7 outbreak in spinach. FDA News. 29 Sept. 2006. ———. 2006b. FDA statement on foodborne E. coli O157:H7 outbreak in spinach. FDA News. 17 Sept. 2006. ———. 2006c. FDA statement on foodborne E. coli O157:H7 outbreak in spinach. FDA News. 24 Sept. 2006. ———. 2006d. FDA statement on foodborne E. coli O157:H7 outbreak in spinach. FDA News. 26 Sept. 2006. ———. 2006e. FDA statement on foodborne E. coli O157:H7 outbreak in spinach. FDA News. 04 Oct. 2006. ———. 2006f. FDA statement on foodborne E. coli O157:H7 outbreak in spinach. FDA News. 05 Oct. 2006. ———. 2006g. FDA statement on foodborne E. coli O157:H7 outbreak in spinach. FDA News. 06 Oct. 2006. ———. 2006h. FDA statement on foodborne E. coli O157:H7 outbreak in spinach. FDA News. 12 Oct. 2006. ———. 2006i. FDA Warning on Serious Foodborne E. coli O157:H7 Outbreak. FDA News. 14 Sept. 2006. ———. 2006j. Nationwide E. coli O157:H7 outbreak: Questions and Answers. Center for Food Safety and Applied Nutrition. 16 Sept. 2006. Available at: http://www.cfsan.fda.gov/∼dms/spinacqa.html, accessed 17 Sept. 2006. U.S. National Advisory Committee on Microbiological Criteria for Foods. 1999. Microbiological safety evaluations and recommendations on fresh produce. Food Control 10:117–143. Warriner, K., Ibrahim, F., Dickinson, M., Wright, C., and Waites, W.M. 2003a. Internalization of human pathogens within growing salad vegetables. Biotechnology & Genetic Engineering Reviews 20:117–134. ———. 2003b. Interaction of Escherichia coli with growing salad spinach plants. Journal of Food Protection 66:1790–1797. Wood, Daniel B. 2006. E. coli cases prompt calls to regulate farm practices. Christian Science Monitor. 18 Sept. 2006. Available at: http://www.csmonitor.com/2006/0918/p02s01-usgn.html, accessed Sept. 18, 2006. Zhuang, R.-Y., Beuchat, L.R., and Angulo. F.J. 1995. Fate of Salmonella montevideo on and in raw tomatoes as affected by temperature and treatment with chlorine. Applied Environmental Microbiology 61:2127–2131.

21 Contaminated Fresh Produce and Product Liability: A Law-in-Action Perspective Denis W. Stearns

Introduction The challenges of produce safety are not often examined from a legal, economic, and political perspective—or what is sometimes referred to as a “law-in-action” perspective.1 When one looks at the ever-growing number of foodborne illness outbreaks linked to contaminated produce, there is no escaping the conclusion that there are more challenges than understanding of the scientific basis of how and why such outbreaks occur. Although a significant majority of the foodborne illnesses caused each year are never attributed to a particular source, we know that during the last 10 years an increasing proportion of the illnesses have been caused by contaminated produce.2 For example, there have been 24 outbreaks associated with leafy greens since 1996, with twenty of those involving E. coli O157:H7. It was the 200+ illnesses, 5 deaths, and widespread publicity surrounding the 2006 outbreak linked to Dole baby spinach that finally did for spinach what the 1993 Jack in the Box outbreak of E. coli O157:H7 did for ground beef: it put the public on notice. Against this backdrop we can see that the risk of microbial contamination, whether it occurs during growing, harvesting, or processing, is not the only risk that must be understood. The fact of contamination occurring can have significant and far-reaching consequences, both economically and legally. Fortunately, the last 10 years of foodborne illness litigation, an excellent example of law-in-action, provides an effective lens through which to focus and gain the broader perspective necessary to better understand the challenges and the risks associated with growing, harvesting, processing, and selling fresh produce.

Legal Responsibility for Foodborne Illness When we speak of legal responsibility for a foodborne illness we are speaking of an area of the law broadly referred to as product liability. In general, product liability is an obligation, enforceable by a lawsuit, to pay monetary damages to a person as compensation for injuries caused by an unsafe product. To best understand product liability as it applies to food safety, it is helpful to know something about how this area of law has developed, moving from an era where persons injured by products had little in the way of effective remedies, to the current one in which a person injured by defective food has less problem successfully asserting a legal claim for damages. 385

386

Section V. Public, Legal, and Economic Perspectives

Understanding the Origins of Product Liability Nearly since the beginning of Western Civilization the law has existed as a means of enforcing rights or obligations. As far back as ancient Rome, there were two kinds of obligation under civil law: those that arose by agreement (contract) and those that arose out of a wrong (tort). Thus, if a person agreed to sell a bushel of wheat, the buyer could sue if the wheat was not delivered as agreed and collect contract damages. On the other hand, if a person stole wheat from a field, the grower could sue for tort damages. In both of these cases, an obligation was breached, one arising from contract, the other arising from tort, and the remedy was to order compensation for what had been lost. Seeking and obtaining compensation for a loss is what defines the civil law system, in contrast to criminal law, which has punishment as its primary purpose. Product Liability: From Contract to Tort Product liability law evolved from contract law, with the first decisions strongly favoring manufacturers. For a very long time, the “general rule” was that a contract was the sole source of obligation, and that a product need only be as good as promised. The corollary to this rule was that a manufacturer could not be sued, even for negligence, by someone with whom he had no contract. This was called the “rule of privity,” and it was most famously set forth in the 1842 case Winterbottom v. Wright.3 In this case, Mr. Wright was injured when a mail coach he was driving collapsed because of its shoddy construction. He sued the manufacturer of the coach for personal injury damages, but soon had his lawsuit thrown out of court. Showing no sympathy to Mr. Wright at all, the court said the following: … the only safe rule is to confine the right to recover to those who enter into the contract: if we go one step beyond that, there is no reason why we should not go fifty. The only real argument in favor of [allowing] the action is, that this is a case of hardship; but that might have been obviated, if [Mr. Wright] had made himself party to the contract. To a certain extent, this argument made sense, especially about making no exceptions, even in the case of hardship. In law and life, exceptions over time usually do swallow the rule, so it makes sense to avoid them. Also, the world Mr. Wright lived in was still small enough to have not yet made impractical face-to-face dealing with manufacturers of the products used in daily life. So one could more easily seek a promise that a product was safe, and thus have a means to hold the manufacturer to it. But that was soon to change. And it did so with the rise of the merchant class, a new breed of retailers who stood between the manufacturer and the buyer, making direct-dealing impossible. The Origins of Strict Liability in Tainted Food Cases The rule of strict liability has its roots in the judicial creation of a legal remedy for people injured by unsafe food.4 The seminal case is a 1913 decision by the Washington Supreme Court, Mazetti v. Armour & Company.5 It has been called “one of the most important cases in the development of early twentieth century product law in the

Product Liability: A Law-in-Action Perspective

387

United States.”6 The case involved canned tongue that had somehow gone foul, causing the person who ate it to become quite sick. In upholding the injured person’s right to sue the canner of the unsafe product, the court ruled: To the old rule [of privity] should be added another exception—not one arbitrarily worked by the court—but arising … from the changing conditions of society. An exception to a rule will be declared by courts when the case is not an isolated instance, but general in its character and the existing rule does not square with justice. Under such circumstances, a court will, if free from the restraint of some statute, declare a rule that will meet the full intendment of the law. This new rule was said by the court to be necessary due to the “modern method of preparing food for use by the consumer, and the more general and ever increasing use of prepared food products.” The rule was also premised on what the court called “the demands of social justice.” Plainly, the law was being forced to catch up with the rise of mass production and broader distribution of consumer products. A new relationship between producer and user was emerging, and the courts were being called upon to grapple with the sociolegal implications. Where once a person might grow their own food, or buy food from someone with whom they had a personal relationship, in a face-to-face dealing, now packaged food came from myriad sources with nothing to identify the maker except brand names. And so the Mazetti court held that anyone injured by a food product had the right to sue its manufacturer for the breach of an implied warranty that the food was safe, even in the absence of privity. This is the beginning of modern food product liability because the court accepted and addressed the reality the consumer faced in having to simply trust that the food they purchased was safe. Notably, the court also threw to one side the concern expressed in Winterbottom v. Wright about making hardship exceptions to a rule in order to provide an injured person with a legal remedy. This time the exception was made, and a new consumer-friendly remedy was created.

The Modern Rule of Strict Liability Although a precursor to strict liability, the remedy that the court created in the Mazetti decision was still one arising from contract. This changed, however, when the rule of strict liability was first announced by a court in 1963 in a case that involved a defective power tool. The case was Greenman v. Yuba Power Products, and it did away with the legal fiction that a manufacturer ’s liability for injury was based on the implied promise that the product was safe to use. Writing for the California Supreme Court, Chief Justice Roger Traynor, wrote that it was “clear that the liability is not one governed by the law of contract warranties but by the law of strict liability in tort.”7 The stated purpose was “to insure that the costs of injuries resulting from defective products are borne by the manufacturer that put such products on the market rather than by the injured persons who are powerless to protect themselves.” Under the new rule of strict liability, to hold a manufacturer liable, a person injured by a product need only show that: 1) the product was defective, 2) it was used as intended, and 3) the defect caused the injury. The care used in the manufacture of the product is irrelevant to the determination of liability. The only issue in a product

388

Section V. Public, Legal, and Economic Perspectives

liability case is the defectiveness of the product, not the conduct of the manufacturer in somehow allowing the defect to arise. As a result, proof of negligence is not required to recover damages; it is presumed based on the defect’s existence. Defining Products and Defects There are three kinds of product defects that give rise to strict liability: manufacturing defects, design defects, and marketing claims. Food injury claims primarily involve manufacturing defects, the most straightforward and uncontroversial of product claims. As one commentator has aptly pointed out, “when talking about a manufacturing defect, the need for a definition is not obvious. For decades, both courts and commentators considered the meaning of the ‘manufacturing defect’ concept so selfevident as to be self-defining.”8 The inquiry into whether a product is defective closely coincides with what most people would assume using common sense. A product is defective for not being how it was supposed to be. Put in more strictly legal terms, the product is not reasonably safe in construction because, as one state legislature has defined it, “the product deviated in some material way from the design specifications or performance standards of the manufacturer, or deviated in some material way from otherwise identical units of the same product line.”9 Proving the Existence of a Defect in Food Just as it is commonly assumed that proof of negligence is required to establish liability for a product-related injury, it is just as commonly assumed that proving the existence of a defect is difficult in food cases. This assumption might seem reasonable, at first glance, since food products are typically destroyed (i.e., eaten or discarded) and thus direct evidence of the defect rarely exists. Fortunately for the injured person, direct evidence is not required to prove the existence of a product defect, or precisely how or why the product failed. In manufacturing defect cases, the fact of product malfunction, and resulting injury, is by itself enough to give rise to a presumption of negligence and thus liability in most states. This is sometimes referred to as the malfunction doctrine.10 Its fundamental premise is the high correlation between the existence of a defect and a failure of some kind in the manufacturing process.11 This makes the issue of negligence not worth the cost and uncertainties of trying to prove it. Thus, with a manufacturing defect, it is not a useful exercise to ask whether the defect could have been prevented; the existence of the defect is by itself sufficient to impose liability. For cases involving unsafe food, it is nearly always a manufacturing defect at issue, especially when pathogens such as E. coli O157:H7, Salmonella, or hepatitis A are involved. And while it is true that a manufacturer is not liable for a product-related injury unless the product is both defective and unsafe, in food cases this is a distinction without a difference. Food that is unsafe, because it is unfit to eat, is by definition defective. For that reason, it is rare to have a defendant in a food contamination case dispute liability unless there is a serious question of causation, or some other productrelated problem of proof.12 Moreover, since only cases with problems of proof, or uncertain damages, tend to go to trial, this would explain the low win-percentage for plaintiffs who go to trial, and the relatively small damage awards for those cases the

Product Liability: A Law-in-Action Perspective

389

plaintiffs do win.13 But when there is a confirmed illness, and a link to an investigated outbreak, there is no real way to dispute the existence of the defect. Proof of Causation Under the malfunction doctrine, proof of injury is proof of defect only if the injured person can prove the defect caused the injury. At first glance this appears simple, at least to many people contacting a law firm and claiming to have a valid foodborne illness claim. Based on over a decade of experience responding to telephone calls and emails from persons thinking they have a legal claim, I can state with a high degree of confidence that the vast majority of people assume that the last restaurant meal that they ate was “for sure” the cause of their illness. But this is rarely the case for several reasons. First, incubations periods vary significantly depending on the pathogen involved, and can extend upwards to several weeks.14 Unaware of this fact, people wrongly assume a shortened cause-and-effect between their last restaurant meal and their illness. As a result, without knowing what pathogen is involved (assuming one is), it is difficult to prove causation. Second, even assuming the person received medical attention, physicians do not routinely order the stool cultures needed for pathogen identification.15 Third, unless the illness is seen as being potentially part of a foodborne illness outbreak, there is little likelihood of a health department investigating the source of the illness.16 Fourth, the nationwide distribution of mass-produced food makes it difficult to detect multistate outbreaks possibly related to low-level contamination of food.17 It is for these reasons, and no doubt others, that the vast majority of foodborne illness in the United States has an unknown cause.18 Consequently, even with the relative ease of proof associated with an outbreak-related defective food claim, litigation remains a weak economic incentive for increased investment in food safety. As one study concluded: Because firms responsible for microbial contamination compensate relatively few foodborne illnesses, the legal system provides only limited feedback to firms about the need for greater food safety.19 Plainly, where food companies can reasonably assume that an illness caused by their product will never be traced back to them, there is little reason to spend more to prevent or reduce the incident rate of the defects.

Traceability and Product Liability Although proof of negligence is not required to establish strict liability, a person who alleges injury and hopes to recover compensation must still be able to identify the product that caused the injury and prove causation. This is where most potential legal claims fail, and it is the chief reason for turning away a potential client seeking to file a lawsuit. This is especially true when it comes to fresh produce. One major impediment to imposing legal liability in a case involving contaminated fresh produce is the lack of traceability in the distribution system. In most such cases, at least those that involve a branded product, the manufacturer is easily identified, as

390

Section V. Public, Legal, and Economic Perspectives

in the 2005 Dole Romaine lettuce outbreak. But an effective traceback all the way to the grower has been a rare event. One reason for this is the absence of any statute or regulation, state or federal, imposing detailed record-keeping requirements related to distribution. Such requirements as they exist are imposed by the Perishable Agricultural Commodities Act (PACA) regulations.20 But the primary purpose of these recordkeeping regulations is to ensure that growers receive proper payments for the produce shipped.21 In response to the growing number of outbreaks linked to microbial contamination of produce, in 1998, the FDA developed guidelines to improve traceability. One stated reason for the need for effective traceback was described as follows: Despite the best of efforts by food industry operators, food may never be completely free of microbial hazards. However, an effective traceback system, even if only some items carry identification, can give investigators clues that may lead to a specific region, packing facility, even field, rather than an entire commodity group. Narrowing the potential scope of an outbreak could lessen the economic burden on those industry operators not responsible for the problem.22 Notwithstanding the importance ascribed by the FDA to effective traceback records, the guidelines remained wholly voluntary. In fact, the 2006 Dole spinach outbreak is notable for being one of the few investigated outbreaks in which a specific grower was implicated, thereby making it possible to hold the grower of the implicated spinach legally liable, too.23 CalFERT (the California Food Emergency Response Team) was assembled and comprised members from the FDA and the California Department of Health Services. This success in tracing the contamination source to a particular field may have been a harbinger of future successes, in large part attributable to greater traceback efforts being expended by public health officials, like CalFERT. For example, the investigation into the 2006 Taco John’s E. coli O157:H7 outbreak was also able to trace the contaminated lettuce back to a particular field.24 In both of these cases it was environmental testing of irrigation water and swab testing of cattle that ultimately provided the link when E. coli O157:H7 was found and then subjected to Pulsed Field Gel Electrophoresis testing that showed a match to bacterial isolates obtained by confirmed outbreak cases. Because it is relatively unusual for a particular grower to be identified as the source of contaminated produce, those further up the chain of distribution are going to be those most likely to be held liable. For example, this was the case in the 2005 Dole romaine lettuce outbreak, where it was Dole that was the sole defendant named in several of the lawsuits filed against it. The product traceback performed with regard to this outbreak found that: The implicated lettuce reportedly could have been harvested from any one of seven fields in the Salinas Valley of California. The ultimate source of the contaminated lettuce was not identified.25 Finally, it should be noted that it is only when there is a branded product involved that there is a substantial likelihood that illnesses will be linked to a particular product. Brand names act as proxies in the absence of other records, allowing people to better remember their food exposures, and making it easier for epidemiologists to spot foods

Product Liability: A Law-in-Action Perspective

391

that ill individuals may have in common. Added to that is the increased use of customer loyalty programs at grocery stores that give consumers the ability to obtain proof of a particular purchase on a particular date. In contrast, if we are dealing with a commodity produce product like heads of iceberg lettuce, or bunches of green onions, it becomes less likely that such products will be identified as potential sources of infection.

The Persistent Role of Negligence in Product Liability Negligence is the failure to exercise ordinary care, and in most cases knowledge of the risk defines ordinary care. Actual knowledge is not required, however. The law attributes to one who acts both what is known and what should be known or have been discovered. In other words, ignorance is no defense where the facts known or available would have alerted a reasonable person to the likelihood of danger. When dealing with a strict liability claim involving a manufacturing defect we know that proof of negligence is not required. This does not mean though that there is no fault; it means only that the plaintiff need not prove fault to hold the manufacturer liable for damages. And since a manufacturer cannot be held liable more than once, and so provide an injured person with a kind of double recovery, proving the elements of negligence, in addition to the elements of strict liability, gains nothing. That said, there are times when proof of negligence is necessary, as in when the entity being sued is not a manufacturer, and strict liability does not apply. It can also be necessary when trying to show that one or more entities involved in the manufacture or sale of the product have varying degrees of fault. And, from a psychological point of view, a jury is more likely to award higher damage amounts if jurors are angered by the bad behavior of the company being sued. As a result, negligence continues to play a large law-in-action role. Statutory Law and Standards of Care: How Careful Is Careful Enough? Both tort and statutory law have regulatory effects. For example, a statute like the Federal Food, Drug, and Cosmetic Act (FDCA) prohibits the introduction into interstate commerce of food that is adulterated by virtue of containing a poisonous or deleterious substance that may render it injurious to health, or by virtue of having been prepared, packed, or held under unsanitary conditions whereby it may have been rendered injurious to health, among other things.26 But that statutory prohibition does not dictate a particular way to prevent food from becoming adulterated. Put another way, it does not describe the standard of care that a food processor must exercise; the statute simply states that the food should not be adulterated. Thus, it was left to the FDA, which has jurisdiction under the FDCA, to define the standard of care through the promulgation of regulations, including, among other things, good manufacturing practices (GMP) and regulations that require implementation of Hazard Analysis and Critical Control Point (HACCP) systems.27 The GMPs are not the end of the story, however. Although the GMPs are not often revised, the FDA continually issues guidance documents in the form of policy statements, letters to industry, and industry guidelines, none of which are intended to have binding regulatory effects, but all of which make recommendations about the level of

392

Section V. Public, Legal, and Economic Perspectives

care a producer should use in processing fresh produce.28 To the extent that the FDA is defining an expected standard of care, it is one that is in a near-constant state of reactive flux. Either way, though, food processors should always expect to have their conduct judged against whatever guidelines or recommendations exist, whether they are legally binding or not. Still, one might assume, if a regulation exists, that standard would control the determination of whether a grower or processor used sufficient care to avoid liability. But this is not necessarily the case either. Although a law or regulation certainly establishes a standard, it does not establish the only standard.29 As a result, proven compliance with an applicable regulation is not a legally sufficient defense to liability. In contrast, proven noncompliance with an applicable regulation is a legally sufficient way to establish liability. Such liability is referred to as negligence per se. The doctrine of negligence per se is a precursor to strict liability and is often compared to it. To employ the doctrine, a person injured by a defective food product need only show that: 1. There is an applicable statute that was enacted to protect a class of persons that includes the person claiming injury; 2. The injury complained of is the kind contemplated by the statute; 3. The company that is alleged to be negligent violated the statute; and 4. The violation must have caused the injury.30 Court decisions have, however, been generally disinclined to find that a violation of the FDCA constitutes negligence per se.31 There has been no such disinclination to find that a violation of a regulation promulgated pursuant to the FDCA is legally sufficient evidence for a jury to find negligence. For example, in Allen v. Delchamps, Inc., the Supreme Court of Alabama reversed a trial court’s dismissal of a lawsuit against a grocery store for injuries related to an allergic reaction to sulfides on fresh celery.32 In any case, regardless of whether there is a regulatory violation or compliance, it must be remembered that neither is relevant to the issue of strict liability. Although the doctrine of negligence per se is still around, it is unnecessary. Recall that a person does not get to recover damages for each liability theory established. Only one theory of liability is needed to recover damages. Thus, the real lesson to be learned here is that regulatory compliance is no defense. The only real defense is to exercise the care necessary to prevent the defect. Only without injury can there be no liability. The Politics of Regulations: Setting the Safety Bar Low The inability of consumers to discern the relative safety of their food purchases limits all but a generalized demand for safer food. This generalized demand for safer food, applicable to an entire industry or product category, is ineffective in causing the market to enhance food safety. As a result, there is little economic incentive for producers to manufacture food that is safer than that required by government regulations. Such regulations, therefore, tend to act as a ceiling not a floor, and they effectively suppress most competition in the realm of food safety. Rather than worrying about a competitor doing more to improve the relative safety of a product category (e.g., bagged fresh

Product Liability: A Law-in-Action Perspective

393

produce), regulations impose a predictable cost that companies can meet, but need not exceed. Thus, even though certain spinach growers had invested far more than others before the 2006 Dole baby spinach outbreak occurred, the outbreak hurt the market as a whole.33 Nonetheless, in the wake of the 2006 spinach outbreak, the spinach industry actively sought regulations, albeit ones that would be created, policed, and enforced by industry, not the government. Sensing that certain competitors had already invested in many of the food safety improvements that proposed regulations might require, and that the food safety bar might be set too high, the green leafy produce industry drafted marketing agreements that would put in place a set of minimum requirements that all market participants would have to meet to sell their produce.34 It is notable that the minimum requirements were less stringent than what one major market participant already had in place. For example, in one article examining the changes the following was noted: Fresh Express requires an 800-foot buffer between fields of leafy greens and pastures and one-mile buffers between leafy greens and feed lots. “The (leafy greens) metrics could do better, but they certainly set a floor,” says Jim Lugg, food-safety chief for Fresh Express.35 By setting the safety bar lower, and ceding the more stringent requirements to the then market leader, the marketing agreement had the effect of leveling the playing field for the rest of the market, and ensuring that all would bear similar costs in meeting the improved safety requirements. This was, in fact, an anticompetitive move that created a set of safety requirements that were less stringent than what would have likely resulted if market participants had competed in an open market for safety. Such a move was also an attempt to head off the efforts of the Food Marketing Institute, which represents large retailers and wholesalers, and the National Restaurant Association to develop a separate set of safety requirements that could be imposed as contractual purchasing requirements. In short, the spinach industry wanted to make sure that safety improvements cost as little as possible.

Reducing Liability by Spreading the Blame Given the strictness that is strict liability, in practice, the best defense to such a claim is the ability to shift the liability to elsewhere. This can be attempted in a variety of ways, some likely to be more successful than others. Indemnification, Contribution, and Allocation of Fault Not long after its widespread adoption, the doctrine of strict liability expanded to include everyone in the chain of distribution, meaning that the sale of a defective product was enough to create liability, even in cases where the seller had no reason to suspect that the product was defective and no opportunity to prevent or warn about the defect. Not surprisingly, when deciding whom to sue over an injury-causing defective product, the immediate seller was the most popular choice. One big reason for this

394

Section V. Public, Legal, and Economic Perspectives

was the rule of joint and several liability. With joint liability, anyone who in any way contributed to causing the injury is liable for the entire amount of the damages. In contrast, several liability is the liability for only that portion of the damages for which you can be deemed responsible for causing. The rule of joint and several liability is a confusing name meaning that a company is both jointly and severally liable for damages. It is because of the rule of joint and several liability that we get the term “deep pocket” defendant. When a severely injured person is looking to sue he will want to pick the defendant who has either the most money or insurance (i.e., one with “deep pockets”) so he can recover all of his damages from one defendant, rather than being forced to sue multiple defendants in order to fully recover his damages. What, then, was the seller hit with a big product-related damage award to do? One thing was an indemnity or contribution action, a subsequent lawsuit filed to recover the damages paid on behalf of the person or company actually at fault. So, for example, if you are a grocery store owner and end up getting sued for selling a bag of contaminated baby spinach, you could pay the damage award (or, more likely settle), and then turn around and sue the processor of the baby spinach to recover what you paid. The processor could then turn around and sue the grower who sold the contaminated baby spinach. Or, in the more likely scenario, all of these entities would be joined in the initial lawsuit filed by the injured person, and the jury would then sort out who is responsible for how much of the fault and thus the damages. Rather than leave the allocation of fault to chance, many choose to allocate fault by contract. Thus, for example, the supply agreement between the grower and the processor might state that the grower is responsible for all damages related to the supply of contaminated produce, regardless of what a jury ends up deciding. Such an agreement would be made a condition of the two entities doing business, and might also include the requirement that the grower obtain sufficient insurance to cover any possible award of damages and also attorney fees. Agreements like this (sometimes also referred to as “hold harmless” agreements) are becoming increasingly common in the produce industry. Consumer Responsibility and Contributory Negligence In many food cases, particularly those involving meat products, the issue of consumer responsibility can become an issue. For example, in numerous cases involving ground beef, the manufacturer has taken the position that the injury occurred not because of contamination, but because the consumer did not cook the ground beef properly. In legal terms, this is usually called contributory negligence, because the injured person’s own negligence is alleged to have contributed to his injury. The way this usually plays out is that the jury is asked to assign a percentage amount of fault to the contributory negligence, and the amount of damages awards is reduced by that percentage. Some states disallow any recovery if the percentage of fault is higher than 50. The defense of contributory negligence is unlikely to be an issue in cases involving produce for one simple reason: fresh produce is commonly eaten raw. Indeed, with leafy greens, such as lettuce and spinach, this is nearly always the case. And with products such as bagged salads, the fact of their being prewashed and ready-to-eat is the key selling point. As a result, it would be an exceedingly unusual case where the

Product Liability: A Law-in-Action Perspective

395

manner in which the produce was handled by the consumer could be said to have contributed to a resulting injury. Of course, if the produce was contaminated by the consumer—for example, by cross-contamination from meat—that would be a different situation entirely. But barring these kinds of facts, the defense of contributory negligence is simply unavailable.

The Sustainability and Local Food Movements and Their Potential Effects on the Future of Product Liability Historically, the rise of strict liability can be seen, in large part, as a reaction to the parallel rise in the mass production and mass marketing of consumer products, especially food. As one court has explained in a decision adopting the rule of strict liability: The rule requiring a person injured by a defective product to prove the manufacturer or seller negligent was evolved when products were simple and the manufacturer and the seller generally were the same person. Knowledge of the then purchaser, if not as complete as the seller ’s, was sufficient to enable him not only to locate the defect but to determine whether negligence caused the defect and if so whose. The purchaser of the present day is not in this position.”36 And the purchaser of today still is not in this position; indeed, nearly 40 years later, the purchaser ’s position is arguably much worse. Although a century ago we would buy our fresh leafy greens from the farmer who grew them, now we buy salad in a bag without any knowledge of where in the world the produce was grown and who in the world processed it. Even the brand name on the bag is no indication of who processed the produce, as we saw in the 2006 E. coli O157:H7 outbreak linked to bags of Dole baby spinach. Dole had contracted with another company to process and bag the baby spinach, and that company had in turn contracted with a different company to grow and source the baby spinach. These facts did not mean, however, that Dole could escape having a strict liability claim asserted against it as the “apparent manufacturer” of the product.37 Instead, the facts emphasize just how difficult it is for the consumer to know whom to trust. But now, in reaction to the string of widely publicized outbreaks, the public is turning increasingly back to local farmers as a produce source that can be trusted. This trend is often referred to as the “local food” movement, and it is in many ways a return to past practices. Farmers markets selling locally grown produce were once a hallmark of most major cities.38 A recent nationwide bestseller by Barbara Kingsolver also makes clear the growing allure of knowing the source of one’s food.39 She writes the following: This is a story of a year in which we made every attempt to feed ourselves animals and vegetables whose provenance we really knew … and of how our family was changed by our first year of deliberately eating food produced from the same place where we worked, went to school, loved our neighbors, drank the water, and breathed the air.

396

Section V. Public, Legal, and Economic Perspectives

Far from being a romantic notion having nothing to do with food safety or the law of product liability, the move toward the consumption of local food represents a reaction to the same insecurity that called the rule of strict liability into being: the need for food to be safe and the need for a remedy when it is not.

Conclusion Consumers of fresh produce want choices that they can live with—literally. They will pay more for the freshness and convenience of prewashed and packaged produce, but not at the price of suffering injury or losing their lives. Indeed, if there was a way for consumers to look at a produce product and tell whether it was contaminated, only the safest producers would stay in business. Unfortunately, there is no way yet for buying decisions to be based with confidence on safety considerations. That is one reason why the demand for organic produce has gone up, as has the attractiveness of the local food movement. Such products may not be certainly safe, but they certainly appear to be more trustworthy. In the end, with food, it is still all about trust and a promise of safety and quality kept.

References 1. The “law in action” is a multidisciplinary approach that focuses on investigation of how the law actually works, using empirical studies, economics, sociology, and case studies. See, e.g., Marc Galanter, Real World Torts: An Antidote to Anecdote, 55 Md. L. Rev. 1093 (1996) (using empirical research to refute the existence of a “crisis” in the tort system). This approach is usually associated with the University of Wisconsin Law School, which not coincidentally is the author ’s alma mater. For an essay on what law-in-action means at the law school, see Kenneth Davis, Law in Action: The Dean’s View. Available at http://www.law.wisc.edu/law-in-action/davislawinactionessay.html, accessed September 30, 2008. 2. Linda Calvin, Outbreak Linked to Spinach Forces Reassessment of Food Safety Practices, Amber Waves, June 2007. 3. Meeson and Welsby 109, 152 English Reports 402 (1842). 4. D.G. Owen, Manufacturing Defects, 53 S.C. L. Rev. 851 (Summer 2002). 5. 135 Pac. 633 (Wash. 1913). 6. M. Shapo, THE LAW OF PRODUCT LIABILITY, ¶ 6.01[2] (3d Ed. 1994). 7. 59 Cal.2d 57 [13] A.L.R.3d 1049 (1963). 8. Owen at p. 865–894, supra at Note 4. 9. Revised Code of Washington, 7.72.030(2)(a) (defining one standard of strict liability for a product manufacturer). See also Owen at 866–70, supra at Note 4 (discussing the historical development of the “departure from design” test). 10. Owen at 871–74, supra at Note 7. This doctrine is usually understood as a variation on the doctrine of res ipsa loquitor, which means “the thing speaks for itself.” 11. G. Schwarz, New Products, Old Products, Evolving Law, Retroactive Law, 58 N.Y.U. L. Rev. 796, 810 (1983). 12. Owen at 855–56 and fn. 27, supra at Note 7. The food cases at Marler Clark have consistently borne this out. See, e.g., Almquist v. Finley School District, 57 P.3d 1191 (2002) (conceding E. coli O157:H7 in a school lunch taco meat would make it defective but denying that the taco meat was the cause of the outbreak in question). 13. J.C. Buzby et al., Product Liability and Microbial Foodborne Illness, Economic Research Service/ USDA, AER-799, at 13–23 (noting that, of 175 foodborne illness lawsuits that went to verdict from 1988–1997, only 31.4% were won by plaintiffs, and the median damage award was $25,560). 14. Id. at 6 (graphical representation of widely differing incubation periods, from less than an hour to over 3 weeks).

Product Liability: A Law-in-Action Perspective

397

15. Thomas Hennessy, et al., Survey of Physician Diagnostic Practices for Patients with Acute Diarrhea: Clinical and Public Health Implications, CID, 2004:38, S203-11 (Suppl 3) (stool cultures were done for less than 50% of patients with diarrhea). 16. Buzby at 7, supra at note 12. 17. Ban Mishu et al., Surveillance for Sporadic Foodborne Disease in the 21st Century: The FoodNet Perspective, CID, 2004:38, at S119. 18. Paul S. Mead et al., Food-Related Illness and Death in the United States, 5 Emerging Infect. Dis. (No. 5) 607, 614 (1999) (estimating that there are 76 million cases of foodborne illnesses each year with over 80% caused by no known source). 19. Buzby at 27, supra at note 12. The Buzby study, while good, is flawed for its failure to sufficiently recognize the implications of the fact that the vast majority of legitimate food-injury claims never go to trial and are privately settled. In the 10-year history of Marler Clark, only one of the firm’s cases ever went through trial to verdict; all others settled. 20. 7 C.F.R. Part 46 (promulgated pursuant to PACA, 7 U.S.C. 499, et seq.) 21. E. Golan, et al., Traceability in the U.S. Food Supply: Economic Theory and Industry Studies, USDAERS, AER-830, at 12 (March 2004). 22. FDA/CFSAN, “Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables,” Section IX (hereinafter referred to as FDA Safe Produce Guidelines). Available at http://vm.cfsan.fda. gov/∼dms/prodguid.html, accessed September 30, 2008. 23. See California Food Emergency Response Team (CalFERT), Investigation of an Escherichia coli O157:H7 Outbreak Associated with Dole Pre-Packaged Spinach, Final Report, March 21, 2007 (hereinafter “CalFERT Report”). 24. See CalFERT, Investigation of the Taco John’s Escherichia coli O157:H7 Outbreak Associated with Iceberg Lettuce, Final Report (February 15, 2008). 25. Minnesota Department of Health, Final Report, A Multistate Outbreak of Escherichia coli O157:H7 Infections Associated with Dole Brand Prepackaged, Prewashed Lettuce Salad—Minnesota, Oregon, Wisconsin, at 6 (September 2005). 26. 21 U.S.C. 342(a)(1) (defining one kind of “adulteration”). 27. Current Good manufacturing practices (GMPs). 21 C.F.R. Part 110, promulgated pursuant to Federal Food, Drug and Cosmetics Act, 21 U.S.C. § 301, et seq. 28. See, e.g., FDA Safe Produce Guidelines; and the 2004 FDA Produce Safety Action Plan, Produce Safety From Production to Consumption: 2004 Action Plan to Minimize Foodborne Illness Associated with Fresh Produce Consumption. Available at http://www.cfsan.fda.gov/∼dms/prodpla2.html, accessed September 30. 2008. 29. There is one exception, however, and that is in the case of what is called federal preemption. A full discussion of preemption is beyond the scope of this article, but briefly, this involves a situation in which federal law sets a standard that preempts any different standard under state law, including product liability law. For a discussion of preemption in the context of USDA rule-making, see D. Stearns, Preempting Food Safety: An Examination of USDA Rulemaking and Its E. coli O157:H7 Policy in Light of Estate of Kriefall ex rel. Kriefall v. Excel Corp., 1 Food Law & Policy, 375 (Fall 2005) (discussing USDA policy allowing E. coli O157:H7 on intact cuts of meat). 30. See, generally, RESTATEMENT (2nd) OF TORTS § 286 (1965). 31. See David G. Owen, Proving Negligence in Modern Products Liability Litigation, 36 Ariz. St. L.J. 1003, note 36 and corresponding text. 32. 624 So.2d 1065, 1068 (Ala. 1993) (citing FDA regulations in holding that the question of “whether sulfite preservatives on fresh produce are within the reasonable expectation of a consumer is a question for a jury”). 33. Calvin, supra note 2 (“With the fall 2006 outbreak, all spinach growers suffered from decreased consumer demand for their product, even though only one grower ’s spinach was contaminated.”). 34. See, generally, Matthew Kohnke, Reeling in a Rogue Industry: Lethal E. coli in California’s Leafy Green Produce and the Regulatory Response, 12 Drake J. Agric. L. 493 (Fall 2007) (arguing that the safety standards put into place by industry are both too lax and not enforceable). 35. Julie Schmit, Last fall’s E. coli outbreak in fresh spinach spurs changes, USA Today, Sept. 20, 2007. 36. Buttrick v. Arthur Lessard & Sons, Inc., 260 A.2d 111, 113 (N.H. 1969).

398

Section V. Public, Legal, and Economic Perspectives

37. See Wojciuk v. United States Rubber Co., 108 N.W.2d 149, 152 (Wis. 1961) (“One who puts out as his own product a chattel manufactured by another is subject to the same liability as though he were its manufacturer.”) (quoting and adopting RESTATEMENT (2d) OF TORTS § 400, at 1086 (1939)). 38. Alison Brown, Counting Farmers Markets, 91 Geographical Review, 655–74 (Oct. 2001) (noting that such markets were eclipsed by wholesale supply systems, but have since begun to make a comeback). Interestingly, French consumers reacted quite differently to the trend toward the industrialization of the food system, and recoiled from it. This gave rise to a government-regulated system of quality signs that were attached to localized products. See Randall E. Westgreen, Delivering Food Safety, Food Quality, and Sustainable Production Practices: The Label Rouge Poultry System in France, 81 American Jnl. Agric. Econ., 1107–111 (1999). 39. Barbara Kingsolver, ANIMAL, VEGETABLE, MIRACLE: A YEAR OF FOOD LIFE (2007).

22 The Economics of Food Safety: The 2006 Foodborne Illness Outbreak Linked to Spinach Linda Calvin, Helen H. Jensen, and Jing Liang

Introduction On September 14, 2006, the U.S. Food and Drug Administration (FDA) announced that consumers should not eat bagged spinach because of a foodborne illness outbreak of the potentially deadly bacterium Escherichia coli O157 : H7. Retail and food-service firms immediately cleared bagged spinach from their shelves and menus. Spinach sales closed down overnight. By the time the outbreak was over, 204 people became ill across 26 states and 1 province in Canada, 104 people were hospitalized, 31 developed the serious complication of Hemolytic Uremic Syndrome (HUS), and 3 died. Eventually, the FDA determined that one 2.8-acre field was the most likely source of all the contaminated spinach, but it could not identify the method of contamination. In the wake of the outbreak, the spinach—and more generally leafy greens—industry, retailers, and food-service buyers, and the government reassessed their strategies to reduce the risk of microbial contamination (Calvin 2007). The widespread impact of a small quantity of contaminated spinach emphasizes the fact that there can be significant public health effects and economic spillovers from the actions of an individual grower. Although spinach and other leafy greens have been associated with numerous foodborne illness outbreaks, the risk of becoming ill from spinach is low. In 2005, U.S. consumers ate 680 million pounds of fresh spinach and the load of contaminated spinach associated with the outbreak totaled only 1,002 pounds. However, leafy greens are the most likely produce category to be associated with an outbreak. From 1996 to 2006, leafy greens have accounted for 34% of all outbreaks due to microbial contamination traced back to a specific fruit or vegetable, 10% of illnesses, and 33% of deaths (Table 22.1). Of the 24 outbreaks traced to leafy greens in the United States since 1996, 20 have been associated with E. coli O157 : H7 contamination (Fig. 22.1). Three other outbreaks were related to Cyclospora and one to Salmonella. None of the previous foodborne illness outbreaks linked to leafy greens had the number of illnesses and deaths, negative publicity, market impact, or industry response of the 2006 outbreak associated with spinach. Over this period, only two outbreaks were associated with spinach, but they accounted for all five deaths associated with leafy greens. Although the spinach outbreak received a great amount of publicity, it was not the largest outbreak linked to produce in terms of illnesses. The 1996 outbreak associated with Cyclospora contamination of Guatemalan raspberries sickened 1,465 people in the U.S. and Canada, but no one died. Nor was the outbreak linked to spinach the most deadly. Although it is sometimes difficult to attribute death to a particular cause, the 2003 outbreak associated with green onions from Mexico contaminated with the 399

Table 22.1. Foodborne illness outbreaks attributed to produce, 1996–2006 Commodity Leafy greens Lettuce Mixed lettuce Romaine lettuce Spinach Cabbage Basil or mesclun lettuce mix Tomatoes Melons Cantaloupe Other melons Raspberries and other berries Herbs Basil Parsley Green onions Almonds Green grapes Snow peas Squash Unknown

Outbreaks 14 1 4 2 1 2 12 7 4 6 4 2 3 2 1 1 1 2

Total

71

Source: U.S. Food and Drug Administration.

Outbreaks

Illnesses

7

400 Outbreaks

6

350 Illnesses

5

300 250

4

200 3

150

2

100

1

50

0

0 1996

1998

2000

2002

2004

2006

Figure 22.1. E. coli O157 : H7 outbreaks (vertical bars) and illnesses (diamonds) linked to leafy greens, 1996–2006. Source: U.S. Food and Drug Administration.

400

The Economics of Food Safety

401

hepatitis A virus was linked with 4 deaths. However, the outbreak linked to spinach was probably the biggest in terms of industry impact, primarily because of the FDA’s announcement to the public to immediately stop consuming bagged spinach. With so much at stake in terms of the loss of consumer confidence and potential for involvement by the federal government, the industry needed to mount a strong response to the FDA action. By the beginning of the next marketing season (April 2007), the California leafy green industry introduced the California Leafy Green Marketing Agreement (LGMA). The LGMA is a voluntary state marketing agreement, which requires that any California leafy greens handled by signatories be produced according to the agreement’s new food safety standards. Although the exact benefits associated with this agreement are uncertain, they are potentially large and reflect an assessment by the industry that the benefits of such action likely outweigh the costs of meeting the new standards. This chapter begins with a description of the U.S. produce industry, with a particular focus on spinach and other leafy greens, and factors that relate to food safety. The next section reviews the response to food safety outbreaks, including the FDA’s good agricultural practices (GAPs) and the leafy greens industry’s LGMA, which contains its own Best Practices. The third section discusses the economics of adoption of GAPs and the LGMA standards. The fourth section examines the economic impact of the 2006 outbreak linked to spinach. We end with some conclusions regarding the role of economic factors in determining approaches to reducing food safety hazards in fresh produce.

The U.S. Produce Industry U.S. consumers are eating more fruit and vegetables, with per capita consumption increasing 7% from 1990 to 2005. The increase in consumption varies by type of product, with vegetable consumption increasing by 8% and fruit consumption increasing by 6%. The vegetable per capita consumption statistics include selected vegetables and potatoes, but do not include mushrooms, sweet potatoes, dry peas and lentils, or dry edible beans. Fruit consumption statistics include selected fruit, but do not include tree nuts. Per capita statistics can be found in two annual ERS publications: Fruit and Tree Nuts Situation and Outlook Yearbook and Vegetables and Melons Situation and Outlook Yearbook. Total spinach consumption has grown more rapidly than the average, increasing 90% from 1992 to 2005, from 1.6 pounds per capita to 3 pounds per capita per annum. If there is a contamination problem associated with a certain commodity and consumption of that commodity goes up, the probability of an outbreak also increases correspondingly. Overall, more produce is being consumed fresh than processed (canned, frozen, dried, or juiced) and the fresh produce share is growing. From 1990 to 2005, the percent growth in fresh consumption exceeded the growth in total consumption: 15% for vegetables and 90% for fruit (see Table 22.2). This shift toward fresh products also increases the associated food safety risk. Produce that is consumed uncooked, such as raw spinach, poses more risk than produce that has been treated with a kill step, such as cooking for fresh produce, heating for canned and frozen fruit and vegetables, or pasteurization for juice. Spinach consumption patterns have exhibited a

Table 22.2. Fresh fruit and vegetable consumption and imports

Item Fruit:1 Bananas Apples Oranges Grapes Strawberries Pineapples Peaches2 Avocados Pears Lemons Grapefruit Tangerines and tangelos Limes Mangoes Cherries Total fresh fruit3 Vegetables and melons:4 Potatoes All lettuce Onions Tomatoes Watermelon Cantaloupe Carrots Sweet corn Cabbage Cucumbers Broccoli Spinach Snap beans Cauliflower Asparagus Total fresh vegetables and melons

Per capita consumption

Imported share of consumption

2005

1990

2005

25.1 16.7 11.7 8.6 5.8 4.9 4.8 3.5 2.9 2.9 2.6 2.5 2.2 1.9 0.9 100.5

24.4 19.6 12.8 7.8 3.2 2.0 5.5 1.4 3.2 2.6 4.4 1.3 0.7 0.5 0.4 92.5

99.7 7.0 4.6 54.9 7.1 87.7 11.0 40.9 21.3 9.7 4.0 29.0 100.0 100.0 8.1 45.6

1990 percent 99.8 4.7 0.9 37.0 4.0 49.0 8.0 10.5 12.5 3.6 0.9 11.4 53.0 97.4 0.2 34.9

42.4 31.6 21.0 20.2 14.0 9.8 8.8 8.8 8.1 6.3 5.6 2.3 1.8 1.5 1.1 215.6

46.7 31.5 15.1 15.5 13.3 9.2 8.3 6.7 8.3 4.7 3.4 0.8 1.1 2.2 0.6 188.2

6.3 1.8 11.0 35.0 15.9 32.8 7.5 2.2 4.8 50.9 11.0 3.7 11.3 4.2 72.2 16.9

5.9 0.0 10.1 20.5 6.9 23.0 5.9 0.9 4.2 33.7 2.5 1.5 11.2 4.0 29.8 9.9

pounds

5

NA = not available For citrus, the year reflects the end of the harvest; for noncitrus, the beginning of the harvest. 2 Trade numbers include nectarines. 3 Includes bananas. 4 ERS traditionally reports melons with vegetables. Consumption is on a calendar-year basis. 5 Does not include potatoes, sweet potatoes, or mushrooms. Source: Fruit and Tree Nut Yearbook, and Vegetable and Specialties Yearbook, ERS, USDA. 1

402

The Economics of Food Safety

403

Lbs per capita 2.5 2.0

Fresh 1.5

Frozen

1.0 0.5

Canned 0.0 1992

1994

1996

1998

2000

2002

2004

Figure 22.2. Changing spinach consumption pattern 1992–2005. Source: USDA— National Agricultural Statistics Service.

large change, with fresh per capita consumption increasing significantly from 1992 to 2005 (for example, see Fig. 22.2). Fresh-cut produce, including bagged salads, has become a more important part of the U.S. diet. The fresh-cut and bagging technologies are relatively new and the added convenience of washed and bagged spinach has probably contributed to the growth in fresh consumption. Consumers are eating a more varied diet. The typical grocery store carried 345 produce items in 1998 compared with 173 in 1987. Some of these items are new and exotic products, new varieties of more familiar products, or new formats, such as more bagged or ready-to-eat produce items. Spinach and leafy greens have also followed this pattern. A typical bag of spring mix may contain arugula, radicchio, mizuna, frisée, etc.—products that were unknown to most consumers in 1990. In addition to variety, consumers want produce on a year-round basis. Items that used to be available only seasonally are now imported to meet that year-round demand of consumers. In fact, this is not a new trend. In 2005, 46% of fresh fruit consumption was imported and 17% of fresh vegetable consumption was imported. In 1975, almost 22% of U.S. fresh tomato consumption was imported. Imported products help meet the demand for year-round fresh product, and they may also help dampen fluctuations in seasonal prices of fruits and vegetables. If the U.S. had to depend only on Florida tomatoes in the winter, consumers would face higher prices. Augmenting the winter tomato supply with imports from Mexico benefits consumers, but it does not benefit Florida producers. Although the general trend is toward an increasing share of imports in the total supply of fruit and vegetables in the marketplace, the role of import shares varies widely across the spectrum of products (Table 22.2) and depends on many factors including production possibilities (e.g., season), production costs including labor, transportation, and opportunities for storing product. Imports play a very small role

404

Section V. Public, Legal, and Economic Perspectives

in the spinach and leafy greens market. In 2006, only small volumes of fresh-market spinach and lettuce were imported: 3% of spinach for domestic consumption came from imports, 2% of head lettuce, and 1% of leaf and romaine lettuce. Despite several widely publicized cases, imports do not necessarily pose more of a risk than domestic products for food safety hazards, and no statistically reliable surveys are available to compare safety. Many U.S. buyers require the same food safety practices for imports as they do for domestic production. New production anywhere may be more problematic than production practiced by experienced growers. Food safety is a learning process and adjusting to local microbial risks may take time. In addition, many locations lack infrastructure (such as safe water supplies), and adopting food safety practices may be particularly challenging for foreign growers in such areas producing for the U.S. market (Dong and Jensen 2008). However, many well-established export industries in foreign countries have met this challenge, as evidenced by the significant share of product imported without incidents or food safety problems. The case of raspberries from Guatemala provides an example of a problem with imported produce (Calvin 2003). In the late 1980s Guatemala started to export fresh raspberries to the United States in the spring and fall. It was a new export crop for Guatemala and it filled a lucrative market niche between Chilean winter supplies and the beginning of U.S. summer production. Annual outbreaks in the U.S. and Canada from 1996 through 2000 were linked to Guatemalan raspberries contaminated with Cyclospora, a parasite that no one knew much about at that time. The FDA issued import alerts on fresh raspberries from Guatemala for several years. Extensive and costly efforts to improve food safety did not solve the problem, and 2003 was the last year with substantial exports of the product from Guatemala. Mexico, which began raspberry exports to the U.S. about the same time as Guatemala, has not been linked to any outbreaks and is now the largest supplier to the U.S. market, followed by Chile (Fig. 22.3).

Metric tons 6,000 5,000

Mexico

4,000 3,000 2,000

Chile

1,000

Guatemala

0 1990

1992

1994

1996

1998

2000

2002

2004

2006

Figure 22.3. U.S. import of raspberries from Guatemala, Mexico, and Chile, 1990– 2006. Source: U.S. Department of Commerce.

The Economics of Food Safety

405

Much of the U.S. produce industry is concentrated geographically in just a few states. For example, in 2006 California accounted for 51% of the value of fresh-market vegetables produced in the United States. Large firms that provide a year-round or extended-season supply dominate in these areas. Many smaller growers produce fruit and vegetables seasonally in other states, often for local markets. Spinach demonstrates this concentration. In 2005, 75% of the U.S. fresh-market supply of spinach grew in California (Fig. 22.4), with 54% of the U.S. total grown in the three adjacent counties of Monterey, Santa Clara, and San Benito. Arizona, Texas, New Jersey, Colorado, and Maryland combined to account for about 25% of the 2005 fresh-market spinach production. These statistics come from the USDA’s Agricultural Marketing Service, which records shipments just from the largest production area, unlike the U.S. Census, which records all production areas. Today, fresh fruit and vegetable products move quickly from producing regions directly or via market intermediaries to retail and food-service buyers (Calvin and others 2001). Retail consolidation has resulted in shipper consolidation. Retailers and food-service buyers do not want to deal with a large number of small shippers when a few larger shippers could supply their needs. As both retailers and shippers consolidate, it has become easier to specify desired production practices to obtain a uniform product. Some, but not all, of this coordination has been achieved through increased use of contracting and vertical integration within the marketing and procurement channel. Industry structure has several implications for food safety. It is generally easier to get major players to cooperate in a food safety initiative when production is concentrated in a state. Crossing state lines may involve more types of growers, different production practices, and less agreement on issues. The LGMA illustrates these challenges. The voluntary marketing agreement in California had virtually 100% participation in its first year. Arizona also initiated a similar marketing agreement. Many of

Texas, New Jersey, Colorado, and Maryland Arizona

9% 16% 54%

Rest of California

21%

California counties of Monterey, Santa Clara, and San Benito

Figure 22.4. U.S. fresh-market spinach production, 2005. Sources: USDA—National Agricultural Statistics Service and California Agricultural County Commissioners’ data.

406

Section V. Public, Legal, and Economic Perspectives

the large California shippers also operate in Arizona. However, when the U.S. Department of Agriculture (USDA) issued a call for comments on the idea of having a national marketing agreement or marketing order, many smaller producers protested due, in part, to the expected high costs of complying with such an agreement. Consolidation in the produce industry means that any food safety problem that affects one grower or shipper may affect a large number of consumers, and be more likely to be detected. The capital-intensive bagged salad industry is particularly concentrated. In 1997, the top two firms accounted for 76% of the retail sales of bagged salads, the top five firms accounted for 88% of sales, and private label firms accounted for 10% of sales (Calvin and others 2001). At the same time, large firms may have the financial resources and the volume of sales to adopt some of the food safety practices, which may be relatively expensive for smaller companies.

Government and Industry Response to Food Safety Problems In the mid-1990s, outbreaks of foodborne illnesses linked to microbial contamination of both domestic and imported produce focused attention on the potential for contamination at the farm level. In 1996, E. coli O157 : H7 was linked to California lettuce associated with farm-level contamination. This was in addition to the foodborne illness outbreak linked to imported Guatemalan raspberries, also contaminated at the farm level. The economic impacts of the outbreaks made it clear to the produce industry, particularly those sectors associated with the contaminated product, that improved food safety programs were necessary. The U.S. government also became more involved in produce food safety at the farm level. The appropriate regulatory approach to promote food safety depends on both the type of product and the hazard. The government can either regulate the product or the production process (Unnevehr and Jensen 2005). When monitoring the quality of products is feasible, product standards are likely to be more efficient than process standards because they allow firms to meet the minimum quality or tolerance levels but choose the least expensive method to do so. In addition, since the potential for contamination may vary across farms, locking all growers into the same process standard may not be appropriate. However, in instances when determining product quality is difficult or very costly, requiring certain processes to be followed provides an effective strategy for reducing product risk. Two characteristics of fresh produce work against product standards for microbial contamination. First, it is very hard to detect microbial contamination on produce. The FDA product standard is zero tolerance for microbial contamination, but this is largely unenforceable. Microbial contamination on produce can be difficult to detect. In contrast, testing for pesticide residues is relatively efficient. If a field is sprayed with too much pesticide, any produce from that field will turn up positive for excessive residues. Microbial contamination on produce can be at low levels and occur sporadically. Only a small section of a field, or even just one leaf of lettuce, may be contaminated, and the chance of detecting that contaminant in random testing is low. Second, there is no generally approved methodology for removing microbial contamination from fresh produce, so the most effective strategy for reducing the risk of microbial contamination is prevention. Until recently, the FDA had approved irradiation of fresh

The Economics of Food Safety

407

produce only for controlling quarantine pests (e.g., irradiating mangoes for fruit fly). In 2008, the FDA approved irradiation of leafy greens to reduce microbial contamination, which may provide an alternative strategy for this segment of the produce industry. The FDA does not have a process standard, but the produce and buying industries are developing process standards. In 1998, the FDA published voluntary guidelines, GAPs, to help growers, both domestic and foreign, reduce the risk of microbial contamination at the farm level. The FDA specified particular areas of risk (water, manure, and municipal biosolids, worker health and hygiene, sanitary facilities, field sanitation, packinghouse sanitation, transportation, and traceback) that should be addressed but did not specify particular production practices. GAPs are guidance, not a process standard. This was a general, common-sense guide, because, at that time, there was very little specific research to provide more concrete advice. Over time, there has been a push to develop more specific guidelines. In 2004, the FDA and the Centers for Disease Control and Prevention (CDC) met with produce industry leaders to discuss numerous foodborne illness outbreaks associated with produce. At that meeting, industry representatives agreed to take the lead on developing commodity-specific GAPs that would provide additional guidelines tailored to individual commodities that had been implicated in recent foodborne illness outbreaks. After the 2006 outbreak linked to spinach, the California leafy greens industry developed a set of Best Practices (also known as the metrics) that would become the standard for the LGMA. This is a process standard required for all participants and verified with mandatory audits; it is not guidance. Unlike the FDA’s initial guidance document and the commodity-specific GAPs, the new Best Practices defines practices with specific criteria and target values for controls and monitoring. As an example, the original GAPs document warned farmers that “water quality should be adequate for its intended use.” At the time, the FDA was justifiably reluctant to specify what adequate water quality was because it did not have enough data to support specific thresholds. The new Best Practices are much more specific, but the science is still relatively weak. For example, the standards for well water require testing before production begins and monthly testing during the production season. The document recommends specific tests for measuring levels of generic E. coli in the water and an action plan to be applied if counts reach certain numerical thresholds. However, the effectiveness of these practices under different growing conditions as well as the costs are not well understood. There have been some industry and consumer calls for the FDA to step in with mandatory process standards. Although the FDA does not currently have food safety process standards with respect to microbial contamination for the fresh produce industry, it could impose mandatory food safety standards if deemed necessary. However, developing process standards without adequate scientific support could undermine public confidence in food safety regulators if another outbreak occurred. The FDA has imposed mandatory standards on fruit juice. In the late 1990s there were three foodborne illness outbreaks associated with unpasteurized fruit juice. In 1998, the FDA published a rule requiring juice processors to use Hazard Analysis and Critical Control Point (HACCP) principles to reduce risk. Processors are required to use processes that

408

Section V. Public, Legal, and Economic Perspectives

achieve a 5-log reduction in the numbers of the most resistant pathogen in their finished products, compared to levels that may be present in untreated juice. In most cases, this level of reduction is achieved with pasteurization. There is no corresponding easy solution for fresh produce. There is a proliferation of other process standards. Many private firms (growers and buyers) have developed their own process standards. Buyer organizations have also developed process standards. The Food Marketing Institute holds the U.S. licensing rights for Safe Quality Food (SQF), an Australian process standard. The Food Safety Leadership Council proposed its own process standard but faced protests from the produce industry. The future of that specific process standard is uncertain. As a result of the explosion of standards, growers are paying for numerous third-party audits of food safety practices, which can lead to a considerable expense.

Economics of Adoption of GAPs and the LGMA GAPs are now an important part of the produce industry in the United States and countries that export to the U.S. The private third-party audit industry has taken the guidelines and developed audits to certify whether growers are complying with the FDA guidelines or any other guidelines or process standards that a grower or buyer might use. Foodborne illness outbreaks related to produce continue. Either growers are not using GAPs or are not using them correctly and consistently, or GAPs do not sufficiently target the relevant risks. The conventional wisdom suggests that large producers in high-risk commodities use GAPs, but there is no statistical evidence to support this assumption for the U.S. fresh produce industry. A survey of green onion exporters in Mexico (the major source of green onions consumed in the U.S.) in 2002, the year before the large outbreak linked to Mexican green onions, found that three of seven growers had already adopted GAPs and two of seven growers were in the process of doing so (Calvin and others 2004). Although the survey’s sample size is small, the industry was concentrated with just 26 growers in 2002. None of the growers associated with the 2003 outbreak linked to Mexican green onions had third-party audits of GAPs. After the outbreak, Mexican growers developed a mandatory food safety program, so that the practices of a few growers could not undermine the reputations of others. Many California leafy greens growers already had GAPs before the 2006 outbreaks linked to spinach. With repeated outbreaks traced back to leafy greens, it is thought that most large retailers and food-service buyers were demanding GAPs from their suppliers. Also, because there are very large bagged salad producers with consumerrecognized brand names, these producers would have more incentive to adopt GAPs than others who were not invested in maintaining a brand name. Private Benefits and Costs When individual growers consider whether to adopt additional voluntary food safety practices, they weigh their private benefits and costs. Typically, growers adopting a new production practice expect to either receive a higher price (i.e., price premium) for a higher-quality good, reduce risk, or lower their costs of production. In the case

The Economics of Food Safety

409

of adopting food safety practices, growers have generally not been able to charge a higher price for a product grown with more attention to food safety (discussed in more detail below). Other benefits may influence growers’ decisions to adopt better food safety practices. These benefits are mostly related to reducing risk. Growers want to reduce the probability of an outbreak being traced to their firm that could cause lost sales, damage to reputation and brand name, costly lawsuits, etc. These benefits accrue only in the event of an outbreak. Before an outbreak occurs, some growers may think that the probability of experiencing these benefits is very low. A more immediate benefit of adopting better food safety practices is that it satisfies many retailers and food-service buyers who require third-party audits of grower food safety practices as a precondition of purchase. Having higher food safety practices gives growers broader market access, which is an important competitive advantage and incentive to adopt GAPs. A primary factor weighing against the potential benefits of adopting new food safety standards is the costs, which are immediate and often large. Growers want to reduce the risk of outbreaks, but unless the contamination mechanism is understood, it is not clear whether additional practices will reduce risk or just raise costs. The lack of relevant science may limit opportunities to control risk (Hennessy and others 2003). Perfect safety for products grown in a natural environment is not attainable (the FDA acknowledges that no practices guarantee perfect safety) and a grower could go broke trying to approach that elusive goal. Costs of adopting new food safety practices include both recurrent and nonrecurrent costs. Nonrecurrent costs may entail investments in water quality and waste management infrastructure, harvest machinery, and packinghouse facilities. Recurrent costs of compliance may include higher costs for water, water testing, training workers in hygiene in the field, upgrading data collection and record keeping systems, etc. In the case of the new California LGMA, several new practices are expected to be quite expensive, such as water testing, record keeping, and buffer zones. Before the outbreak, some growers had already adopted some of these new practices, and their costs of adopting the rest will be lower than the costs for growers who had more limited food safety programs and have to adopt all the safety practices at the same time. Costs will also vary among farmers. Smaller farms may find the costs of record keeping and related training difficult to meet as they try to spread the cost over a lower production base. The buffer zone costs will vary by location, with farms in outlying areas with more risk of wild animal intrusion or near cattle operations needing to consider more remedial actions than growers surrounded only by other leafy greens fields. Some retailers and food-service buyers are requesting additional practices such as final product testing, which will also raise the costs of being a leafy greens grower in California. Growers adopt new food safety practices if expected benefits exceed expected costs. However, not all growers make the same decisions with respect to adopting more food safety practices. Even among growers of the same crop, benefit-cost analyses upon which decisions are based can vary depending on characteristics of the growers and their operation. Early adopters have more choices. At some point, new practices become the industry standard, and those who did not adopt early in the process must finally adopt if they want to remain competitive.

410

Section V. Public, Legal, and Economic Perspectives

A market that accommodates separate prices for products grown with different levels of food safety practices with respect to microbial food safety has not developed. Because growers cannot necessarily recoup the costs of increased food safety, not all adopt new practices. The outbreak linked to spinach provides an interesting example of this problem. Other recent outbreaks have largely dealt with bulk commodities. The leafy greens industry has both bulk and value-added bagged salad products, and each segment has fared differently in its ability to increase prices to cover additional costs of improved food safety practices. Several large firms of bagged salads raised their prices after the outbreak from spinach. Consumers might miss their favorite salad if a retailer refused to pay the price increase. Also, with such a high level of concentration, if a very large retailer rejected a request for a higher price from a large bagged salad company, the retailer might have a hard time getting enough replacement product. Bagged salads are largely sold via long-term contracts, and if firms did not raise their prices after the outbreak, they tried to raise them when their contracts were next renewed. It is not clear whether smaller companies were as successful as the larger companies in renegotiating contract prices. With bulk leafy greens, only a small portion is sold via long-term contracts and firms are gradually trying to raise prices as contracts are renewed, just as the bagged firms did. But the majority of bulk leafy greens are sold at the daily market price where price is set by supply and demand; there has been no price increase for bulk spinach or lettuce where one grower ’s product is essentially indistinguishable from another ’s. Of course, in the long run, prices must rise to cover costs, or growers will stop growing leafy greens, although this can sometimes be a slow adjustment process. Immediately after the outbreak, many retail and food-service buyers joined a call for better food safety standards in the leafy greens industry. When the California industry put the LGMA into operation, they asked buyers to agree to buy only leafy greens from California from signatories to the agreement. With only a few exceptions, buyers generally refused to sign such an agreement. But if a large share of buyers had signed such an agreement and LGMA participation fell substantially, there could be a situation where there would be two markets, with one price for products grown with the LGMA and one price for those grown outside the agreement. The price in the segment grown with the LGMA would vary depending on supply and demand conditions in that segment alone. Buyers pledged to purchase only from LGMA members would not be able to buy from the others while honoring their agreement. The segment grown outside the LGMA would respond to its own supply and demand conditions as well as those in the LGMA group, with buyers able to buy from any supplier. Public Benefits and Costs The decisions individual growers make about food safety practices may not ensure the level of food safety desired by consumers and society at large (Caswell and Mojduszka 1996). Markets do not always work smoothly for all goods. Private decisions by growers may not be socially optimal because of imperfect information and negative externalities. Imperfect information, which exists when buyers and sellers cannot identify certain characteristics of a product, may reduce the incentives to adopt new food safety

The Economics of Food Safety

411

practices by hindering the development of different prices for different levels of food safety. In the past there was never a separate spot-market price for produce grown with GAPs and produce grown without GAPs, even when imperfect information is reduced with the use of third-party audits. Perhaps food safety is an inherently different quality than other unobserved characteristics. For example, organic production, an unobserved characteristic, enjoys a price premium over conventional production. But organic is something beyond and above conventional production. Consumers may feel that only safe food should be offered and are unwilling to pay extra for that characteristic. Or retailers and food-service buyers may not be willing to pay more for food produced with more food safety attention. Advertising product from one producer as being safer may be a risky strategy for a retailer or food-service firm because in most cases it is not actually possible to guarantee food safety. Retailers often have to change suppliers if there are weather problems in particular areas. What if that product was produced with less attention to safety? A retailer or food-service buyer would hardly want to advertise that fact. Also, advertising differences in safety among sources of a particular produce item may provide consumers with information that undermines their confidence in the product in general, regardless of the food safety claims, especially if consumers never knew that kind of contamination was possible on produce (Golan and others 2004). Negative externalities also affect the incentives to adopt additional food safety practices. Negative externalities occur when one party’s production or consumption choices have a negative impact on another party’s well-being. Society as a whole may demand higher levels of food safety than consumers in grocery stores or food-service establishments. Of course, in the event of an outbreak of foodborne illness, consumers are on the front line facing health problems and medical bills, lost days of work, etc. Everyone along the marketing chain associated with the contaminated product will face potential costs. Even those not directly associated with contaminated product may suffer. For example, if a foodborne illness is traced to a particular product, but not a particular grower, all producers of that food item may feel the effects of decreased demand, as shown by the drop in shipments and fall in price of fresh bunched spinach shipments following the FDA announcement in September 2006 (Fig. 22.5). The CDC and FDA incur substantial costs in tracing back the outbreak to the contaminated product. They also investigate farm and packinghouse operations and review inspection results. Some level of government often ends up paying for many of the medical costs incurred in an outbreak. In their private benefit-costs analyses, growers do not consider the benefits and costs that others might incur if food safety were improved, and may, therefore, provide less food safety than society desires. When there are outbreaks of foodborne illness, other groups in the produce industry, marketing chain, or government may face increased costs and may try to impose new rules on growers to encourage or force them to implement food safety measures more in line with society’s total demand for food safety. Essentially, the costs of organizing to bring about the changes have decreased. For example, grower organizations may put into place voluntary or mandatory practices to reduce the negative impact of one producer with contaminated produce on other growers of the same product. The LGMA is an example of this type of response. Retailers and food-service buyers may require third-party audits showing grower compliance with GAPs and

412

Section V. Public, Legal, and Economic Perspectives

Price $/carton

Shipments Tons

10

180

9

160

FDA announcement

8

140

7 120

Price

100

6 5

Shipments

80

4

60

3

40

2

20

1

0

0

4

11

18

25

September

2

October

Figure 22.5. Fresh bunched spinach shipments: September–October, 2006. Source: USDA—Agricultural Marketing Service.

packinghouse compliance with good manufacturing practices (GMPs) to reduce the chance that their businesses will be associated with an outbreak.

Economic Impacts of the Spinach Outbreak After the FDA’s announcement on September 14 not to eat bagged spinach (followed the next day by the announcement not to eat any fresh spinach) there was no U.S. spinach on the market for 5 days until after the FDA announced on September 19 that all spinach from outside California was safe to eat. Figure 22.5 shows a timeline of bulk spinach shipments in September–October 2006 (USDA has shipment data only for bulk spinach, which is estimated at 10–25% of total fresh-market spinach). Data do not show shipments from outside of California that might have resumed after September 19. On September 22, the FDA announced that spinach from California, except Monterey, Santa Clara, and San Benito, was safe, and small sales resumed. On September 29, the FDA announced that “spinach on the shelves is as safe as it was before this event.” At that stage there were no restrictions on any spinach, except for the four fields that the FDA was still investigating, and bulk spinach sales began to grow slowly. Retail sales data show a more complete picture of the impact on the sector ’s sales and loss in economic value (Table 22.3). Retail data are available from Information Resources, Inc. and FreshLook Marketing. The data cover sales by major retailers but not “big box” stores. The food-service market is very important for leafy greens but no data were available. Members of the leafy greens industry reported that the foodservice market recovered faster than the retail market. In 2005, the year before the

The Economics of Food Safety

413

Table 22.3. Leafy greens retail market shares and changes in sales Change in Sales Quantity Commodity

Romaine hearts Spinach in bags1 Bagged spring mix Salad without spinach Bulk romaine Bulk iceberg lettuce Bulk leaf and other bulk lettuce Bulk spring mix Bulk spinach2 All leafy greens All other vegetables

Share of Leafy Greens Sales 2005 percent 7 5 2 47 6 24 7

2004–05

2005–06 (Jan–Aug)

2005–06 (Sep–Dec)

2005–2007

13 7 6 1 0 −3 −4

10 8 22 −6 0 −6 −5

10 −49 −14 −8 14 −4 7

7 −13 13 −9 −3 −11 −5

1 1 NA NA

−4 −9 1 3

−7 −2 −3 0

−15 −44 −6 −1

−14 −21 −7 3

NA = not applicable. Does not include bagged spring mix. 2 Does not include bulk spring mix. Source: IRI and FreshLook. 1

outbreak, products with spinach accounted for only 9% of the total volume of leafy greens sales. Bulk spinach sales accounted for only 1% of the retail sales volume (in pounds) and its share was declining. Spinach in bags represented 5% of spinach and lettuce product sales. This category includes bagged salads, both those containing just spinach and those containing a spinach-lettuce blend, as well as bags of spinach that might be intended for cooking. Bagged spring mix, which usually contains spinach, represented another 2% of the market. Bulk spring mix accounted for about 1%. Between 2004 and 2005, sales of spinach in bags grew 7%. Spring mix in bags was growing rapidly. It was the third fastest growth item in 2005, after romaine hearts and spinach in bags, and was up 6% over the previous year. In the first 8 months of 2006, spring mix was the most rapidly growing category, up 22% from the same period in 2005. Bulk spinach and bulk spring mix were declining in sales as were all bulk lettuces. Lettuce shows a very similar pattern to spinach, with all bulk products declining in share and all value-added products increasing in share. Although bagged salads without spinach accounted for 47% of total spinach and lettuce sales in 2005, it was not growing very much. Romaine hearts (value-added product often sold in bags) were the most important growth area for lettuce. Iceberg was the largest category of bulk lettuce, followed by leaf and other lettuces, and then bulk romaine. Evaluating the last 4 months of 2006 provides a better view of the impact of the outbreak. In the last 4 months of 2006 after the FDA announcement, sales of all products containing spinach plummeted compared to the same period in 2005. Spinach

414

Section V. Public, Legal, and Economic Perspectives

in bags, the product that was implicated in the outbreak, was down 49% and had the largest decline. Bulk spinach was down 44%. Bagged and bulk spring mix sales were down 14% and 15%, respectively. The smaller impact on spring mix may be because consumers were unaware that this product usually contained spinach or they may have responded to the leafy greens industry taking spinach out of the mix. Other lettuce products were also affected by the problems associated with spinach, most notably the increase in sales for bulk romaine, bulk leaf, and other bulk lettuces. These are the only two categories that grew in sales. Also, with the outbreaks of foodborne illness associated with iceberg lettuce in December, they were the only categories of lettuce untouched by food safety shocks. The growth in romaine hearts sales did not change in the aftermath of the outbreak. Consumers may have been more concerned about romaine in bags than bulk romaine. The impact of the outbreak on spinach has been quite long-lasting. Figure 22.6 shows sales of spinach in bags from 2005 to 2007. 2006 sales were above the 2005 levels until the September 14, 2006 announcement by the FDA when they plunged immediately. At the end of 2007, over 15 months later, sales volume still lagged behind the levels of 2005. It is not clear whether there is a permanent shift in consumer demand for spinach or whether consumers are still adjusting to the shock and may eventually buy more bagged spinach and continue preshock trends. Producers also cut back on spinach acreage with total U.S. fresh-market spinach production down 16% from 2005 to 2007.

Million lbs. 2.50 2.00 1.50 1.00 0.50 0.00 1

10

20

30

40

2006

2007

50

Weeks 2005

Figure 22.6. Spinach in bags; retail sales 2005-2007. Source: IRI and FreshLook Marketing.

The Economics of Food Safety

415

Research on the observed market impact of outbreaks is limited. In the case of the 2003 outbreak of hepatitis A associated with green onions from Mexico (the major supplier of green onions in the U.S. market) shipments lagged behind the previous year ’s level for about 5 months (Calvin and others 2004). Research on strawberries in the 1990s showed that consumption is affected more by negative news than positive news after an outbreak (Richards and Patterson 1999). In the fall of 2006, the leafy greens industry faced three different batches of bad food safety news after the spinach problem. One recall turned out to be a false alarm. In December, two outbreaks in fast-food restaurants were linked to lettuce. In 2007 there was another false alarm and two recalls of contaminated products but no illnesses were reported in either case. Outbreaks can also have an impact on U.S. export markets. Canada is the largest export market for U.S. leafy greens. In the case of the 2006 spinach outbreak, Canada briefly blocked imports of spinach from the U.S. Even after the market reopened, trade was low (Fig. 22.7). Canadian consumers, like U.S. consumers, were probably less likely to consume spinach after all the adverse publicity. Volume over the last 4 months of 2006 was down 49% from the previous year. Beginning in June 2007, Canada started limiting imports of leafy greens from California to signatories of the LGMA. Although Mexico is a very tiny market for U.S. leafy greens it also followed the Canadian example, first briefly banning imports of leafy greens and now limiting California imports to those from members of the LGMA.

Metric tons 3,000 2,500 2005 2,000 1,500 1,000 500

2006

0 Jan

Mar

May

Jul

Sep

Nov

Figure 22.7. U.S. fresh spinach exports to Canada, 2005–2006. Source: U.S. Department of Commerce.

416

Section V. Public, Legal, and Economic Perspectives

Conclusions The market for fresh produce today includes a wide range of products available yearround. An increasing share of the products are consumed fresh and not processed. Many of the products are imported at some time during the year. Several wellpublicized cases of foodborne illnesses illustrate the food safety hazards with fresh produce. Raspberries from Guatemala, green onions from Mexico, and spinach/leafy greens from California all show the potential extent and magnitude of harm caused by breakdowns in food safety. The contamination of fresh spinach with E. coli O157 : H7 in September 2006 led to over 200 people becoming ill and a shutdown of the spinach industry for a short period of time. These cases all show the potential for high costs in terms of public health and to the industry from failure to control the hazards. We draw three main conclusions and observations about economic conditions and factors that affect food safety in fresh produce. First, in the case of the spinach and leafy greens industry, the concentration of the major production in a relatively limited number of states and the consolidation of major growers in the marketing and procurement channels facilitated the organization of growers developing and enforcing a voluntary industry agreement through the LGMA on the use of Best Practices in production and processing. Where such voluntary agreements may be more difficult to develop, individual growers will continue to determine their own level of food safety. A second conclusion is that the distribution of costs of controlling food safety in the system can have an important impact on the structure of the industry. With new concerns about safety, some areas or sizes of firms may not be as competitive as they once were. For example, certain areas may have more problems controlling water quality. Other environmental conditions in some areas may make production more prone to food safety problems. Also, economies of scale in certain processing technologies or practices associated with reduced product risk would favor large firms. In an extreme case, the inability of the Guatemalan raspberry producers to control the Cyclospora contamination problem led to the demise of their industry in the U.S. marketplace. This problem gave the raspberry industry in Mexico a competitive edge in the U.S. marketplace, at least temporarily. Finally, increasingly integrated global markets for fresh produce require that suppliers and buyers make and receive assurances of food safety practices. Use of thirdparty audits of various food safety practices is increasing among U.S. growers and foreign growers who produce for the U.S. market. As illustrated by the 2006 food safety outbreak related to spinach, the effects of an outbreak on an industry can be significant. To date, the food safety outbreaks for leafy green produce have hastened and encouraged the ongoing adoption of safety-related practices and technologies. With so much at stake for the health of the consumer and the industry, it is important that science lead the way in identifying good practices to reduce the risk of microbial contamination and ensure the cost-effectiveness of new investments.

Acknowledgments Linda Calvin is an agricultural economist with USDA’s Economic Research Service. Helen H. Jensen and Jing Liang are professor and doctoral student, respectively, in

The Economics of Food Safety

417

the Department of Economics and Center for Agricultural and Rural Development at Iowa State University. The views expressed herein are those of the authors, who do not necessarily reflect official policy of the USDA or any other federal agency or government entity.

References Calvin L. 2003. Produce, food safety, and international trade: response to U.S. foodborne illness outbreaks associated with imported produce. In: Buzby J, editor. International Trade and Food Safety. USDA, ERS, AER No. 828, Nov. (available at http://www.ers.usda.gov/publications/aer828/aer828g.pdf, and accessed Sept. 26, 2008). ———. 2007. Outbreak linked to spinach forces reassessment of food safety practices. Amber Waves, USDA, ERS, June (available at http://www.ers.usda.gov/AmberWaves/June07/Features/Spinach.htm, and accessed Sept. 26, 2008). Calvin L, Avendaño B, Schwentesius R. 2004. The economics of food safety: the case of green onions and hepatitis A outbreaks. USDA, ERS, E-Outlook, VGS-305-01, Dec. (available at http://www.ers.usda. gov/publications/vgs/nov04/vgs30501/vgs30501.pdf, and accessed Sept. 26, 2008). Calvin L, Cook R, Denbaly M, Dimitri C, Glaser L, Handy C, Jekanowski M, Kaufman P, Krissoff B, Thompson G, Thornsbury S. 2001. U.S. Fresh Fruit and Vegetable Marketing: Emerging Trade Practices, Trends, and Issues. USDA, ERS, AER Rpt. No. 795, 2001 (available at http://www.ers.usda.gov/ publications/aer795/, and accessed Sept. 26, 2008). Caswell J, Mojduszka E. 1996. Using informational labeling to influence the market for quality in food products. American Journal of Agricultural Economics 78:1248–53. Dong F, Jensen H. 2008. Sanitation and hygiene deficiencies as contributing factors in contamination of imported foods. In: Doyle M, Erickson M, editors. Imported Foods: Microbiological Issues and Challenges. Washington, D.C.: ASM Press. p 139–58. Golan E, Roberts T, Salay E, Caswell J, Ollinger M, Moore D. 2004. Food safety innovation in the United States: evidence from the meat industry. USDA, ERS, AER Rpt. No. 831. Hennessy D, Roosen J, Jensen H. 2003. Systemic failure in the provision of safe food. Food Policy 28:77–96. Richards T, Patterson P. 1999. The economic value of public relations expenditures: food safety and the strawberry case. Journal of Agricultural Research Economics 24:440–62. Unnevehr L, Jensen H. 2005. Industry costs to make food safe: now and under a risk-based system. In: Hoffmann S, Taylor M, editors. Toward Safer Food: perspectives on risk and priority setting. Washington, D.C.: Resources for the Future. p 105–28.

Section VI Research Challenges and Directions

23 Research Needs and Future Directions Brendan A. Niemira, Christopher J. Doona, Florence E. Feeherry, Xuetong Fan, and Robert B. Gravani

The recent produce-associated outbreaks demonstrate the critical need for increased research in multiple areas to ensure the safety of fresh produce. Unquestionably, the food safety challenges facing the growers, packers, processors, retailers, and consumers of fresh and fresh-cut produce are complex and multifaceted. Although established and ongoing research projects have provided insights on produce contamination at multiple steps in the supply chain, the goals of future research activities are to develop science-based intervention strategies that minimize the risks of potential contamination and strengthen the safety of fresh fruits and vegetables. Produce safety research grants funded by both the private (FEPSRI 2007) and public sectors (NRI 2008) have identified four key research areas to be pursued: internalization of pathogens into produce, interventions, vectors, and environmental risk factors. There is particular interest in the microbial ecology of pathogens, including their interactions with nonpathogenic microflora and identifying their routes of contamination (Gourabathini and others 2008; Cooley and others 2006). In the context of produce safety, risk reduction can take one of three forms: prevention (to include microbial ecology), containment, and eradication. The most common postharvest intervention for produce, washing in chlorinated water, does not completely eliminate pathogens from the surface of produce. The general lack of a commercially acceptable kill step has limited the response options for growers, packers, and processors. Irradiation was recently approved by the FDA for use on lettuce and spinach to inactivate pathogens and extend shelf life (FDA 2008). Irradiation, alternative sanitizers including ozone, electrolyzed water, chlorine dioxide, peroxyacetic acid, and other nonthermal food-processing technologies, such as High Pressure Processing (HPP), biocontrol, etc., may provide additional tools to ensure the safety of fresh produce and protect public health. The fresh produce industry is very interested in practical and useful interventions that can be readily applied in the field, packinghouse, or processing plant (Gombas 2008). It is important to provide research answers to the most pressing questions, to develop solutions that are relevant to the industry needs, and for scientists in academia and government to collaborate with industry partners on these mitigation strategies. Intervention technologies or treatments developed in the laboratory often do not result in effects of the same magnitude when applied in commercial operations. Conducting studies in actual commercial settings may involve the introduction of surrogates to fresh produce in the field, packinghouse, or processing plant. Alternatively, controlled growing chambers, such as greenhouses or pilot-scale processing facilities may be developed to allow closer study of human pathogens in realistic settings. 421

422

Section VI. Research Challenges and Directions

In recent years, the produce industry has held several high-level meetings to prioritize produce safety research. At these meetings, producers, processors, retailers, regulatory agency officials, and academics identified and assessed the key research gaps in our current state of knowledge and addressed barriers to improving the safety of fresh and fresh-cut produce. Two notable commodity-specific meetings held in 2007 included one focused on tomato safety research needs and hosted by the Joint Institute of Food Safety and Applied Nutrition (JIFSAN), with the University of Florida, Institute of Food and Agricultural Sciences (IFAS) in College Park, MD, (JIFSAN 2007), and a second focused on leafy greens safety research needs, and hosted by the United Fresh Produce Association in Herndon, VA, (UFPA 2007). By drawing on the collective expertise, knowledge, and experiences of the attendees, a prioritized list of recommendations that emphasized research efforts with high value and broad applicability was developed. So, with the conclusions and recommendations of these meetings in mind, a brief summary of the information presented in the preceding chapters provides the foundation for fresh fruit and vegetable safety research needs.

Prevention and Microbial Ecology The epidemiology of several produce outbreaks suggests that focusing exclusively on prevention strategies is insufficient in dealing with the full range of problems associated with produce contamination. Before contamination can be prevented, the avenues of contamination must be more fully understood. A clearer understanding of the ecology of human pathogens in the field (Ch. 1, 2), on the surfaces of fruits and vegetables (Ch. 3), and on food contact surfaces (Ch. 18) is required. This understanding also provides a basis for improved practices and infrastructure in fields (Ch. 4–8), in packinghouses, in processing plants (Ch. 16–18), and at the point of sale (Ch. 15). The following are some of the critical questions pertaining to prevention: • What are the physical, cultural, and economic barriers that have prevented adoption of available risk control measures such as good agricultural practices (GAPs) on farms? • What soil, water, climatic, or environmental factors allow human pathogens to persist in and near fruit and vegetable production fields? • What are the reservoirs for pathogens in the production environment? • Are current recommendations for setbacks and buffer zones scientifically determined and verifiably adequate? • Are some strains of pathogens more likely to be associated with certain fruits and vegetables? If so, can this association/relationship be used to develop new control strategies? • How long can pathogens survive and grow in agricultural soils and on crop plants? Does survival on associated weed plants increase risk for fruits and vegetables? • What aspects of crop production (tillage, chemical inputs, soil amendments, crop rotations, etc.) most directly influence this residence time? • What are the most significant animal vectors for introducing pathogens to fruits and vegetables in the production environment? • Do insect vectors contribute to pathogen transmission in the field?

Research Needs and Future Directions

423

• What are effective exclusion measures for feral animals? • What new animal husbandry practices can be used in livestock operations to reduce microbial hazards and risks associated with raw manure? • Under normal growing conditions, is internalization via the roots a significant risk factor in the field? If so, how does this compare to surface contamination and also to known internalization risks via wounds, stomata, etc., that can occur during harvesting, sorting, washing, processing, and packaging? • Are current irrigation water quality standards adequate to prevent pathogen contamination of fruits and vegetables? What microbial water quality standards should be recommended for irrigation water delivered to plants by surface or overhead methods? Can water treatment procedures be implemented in-field? • What are the transfer coefficients of microorganisms in water, soil, crops, equipment, and food contact surfaces? • What is the best way to balance the needs for meeting new food safety rules with environmentally sustainable, best land management practices? • How can compost verification programs be improved? • What improvements in sanitary design and sanitation can be made to produce handling equipment in fields, packinghouses, processing operations, and in the cold chain? • How do the various risk factors compare in significance and level of reduction? • What are the components of a comprehensive risk reduction model for any given produce commodity? • Can cold-chain monitoring and compliance be improved with new wireless sensor technologies? • What are the risk factors in retail food stores that can be addressed?

Containment The containment of pathogens has two aspects: 1) rapid and accurate testing and detection (Ch. 16–18) and 2) developing appropriate response plans when contamination is detected. Well-developed action plans are required to deal with pathogencontaminated produce in the short (Ch. 6, 19), medium (Ch. 20), and long term (Ch. 21, 22). The following are some of the critical questions pertaining to containment: • How can fields be tested and certified prior to planting, particularly after potential contamination events occur? • Is detection of aerial contamination from windblown manure or dust a practical or meaningful risk reduction approach? • How can microbial sampling of equipment be improved? • How can the transmission of pathogens from farm, packinghouse, retail food store, and food-service workers be reduced? • What seasonal or environmental factors would trigger enhanced monitoring in the field? • What are the parameters of a testing program (methodology, sampling plan, frequency of testing, sample size, etc.) that would adequately detect pathogen

424

• • • • •

•

Section VI. Research Challenges and Directions

contamination on produce in the field? During sorting, packing, processing, and in the finished, packaged product? What role can in-field mobile testing labs play in improving speed, efficiency, and accuracy of microbial testing? How can detection and testing tools developed for food safety be adapted for use in agroterrorism or food defense scenarios? What monitoring and testing program would detect amplification of pathogens between harvesting, sorting, and processing or packaging? Can accurate and rapid testing methods be developed for products with a short shelf life? Can communication and information sharing among growers, processors, retailers, and regulators be improved? Are current communication and information-sharing procedures adequate for produce recalls, traceback, and source identification? Are current labeling practices sufficient to facilitate traceback and recall of produce commodities? Can Radio Frequency Identification (RFID) or other enhanced technology labeling be used as an effective risk reduction tool?

Eradication Changing industry standard practices to eradicate pathogen contamination will include improved cleaning and sanitizing procedures as well as other chemical treatments (Ch. 9), improved thermal and nonthermal physical treatments (Ch. 10, 12, 13, 14), and advanced research on biological control measures (Ch. 11). The following are some of the critical questions pertaining to eradication: • Can current processing systems be modified to reduce risks of cross-contamination? Can water flumes be replaced by belts, air beds, or other systems? • Can aqueous chemical treatments be improved by modification to gas-phase treatments? • Can a combination of treatments, similar to hurdle technologies used in other foodprocessing applications, be used to reduce pathogen contamination of fruits and vegetables? • Can the tools of molecular biology from the study of microbial ecology and biofilm formation also be used to develop new interventions that inhibit or prevent pathogen attachment to fruits and vegetables? • Can plant pathology and plant breeding tools be used to develop fruits and vegetables that are “resistant” to human pathogens? • How can new nonthermal processes (pulsed light, pulsed UV, high pressure processing, cold plasma, radio frequency treatment, etc.) be adapted for use with produce? • Are biological controls suitable for use in controlling or eliminating human pathogens? Can phage technology be used as a pre- or postharvest intervention strategy? • What are the barriers to the adoption of irradiation to produce other than leafy greens? How can these be addressed? • Aside from irradiation, are there additional antimicrobial processes that are effective against protected pathogens, such as those that are internalized or that are associated with biofilms?

Research Needs and Future Directions

425

• How are various intervention methods compared? What standards or metrics can be used to verify the levels of risk reduction achieved through the interventions that are used? • What nonpathogenic organism(s), i.e., surrogate(s), would be a verifiable indicator for E. coli O157 : H7? For Salmonella? • Can food contact surfaces be modified to provide antimicrobial characteristics or barriers to microbial attachment? • What aspects of modified atmosphere packaging (MAP) can be used to enhance active packaging technology? This summary of information presented in previous chapters provides a framework for identifying key research needs and a starting point for future research approaches that will improve the safety of fresh produce. This list of recommendations is neither exhaustive nor immutable. It must be adaptable and responsive to changing conditions and newly emerging information on produce contamination and safety. The significance and complexity of the problem facing researchers in industry, government and academia may seem daunting, but focused research efforts by recognized experts will provide a better scientific understanding of the issues. This will guide advances in the most important areas of research and will ultimately provide the development of tools to improve the safety of fresh and fresh-cut produce for consumers.

References Cooley MB, Chao DK, Mandrell RE. 2006. Escherichia coli O157 : H7 survival and growth on lettuce is altered by the presence of epiphytic bacteria. Journal of Food Protection 69:10:2329–2335. FDA. 2008. U.S. Food and Drug Administration—Final Rule (73 FR 49593), Irradiation in the Production, Processing and Handling of Food 21 CFR Part 179. Available at http://www.cfsan.fda.gov/∼lrd/fr080822. html, accessed Oct 22, 2008. FEPSRI. 2007. Fresh Express Produce Safety Research Initiative. Available at http://www.freshexpress. com/research/research.asp, accessed Oct 22, 2008. Gombas D. 2008. Researching the right questions. Fresh-Cut, February 2008, p. 19. Gourabathini P, Brandl MT, Redding KS, Gunderson JH, Berk SG. 2008. Interactions between foodborne pathogens and protozoa isolated from lettuce and spinach. Applied and Environmental Microbiology 75(8):2518–2525. JIFSAN. 2007. Joint Institute for Food Safety and Applied Nutrition—Tomato Safety Research Needs Workshop Available at http://www.jifsan.umd.edu/tomato_wkp2007.htm, accessed Oct 22, 2008. NRI. 2008. National Research Initiative, USDA-CSREES. Available at http://www.csrees.usda.gov/fo/ foodsafetynri.cfm, accessed Oct 22, 2008. UFPA. 2007. United Fresh Produce Association—Leafy Greens Food Safety Research Conference. Available at http://www.unitedfresh.org/newsviews/leafy_greens_food_safety_research, accessed Oct 22, 2008.

Index

Acidic electrolyzed water (AEW), 172, 175–76 Acidified sodium chlorite, aqueous antimicrobial treatment with, 171t, 178–79 Aeromonas hydrophila, manure with, 147 AEW. See Acidic electrolyzed water Agricultural practices. See Good agricultural practices Airborne contamination, preharvest produce influenced by, 45f, 47–48 Alfalfa sprouts, IMB for detection of pathogens in, 339–40 Alginate, human pathogen effected by, 232t Alginate-apple puree, human pathogen effected by, 232t Almonds foodborne illness outbreaks with, 105t, 400t S. enteritidis contaminated, 8t–9t, 12 Animal and pest management, 110f, 113–14 Animal wastes, preharvest produce contamination from, 45–46, 45f Antimicrobial edible films/coatings antimicrobial plan essential oils in edible films, 228–30 biopolymers used for, 225–26 edible coatings for fruit/vegetables, 226 edible films casting methods for, 228 edible films for fruit/vegetables, 226–27 evaluation of antimicrobial activity of volatile components with, 230–31 fruit/vegetable-based, 227 human pathogen control with, 225–35 measuring antimicrobial activity of, 231–32, 232t physical properties with essential oils in, 229–30 use on fruits/vegetables of, 233–35, 233t, 234t

Apple juice Cryptosporidium parvum contaminated, 44t manure contamination of, 145t–146t thermal resistance of, 244t Apples alginate-apple puree, 232t antimicrobial edible films used on, 233t, 234t consumption/imports of, 402t hot-water treatment for, 254 internalization of pathogens in, 67–68 pectin-apple puree, 232t Aqueous antimicrobial treatment acidified sodium chlorite, 171t, 178–79 chlorine and chlorine compounds, 173–75 chlorine dioxide, 171t, 177–78 electrolyzed water, 175–76, 175t factors influencing, 170–73, 171t microbial type, level and attachment, 172–73 produce surface properties, 173 washing conditions, 170–71 washing conditions—antimicrobial concentration, 170 washing conditions—pH, 171 washing conditions—sanitizer flow hydrodynamics, 171 washing conditions—temperature, 170 washing conditions—time, 170 water quality, 171–72, 171t water quality—organic loading, 171–72, 171t water quality—water pH and chemistry, 172 hydrogen peroxide, 171t, 182–83 organic acids, 183–84 ozone, 171t, 181–82 peroxyacetic acid, 171t, 179–81 produce safety with, 169–84

427

428

Index

Arobacter butzleria, manure with, 147 ATR. See Attenuated total reflection Attenuated total reflection (ATR), 332 Audit, 322 Auditee, 322 Auditing body, 322 Auditor, 322 Bacillus, 271 Bacillus anthracis, manure with, 147 Bacillus subtilis, 64 Bacterial antagonists, 209–13, 211t, 212t Bacterial spores, 265–71, 265f, 266f killing mechanisms for, 269–71 DNA damage in, 269 germination disruption in, 269–70 inner membrane damage in, 270–71 morphology of, 266–69, 266f schematic representation of, 265f Bacteriophages (Phages), biological control of pathogens with, 207–13, 211t, 212t Bananas consumption/imports of, 402t modified atmosphere packaging of, 280t, 281–84, 282t, 283f Basil, foodborne illness outbreaks with, 105t, 400t Beans, hot-water treatment for, 250 Beef cattle, E. coli contaminated, 17t “Bertha,” 272–73 Biological hazards understanding, 110f, 111 Biosensors IMB used for, 334–44, 336f–339f, 341f–343f detection of pathogens in irrigation water with, 342–44, 343f detection of pathogens in sprouts with, 341–42, 342f detection of pathogens on cantaloupe with, 335–36 detection of pathogens with alfalfa sprouts with, 339–40 dissociation-enhanced lanthanide fluoroimmunoassay with, 340 NAD(P)H method with, 336–38, 336f, 337f sandwich method with, 338, 338f, 339f time-resolved fluorescence approach with, 340–41, 341f

time-resolved fluorescence of lanthanide cations with, 340 pathogen detection in produce with, 332–34 Birds, preharvest contamination by, 49, 49t, 50f Broccoli consumption/imports of, 402t hot-water treatment for, 250 modified atmosphere packaging of, 280t thermal resistance of, 245t Brucella abortus, manure with, 147 Cabbage consumption/imports of, 402t foodborne illness outbreaks with, 105t, 400t manure contamination of, 146t Cabbage leaf, antimicrobial edible films used on, 234t CalFERT. See California Food Emergency Response Team Calicivirus foodborne outbreaks associated with, 105t fruit contaminated with, 44t California Department of Health Services, Food and Drug Branch (CDHSFDB), 375 California Food Emergency Response Team (CalFERT), 390 California Leafy Green Marketing Agreement (LGMA), 401 economics of adoption of, 408–12 private benefits/costs, 408–10 public benefits/costs, 410–12 Campylobacter animal sources of, 18t, 19–20 cattle, 18t chickens, 18t ducks, 18t pigs, 18t sheep, 18t foodborne outbreaks associated with, 105t inactivation of during compost of, 157t incidence on produce of, 11 number of U.S. outbreaks of, 9 Campylobacter jejuni fruit salad contaminated with, 44t manure with, 147 melon contaminated with, 44t

Index

orange juice contaminated with, 44t strawberries contaminated with, 44t Cantaloupe Cladosporium cladosporioides contaminated, 68 consumption/imports of, 402t E. coli contaminated, 44t foodborne illness outbreaks with, 105t, 400t hot-water treatment for, 251–54, 253t, 254t IMB for detection of pathogens in, 335–36 internalization of pathogens in, 68 manure contamination of, 146t modified atmosphere packaging of, 280t Penicillium expansum contaminated, 68 S. Litchfield contaminated, 8t S. Poona contaminated, 7t S. Saphra contaminated, 7t Salmonella contaminated, 5, 6, 10, 12, 68 thermal treatment to eliminate Salmonella from, 246, 246f–248f Carrot juice, Clostridium botulinum contaminated, 44t Carrots antimicrobial edible films used on, 234t consumption/imports of, 402t manure contamination of, 146t Shigella sonnei contaminated, 9t, 10 Yersinia pseudotuberculosis contaminated, 9t, 10 CATI. See Computer Assisted Telephone Interviews Cattle Campylobacter contaminated, 18t E. coli contaminated, 17t manure, 88t Salmonella enterica contaminated, 18t Cauliflower, consumption/imports of, 402t CDC. See Centers for Disease Control CDHS-FDB. See California Department of Health Services, Food and Drug Branch Center for Food Safety and Applied Nutrition (CFSAN), 102 Center for Science in the Public Interest (CSPI), 103 Centers for Disease Control (CDC), 101, 351

429

Foodborne Outbreak Surveillance System, 6, 9 CFSAN. See Center for Food Safety and Applied Nutrition CFU. See Colony forming units Cherry antimicrobial edible films used on, 234t consumption/imports of, 402t Cherry orchards case study, dissemination of enteric bacteria in, 48–50, 49t, 50f Chickens Campylobacter contaminated, 18t E. coli contaminated, 17t manure, 88t Chitosan, human pathogen effected by, 232t Chlamydia psittaci, manure with, 147 Chlorine, aqueous antimicrobial treatment with, 173–75 Chlorine dioxide aqueous antimicrobial treatment with, 171t, 177–78 military food safety/shelf life using, 271–78, 272f, 274t, 275f, 276f, 277f, 277t oxidation capacity of, 171t Chlorous acid, oxidation capacity of, 171t Cinnamon, antimicrobial edible films with, 228–29, 232t Citrobacter, 94 Citrobacter freundii, 95 Cladosporium cladosporioides, cantaloupe contaminated with, 68 Cleaning and sanitation, 110f, 112–13 Client, 322 Clostridium, 271 Clostridium botulinum carrot juice contaminated with, 44t manure with, 147 Clove, antimicrobial edible films with, 228–29 CLSM. See Confocal laser scanning microscopy Codex Alimentarius Commission, 84 Coleslaw, manure contamination of, 146t Colony forming units (CFU), 207 Compost control of pathogens in manure by, 155–59, 157t education on safe use of, 159

430

Index

Compost (continued) good agricultural practices with, 110f, 111–12 identifying improper practices with, 156–58, 157t pathogen contamination on produce with, 152–55 under controlled conditions, 153 under field conditions, 153 impact of wildlife on farms with, 154 internalization via roots, 153–54 organic production practices with, 154–55 pathogen regrowth with, 158–59 pathogen survival in, 87–91, 88t, 89t Salmonella and E. coli, 87–89, 88t produce safety with, 143–60 use of, 110f, 111–12 Computer Assisted Telephone Interviews (CATI), 353 Confocal laser scanning microscopy (CLSM), 63 Consumer/food-service, 291–303 consumer handling practices of, 296–99 bringing produce home, 296–97 home storage, 297 kitchen sanitation, 297–98 preparation, 297 selecting produce, 296 storage of leftovers, 298–99 washing produce, 298 consumer perception of produce safety, 292–96 integrated pest management with, 294–95 microbiological hazards with, 295 organic food in, 293–94 perception of processing for convenience in, 295 waxes with, 295 eating/staying healthy with, 300–301 food-service workers handling practices of, 299–300 safe handling of fruits/vegetables with, 301–3 bacteria are everywhere, 302 everyone is at risk for foodborne illness, 301

home storage in, 302 kitchen setup/preparation, 302 refrigerate leftovers, 303 separate bags at supermarket for, 302 variety in eating, 301 wash all fruits/vegetables, 303 washing hands, 302 selecting produce for, 291–92 Containment, 421, 423–24 Contributory negligence, 394–95 Corrective action reports, 327–28, 328f Coxiella burneti, manure with, 147 Crisis management, 110f, 114–15, 119–28 Incident Management Plans for, 119–28 communication procedures for, 121–22 defining, 120–22 definition of threat or action levels, 121 developing, 122–24 developing document for, 121–22 guiding principles of, 121 implementing during crisis, 124–26 incident management team, 122 media communication policy, 121 purpose and definitions sections of, 121 seven R’s of response, 126–28 reform, 127 regret, 126 repeat, 128 responsibility, 126–27 restitution, 127 review, 127 two categories of crises, 120 Cryptosporidium irrigation water contamination with, 136 process water contamination with, 138 surface water with, 132 Cryptosporidium parvum apple juice contaminated with, 44t number of U.S. outbreaks of, 10 CSPI. See Center for Science in the Public Interest Cyclospora, foodborne outbreaks associated with, 105t Cyclospora cayetanensis fruit salad contaminated with, 44t number of U.S. outbreaks of, 9 raspberries contaminated with, 44, 44t

Index

Dairy cattle E. coli contaminated, 17t manure, 88t Salmonella enterica contaminated, 18t Deer E. coli contaminated, 17t Salmonella enterica contaminated, 18t DELFIA. See Dissociation-enhanced lanthanide fluoroimmunoassay D-FENS sprayer (Disinfectant-sprayer for Foods and Environmentally friendly Sanitation), 264, 277–78, 277t Disinfectant-sprayer for Foods and Environmentally friendly Sanitation. See D-FENS sprayer Dissociation-enhanced lanthanide fluoroimmunoassay (DELFIA), 340 Ducks Campylobacter contaminated, 18t E. coli contaminated, 17t E. coli animal sources of, 16–19, 17t–18t beef cattle, 17t chickens, 17t dairy cattle, 17t deer, 17t ducks, 17t feral swine, 17t fish, 17t goats, 17t pigs, 17t rabbits, 17t–18t rats, 17t sheep, 17t turkeys, 17t zebu, 17t antimicrobial edible films’ effects on, 232t apples contaminated with, 67–68 bacterial protective culture in control of, 211t cantaloupe contaminated with, 44t cherry orchards contaminated with, 48–50, 49t, 50f contamination in fresh fruits/vegetables, 94

431

foodborne illness outbreaks with, 105t, 400f groundwater with, 130–31 inactivation of during compost of, 157t incidence in watersheds of, 21–23, 22t incidence on produce of, 11–19, 14t, 17t–18t irradiation’s influence on, 195–96 irrigation water contamination with, 136–37 leafy greens contaminated with, 5, 7t, 9–11, 15, 65f, 71–74, 370t, 372t lettuce contaminated with, xvii, 7t, 15, 24t, 72–73, 370t, 372t manure with, 24t, 145t, 147 animal feces/slurry with, 148–49 number of U.S. outbreaks of, 9 O157:H7 and non-O157 STEC, 16–19, 17t–18t oranges contaminated with, 69 pineapple contaminated with, 44t process water contamination with, 138 reliability as indicator for contamination of produce, 93–94 romaine lettuce contaminated with, 65f sandwich method used to detect, 338 shedding by animals of, 20–21 soil with, 24t, 148–51 spinach contaminated with, xvii, 7t, 15, 24t, 73–74, 331 public policy on, 377–81 spinach recall of 2006, public response to, 351, 358, 359f, 359t sprouts contaminated with, 10, 70–71 surface water with, 131–33 survival in manure, compost, 87–89, 88t survival in water of, 133–34, 133t thermal resistance of, 244t, 245f transport in environment of, 23–26, 24t USDA Agricultural Marketing Service study on, 14, 14t water with, 24t Earthbound Farm, 119 Economics, food safety, 399–417 foodborne illness outbreak, 400f, 400t government/industry response to, 406–8 U.S. produce industry, 401–6, 402t, 403f–405f

432

Index

Edible films. See Antimicrobial edible films/ coatings Education and training, 110f, 111 Electrolyzed neutral water (ENW), 176 Electrolyzed water, aqueous antimicrobial treatment with, 175–76, 175t Electron beam, 192t Endophytes, 55–56 extraction of, 61–62 Enterobacter, 94 Enterococcus mundtii, bacterial protective culture with, 211t Environmental risk factors, 421 ENW. See Electrolyzed neutral water Eradication, 421, 424–25 Erwinia amylovora, phages in control of, 208 Erysipelothrix rhusiopathiae, manure with, 147 Essential oils antimicrobial edible films with, 228–29, 232t control of pathogens with, 215–17 Evanescent wave (EW), 332 EW. See Evanescent wave External audit (Second- or third-party audit), 323 Farm biosecurity, 110f, 115 Fecal contamination, 15 E. coli as indicator for, 93–94 irrigation water with, 136–38 manure/compost, 144–46, 145t–146t process water with, 138–39 Fecal indicator, 134t Feral swine, E. coli contaminated, 17t Fish, E. coli contaminated, 17t FITC. See Fluorescein isothiocyanate Fluorescein isothiocyanate (FITC), 63 Fluorescent microbeads, 64–65 Fly larvae infestation, mangos with, 68–69 Foodborne illness. See Outbreaks, foodborne illness Foodborne Outbreak Surveillance System, 6, 9 Food safety documents, 316–17 Francisella tularensis, manure with, 147 Free available chlorine, 174 Fruit. See also specific fruit cherry orchards case study, 48–50, 49t, 50f

consumption/imports of, 402t E. coli contamination in, 94 edible coatings for, 226 edible films for, 226–27, 233–35, 233t, 234t fruit/vegetable-based edible films, 227 harvest contamination of, 48 isolation of pathogens from, 91–93, 92t origin/spread of pathogens in, 43–51, 44t, 45f, 49t, 50f preharvest contamination of, 45–48, 45f safe handling for consumer of, 301–3 Fruit salad Campylobacter jejuni contaminated, 44t Cyclospora cayetanensis contaminated, 44t Fungi, biological control of pathogens with, 213–15 Gamma radioisotopes, 192t GAP. See Good agricultural practices Generally recognized as safe (GRAS), 208 Genetic engineering, 84 Giardia irrigation water contamination with, 136 process water contamination with, 138 surface water with, 132 Giardia lamblia, number of U.S. outbreaks of, 9–10 Gluconobacter asaii, bacterial protective culture with, 210, 211t Glucuronidase (GUS), 63 GMP. See Good manufacturing practices Goats E. coli contaminated, 17t Salmonella enterica contaminated, 18t Good agricultural practices (GAP) advancing implementation of, 115–16 components of, 110–15, 110f animal and pest management, 110f, 113–14 biological hazards understanding, 110f, 111 cleaning and sanitation, 110f, 112–13 crisis management, 110f, 114–15, 119–28 education and training, 110f, 111 farm biosecurity, 110f, 115 manure use and composting, 110f, 111–12

Index

microbial water quality, 110f, 111 recall and traceback, 110f, 114 total management commitment, 110, 110f worker health/hygiene, 110f, 112 development/implementation of, 108–10 economics of adoption of, 408–12 private benefits/costs, 408–10 public benefits/costs, 410–12 eight principles of microbial food safety, 109 foodborne outbreaks by causative agent, 104, 105t foodborne outbreaks by commodity, 104, 105t food safety with, 311 minimize risk of pathogen with, xviii nature of pathogen contamination and, 104–7, 105t, 106f produce safety from good, 101–16 Good manufacturing practices (GMP), 309–10 Grape juice, 244t Grapes antimicrobial edible films used on, 234t consumption/imports of, 402t foodborne outbreaks associated with, 105t, 400t hot-water treatment for, 255 GRAS. See Generally recognized as safe Green grapes, foodborne illness outbreaks with, 105t, 400t Green onions foodborne illness outbreaks with, 105t, 106f, 400t Hepatitis A contaminated, 331 hot-water treatment for, 251 Green pepper, antimicrobial edible films used on, 233t Groundwater, foodborne pathogens in, 130–31 Guide to Minimize Microbial Food Safety Hazards for Fresh Fruits and Vegetables (FDA), 108 GUS. See Glucuronidase HACCP. See Hazard analysis and critical control point Harvest, contamination at, 48, 57f

433

Hazard analysis and critical control point (HACCP), 310–11, 335, 391 Hemolytic uremic syndrome (HUS), 331 Hepatitis A foodborne outbreaks associated with, 105t green onions contaminated with, 331 manure/animal source of, 146t number of U.S. outbreaks of, 9 strawberries contaminated with, 44t Honeydew melon foodborne outbreaks associated with, 105t Gluconobacter asaii as protection for, 211t modified atmosphere packaging of, 280t Hot-water treatment, 249–57, 253t, 254t apples, 254 beans, 250 broccoli, 250 cantaloupe, 251–54, 253t, 254t commercial application considerations for, 257 fruits, 251–56, 253t, 254t grapes, 255 green onions, 251 lettuce, 249–50 mangos, 256 oranges, 254–55 pears, 256 pineapple, 256 sprouting seeds, 256–57 sprouts, 251 tomatoes, 255–56 vegetables, 249–51 watercress, 251 HUS. See Hemolytic uremic syndrome Hydrogen peroxide aqueous antimicrobial treatment with, 171t, 182–83 oxidation capacity of, 171t Iceberg lettuce E. coli contaminated, 7t modified atmosphere packaging of, 280–81, 280t, 281t, 282f, 283f retail market shares of, 413t Yersinia pseudotuberculosis contaminated, 9t IFC. See Immunofluorescent colony staining IFPA. See International Fresh-cut Produce Association IMB. See Immunomagnetic beads

434

Index

Immunofluorescent colony staining (IFC), 63 Immunomagnetic beads (IMB), 331–44 detection of pathogens in irrigation water with, 342–44, 343f detection of pathogens in sprouts with, 341–42, 342f detection of pathogens on cantaloupe with, 335–36 detection of pathogens with alfalfa sprouts with, 339–40 dissociation-enhanced lanthanide fluoroimmunoassay with, 340 NAD(P)H method with, 336–38, 336f, 337f sandwich method with, 338, 338f, 339f time-resolved fluorescence approach with, 340–41, 341f time-resolved fluorescence of lanthanide cations with, 340 Incident Management Plans, 119–28 communication procedures for, 121–22 defining, 120–22 definition of threat or action levels, 121 developing, 122–24 developing document for, 121–22 guiding principles of, 121 implementing during crisis, 124–26 incident management team, 122 media communication policy, 121 purpose and definitions sections of, 121 seven R’s of response, 126–28 reform, 127 regret, 126 repeat, 128 responsibility, 126–27 restitution, 127 review, 127 two categories of crises, 120 Index and model organisms, 134t Infrared treatment, 257–59 Integrated pest management (IPM), 294–95 Internal audit (First-party audit), 323 audit day, 327 corrective action reports, 327–28, 328f scheduling of, 326–27 third-party audit and, 325–29, 326f, 328f training record, 326f verification with, 325

Internalization, of pathogens, 421 bacterial attachment/colonization in, 56–57, 58f bacterial endophytes, 55–56 endophytic pathogen sources with, 56, 57f evidence in apples of, 67–68 evidence in cantaloupe of, 68 evidence in leafy greens of, 71–74 evidence in mangos of, 68–69 evidence in oranges of, 69 evidence in produce of, 67–74 evidence in sprouts of, 70–71 evidence in tomatoes of, 69–70 labeling of endophytic bacteria for, 62–63 methods for examining, 60–62 extraction of endophytes, 61–62 surface disinfection/maceration, 61 methods for introducing bacteria into plants for, 63–65, 65f plant and soil microflora interactions in, 66–67 plant stress’ impact on, 57–59 portals of entry for, 59–60 postharvest of entry for, 60 preharvest of entry for, 59 strategies for minimizing, 74 via roots with manure and compost, 153–54 International Fresh-cut Produce Association (IFPA), 307 International Sprout Growers Association (ISGA), 344 Interventions, 421 Ionizing radiation, 84 IPM. See Integrated pest management Irradiation applications/dose limits for food with, 193t consumer acceptance of, 200–201 defined, 191 microbial safety of produce with, 195–98 internalized pathogens, 197–98 pathogens within biofilms, 197 protected pathogens, 196–97 packaging approved for, 199–200, 199t, 200t processing considerations for, 198–99 produce safety using, 191–201 quality of produce with, 194–95 radura logo, 193f

Index

technologies for, 192t Irrigation water, 136–38 fecal contamination in, 136–38 IMB for detection of pathogens in, 342–44, 343f preharvest produce contamination from, 45f, 46–47 standards for quality of, 138 ISGA. See International Sprout Growers Association Jalapeño peppers, Salmonella contaminated, 9t, 10 Klebsiella, 94 L. monocytogenes antimicrobial edible films’ effects on, 232t bacterial protective culture in control of, 211t irradiation’s influence on, 195–96 manure/animal source of, 146t manure with, 147, 149 phages in control of, 207 LAB. See Lactic acid bacteria Lactic acid bacteria (LAB), 209–13, 212t Lactobacillus casei, bacterial protective culture with, 211t Lactobacillus lactis, bacterial protective culture with, 211t Lead auditor, 322 Leafy green produce E. coli contaminated, 5, 7t, 9–11, 15, 71–74, 370t, 372t foodborne illness outbreaks with, 400t internalization of pathogens in, 71–74 Norovirus contaminated, 370t outbreaks related to, 370t retail market shares of, 413t Salmonella contaminated, 72–73, 370t Lemongrass, antimicrobial edible films with, 228–29, 232t Leptospira, manure with, 147 Lettuce antimicrobial edible films used on, 233t, 234t consumption/imports of, 402t E. coli contaminated, xvii, 7t, 15, 24t, 65f, 72–73, 370t, 372t

435

foodborne illness outbreaks with, 105t, 400t hot-water treatment for, 249–50 manure contamination of, 145t–146t Norovirus contaminated, 370t retail market shares of, 413t Salmonella contaminated, 24t, 72, 370t Yersinia pseudotuberculosis contaminated, 9t LGMA. See California Leafy Green Marketing Agreement Listeria inactivation of during compost of, 157t incidence on produce of, 11 phages in control of, 208 Listeria monocytogenes growing at refrigerator temperatures, 107 thermal resistance of, 244t–245t Local food movement, 395–96 Lysozyme, control of pathogens with, 218 Mangos consumption/imports of, 402t foodborne outbreaks associated with, 105t hot-water treatment for, 256 internalization of pathogens in, 68–69 Manure animal wastes, human pathogens in, 147–48 cattle, 88t chicken, 88t control of pathogens by composting with, 155–59, 157t dairy, 88t E. coli with, 24t, 145t, 147 education on safe use of, 159 good agricultural practices with, 110f, 111–12 manure-amended soil, 149–52 length of time in soil influencing pathogens in, 151 microbial activity of soil influencing pathogens in, 150 rates/ways of application influencing pathogens in, 151 rhizosphere’s presence influencing pathogens in, 150 soil temperature’s effect on pathogens in, 149–50

436

Index

Manure (continued) survival of pathogens in, 151–52 type of animal waste influencing pathogens in, 150–51 outbreaks from fecal contamination, 144–46, 145t–146t pathogen contamination on produce with, 152–55 under controlled conditions, 153 under field conditions, 153 impact of wildlife on farms with, 154 internalization via roots, 153–54 organic production practices with, 154–55 pathogen survival in, 87–91, 88t, 89t Salmonella and E. coli, 87–89, 88t preharvest pathogens with, 148–55 animal feces/slurry, 148–49 manure-amended soil, 149–52 produce safety with, 143–60 use of, 110f, 111–12 MAP. See Modified atmosphere packaging Master Sanitation Plan (MSP), 313 Material change, 193 Mazetti v. Armour & Company, 386–87 Melon. See also Cantaloupe Campylobacter jejuni contaminated, 44t foodborne illness outbreaks with, 105t, 106f, 400t Norovirus contaminated, 44t Mesclum lettuce E. coli contaminated, 7t manure contamination of, 145t Methylcellulose coating, 63 Microbial water quality, 110f, 111 Microwave treatment, 257–59 Military food safety/shelf life, 263–84 bacterial spore resistance faced by, 265–71, 265f, 266f DNA damage in killing of, 269 germination disruption in killing of, 269–70 inner membrane damage in killing of, 270–71 killing mechanisms for, 269–71 morphology of, 266–69, 266f schematic representation of, 265f chlorine dioxide technologies used by, 271–78, 272f, 274t, 275f, 276f, 277f, 277t

D-FENS sprayer for, 264, 277–78, 277t modified atmosphere packaging for, 264, 278–84, 279t–282t, 282f, 283f bananas in, 280t, 281–84, 282t, 283f broccoli in, 280t cantaloupe in, 280t honeydew melon in, 280t iceberg lettuce in, 280–81, 280t, 281t, 282f, 283f in-house testing of, 280–84, 280t–282t, 282f, 283f moderate temperature compensation character of, 279 packaging atmospheres for select commodities with, 280t romaine lettuce in, 280t self-life extension obtained with, 280t technologies of, 279, 279t tomatoes in, 280t portable chemical sterilizer for, 264, 272–77, 272f, 274t, 275f–277f “Bertha” compared to, 272–73 description of, 272, 272f fresh fruits/vegetables with, 275–77, 276f, 277t microbiological validation of, 274–75 ontogeny of, 273–74, 274t, 275f three-step procedure for operating, 275f MLST. See MultiLocus Sequence Typing MLVA. See MultiLocus Variable number tandem repeat Analysis Moderate temperature compensation character, 279 Modified atmosphere packaging (MAP), 264, 278–84, 279t–282t, 282f, 283f bananas in, 280t, 281–84, 282t, 283f broccoli in, 280t cantaloupe in, 280t honeydew melon in, 280t iceberg lettuce in, 280–81, 280t, 281t, 282f, 283f in-house testing of, 280–84, 280t–282t, 282f, 283f moderate temperature compensation character of, 279 packaging atmospheres for select commodities with, 280t romaine lettuce in, 280t

Index

self-life extension obtained with, 280t technologies of, 279, 279t tomatoes in, 280t MSP. See Master Sanitation Plan MultiLocus Sequence Typing (MLST), 27 MultiLocus Variable number tandem repeat Analysis (MLVA), 27 Mycobacterium paratuberculosis, manure with, 147 Mycobacterium tuberculosis, manure with, 147 NAD(P)H method, 336–38, 336f, 337f National Organic Program (NOP), 84 National Organic Standard Board (NOSB), 83–84 Natural Selection Foods (NSF), 119, 351, 378–80 Negligence, 391–93 contributory, 394–95 politics of regulations with, 392–93 statutory law/standards of care with, 391–92 NOP. See National Organic Program Norovirus foodborne outbreaks associated with, 105t leafy greens contaminated with, 370t lettuce contaminated with, 370t melon contaminated with, 44t number of U.S. outbreaks of, 9 raspberries contaminated with, 44t NOSB. See National Organic Standard Board NSF. See Natural Selection Foods NSRDEC/CFD. See U. S. Army-Natick Soldier RD&E Center, Department of Defense Combat Feeding Directorate OFPA. See Organic Foods Production Act Onion. See also Green onions consumption/imports of, 402t thermal resistance of, 245t Orange juice Campylobacter jejuni contaminated, 44t manure contamination of, 146t thermal resistance of, 244t, 245f Oranges consumption/imports of, 402t E. coli contaminated, 69

437

hot-water treatment for, 254–55 internalization of pathogens in, 69 Salmonella contaminated, 69 Oregano, antimicrobial edible films with, 228–29, 232t Organic acids, aqueous antimicrobial treatment with, 183–84 Organic foods, 83–96 bacterial prevalence in, 91–94, 92t E. coli’s reliability as indicator for contamination, 93–94 isolation of pathogens from fruit/ vegetables, 91–93, 92t consumer perception of produce safety with, 293–94 contamination from practices of, 94–95 E. coli contamination in, 94 epidemiology of foodborne disease linked to, 95–96 history of, 83–84 market, 84–86, 85t pathogen contamination with manure fertilizer, 154–55 pathogen survival in manure, compost, soil with, 87–91, 88t, 89t organisms in soil, 89–91, 89t Salmonella and E. coli, 87–89, 88t safety issues associated with, 86–87 sales of, 85, 85t Organic Foods Production Act (OFPA), 83 Outbreaks, foodborne illness. See also Spinach recall of 2006, public response agricultural practices’ role in safety from, 101–16 almonds, 105t, 400t basil, 105t, 400t cabbage, 105t, 400t cantaloupe, 105t, 400t by causative agent, 104, 105t causative agent for, 102f, 105t changing consumer preferences with, 107 changing demographics with, 106 changing food systems with, 107 changing microorganisms with, 107 by commodity, 104, 105t economics of, 108, 399–417 fecal contamination from manure/ compost, 144–46, 145t–146t

438

Index

Outbreaks (continued) green grapes, 105t, 400t green onions, 105t, 106f, 400t illnesses by year of, 103f leafy green produce, 370t, 400t lettuce, 105t, 400t melon, 105t, 106f, 400t number of cases of, 6 produce associated, 6–11, 7t–9t raspberries, 105t, 400t research on North American, 369–73, 370t, 372t romaine lettuce, 105t, 400t snow peas, 105t, 400t spinach, 105t, 400t squash, 105t, 400t tomatoes, 105t, 106f, 400t types of produce associated with, 104–5, 106f Ozone aqueous antimicrobial treatment with, 171t, 181–82 oxidation capacity of, 171t PACA. See Perishable Agricultural Commodities Act Packaging materials, approved for use in irradiation, 199–200, 199t, 200t Parsley, foodborne outbreaks associated with, 105t Pasteurization, 193 Pathogens, human animal sources of, 16–20, 17t–18t Campylobacter, 18t, 19–20 E. coli, 16–19, 17t–18t Salmonella enterica, 17t–18t, 19 shedding of E. coli and Salmonella with, 20–21 antimicrobial edible films in control of, 225–35 biological control on produce of, 205–19 connection between control of human/ plant pathogens in, 205–6 location of pathogens in produce for, 206–7 organisms for, 206–15, 211t, 212t technologies for, 206–18, 211t, 212t biologically derived chemicals for control on produce of, 215–18 essential oils/plant extracts, 215–17

lysozyme, 218 polylysine, 218 quorum sensing signaling molecules, 217 siderophores, 217–18 cherry orchards case study, 48–50, 49t, 50f contamination indicators for watersheds with, 26–28 detection in produce of, 331–44 biosensors for, 332–34 biosensors process with IMB for, 334–44, 336f–339f, 341f–343f fruit production systems with, 43–51, 44t, 45f, 49t, 50f harvest contamination by, 48 incidence of generic E. coli, 13–15, 14t incidence on fresh produce of, 11–13 internalization of, 55–74 bacterial attachment/colonization in, 56–57, 58f bacterial endophytes, 55–56 endophytic pathogen sources with, 56, 57f evidence in apples of, 67–68 evidence in cantaloupe of, 68 evidence in leafy greens of, 71–74 evidence in mangos of, 68–69 evidence in oranges of, 69 evidence in produce of, 67–74 evidence in sprouts of, 70–71 evidence in tomatoes of, 69–70 labeling of endophytic bacteria for, 62–63 methods for examining, 60–62 extraction of endophytes, 61–62 surface disinfection/maceration, 61 methods for introducing bacteria into plants for, 63–65, 65f plant and soil microflora interactions in, 66–67 plant stress’ impact on, 57–59 portals of entry for, 59–60 postharvest of entry for, 60 preharvest of entry for, 59 strategies for minimizing, 74 municipal/agricultural watersheds with, 21–23, 22t organisms for control on produce of, 206–15, 211t, 212t

Index

bacterial protective cultures, 209–13, 211t, 212t bacteriophages (phages), 207–8 yeast and fungi, 213–15 outbreaks associated with fresh produce with, 6–11, 7t–9t persistence in agricultural environment of, 13 preharvest contamination by, xvii, 45–48, 45f airborne, 45f, 47–48 animal wastes in, 45–46, 45f birds in, 49, 49t, 50f irrigation water in, 45f, 46–47 pickers’ hands in, 49, 49t, 50f sorters’ hands in, 49, 49t, 50f sources of, 56, 57f preharvest produce and survival of, 28–30 hypotheses from recent outbreaks, 28–29 produce associated, 5–32 source-tracking of, 26–28 thermal treatment for control of, 241–59 transport in environment of, 23–26, 24t water with, 130–34, 133t groundwater, 130–31 pathogen survival in, 133–34, 133t surface water, 131–33 PCS. See Portable chemical sterilizer Pears, hot-water treatment for, 256 Peas. See also Snow peas thermal resistance of, 245t Pectin-apple puree, human pathogen effected by, 232t Pediococcus pentosaceus, bacterial protective culture with, 211t Penicillium expansum, cantaloupe contaminated with, 68 Peppers green, 233t Salmonella contaminated, xvii, 9t, 10 thermal resistance of, 245t Peracetic acid. See Peroxyacetic acid Perishable Agricultural Commodities Act (PACA), 390 Peroxyacetic acid (Peracetic acid, POAA) aqueous antimicrobial treatment with, 171t, 179–81

439

oxidation capacity of, 171t Pest control, 315 PFGE. See Pulsed field gel electrophoresis Phages. See Bacteriophages PHF. See Potentially hazardous foods Pickers’ hands, preharvest contamination by, 49, 49t, 50f Pigs Campylobacter contaminated, 18t E. coli contaminated, 17t Salmonella enterica contaminated, 18t Pineapple consumption/imports of, 402t E. coli contaminated, 44t hot-water treatment for, 256 Plant extracts, control of pathogens with, 215–17 Plant sanitation, 307–20 employee training in GMP, 309–10 facility and equipment sanitation for, 313–15 chemical handling procedures, 314–15 designated cleaning crew, 314 internal inspections, 313–14 Master Sanitation Plan, 313 pest control, 315 sanitation standard operation procedures, 314 monitoring for, 315–17 record-keeping logs and forms, 316–17 traceback and product recall, 316 process controls for, 310–13 finished product specifications, 312–13 GAP, 311 hazard analysis and critical control point, 310–11 raw product specifications, 311 washing and water sanitation, 311–12 risk assessment for, 307–9 chemical hazards in, 308–9 microbiological hazards in, 308 physical hazards in, 308 verification for, 317–19 microbiological testing in, 319 supplier approval programs in, 318–19 third-party audits in, 318 worker hygiene for, 309–10 POAA. See Peroxyacetic acid Polylysine, control of pathogens with, 218

440

Index

Portable chemical sterilizer (PCS), 264, 272–77, 272f, 274t, 275f–277f “Bertha” compared to, 272–73 description of, 272, 272f fresh fruits/vegetables with, 275–77, 276f, 277t microbiological validation of, 274–75 ontogeny of, 273–74, 274t, 275f three-step procedure for operating, 275f Postharvest, contamination at, 57f Potatoes consumption/imports of, 402t manure contamination of, 145t Potentially hazardous foods (PHF), 299 Poultry, Salmonella enterica contaminated, 18t Preharvest produce cherry orchards case study, 48–50, 49t, 50f contamination of, xviii, 45–48, 45f airborne, 45f, 47–48 animal wastes in, 45–46, 45f birds in, 49, 49t, 50f irrigation water in, 45f, 46–47 pickers’ hands in, 49, 49t, 50f sorters’ hands in, 49, 49t, 50f sources of, 56, 57f pathogens with manure and composting with, 148–55 animal feces/slurry, 148–49 manure-amended soil, 149–52 survival of pathogens with, 28–29 Prevention, 422–23 Process indicator, 134t Process water, 138–39 Produce consumer handling of, 291–303 E. coli contamination in, 94 E. coli’s reliability as indicator for contamination in, 93–94 edible films for extended shelf life of, 225–35 food-service handling of, 291–303 human pathogens associated with, 5–32 incidence of Campylobacter on, 11 incidence of E. coli on, 11–19, 14t, 17t–18t incidence of human pathogens on, 11–13 incidence of Listeria on, 11 incidence of Salmonella on, 11–13, 18t, 19

incidence of Yersinia on, 11 internalization of pathogens in, 55–74 bacterial attachment/colonization in, 56–57, 58f bacterial endophytes, 55–56 endophytic pathogen sources with, 56, 57f evidence in apples of, 67–68 evidence in cantaloupe of, 68 evidence in leafy greens of, 71–74 evidence in mangos of, 68–69 evidence in oranges of, 69 evidence in produce of, 67–74 evidence in sprouts of, 70–71 evidence in tomatoes of, 69–70 labeling of endophytic bacteria for, 62–63 methods for examining, 60–62 methods for introducing bacteria into plants for, 63–65, 65f plant and soil microflora interactions in, 66–67 plant stress’ impact on, 57–59 portals of entry for, 59–60 postharvest of entry for, 60 preharvest of entry for, 59 strategies for minimizing, 74 leafy green, 5, 7t, 9–11, 15, 71–74 military, safety/extended shelf life for, 263–84 nature of pathogen contamination of, 104–7, 105t, 106f organic v. conventional in safety of, 83–96 bacterial prevalence in, 91–94, 92t contamination from practices of, 94–95 epidemiology of foodborne disease linked to, 95–96 pathogen survival in manure, compost, soil for, 87–91, 88t, 89t outbreaks associated with, 6–11, 7t–9t changing consumer preferences with, 107 changing demographics with, 106 changing food systems with, 107 changing microorganisms with, 107 economic impact of, 108 pathogen contamination with manure fertilizer, 152–55 under controlled conditions, 153

Index

under field conditions, 153 impact of wildlife on farms with, 154 internalization via roots, 153–54 organic production practices with, 154–55 pathogens detection in, 331–44 biosensors for, 332–34 biosensors process with IMB for, 334–44, 336f–339f, 341f–343f public policy on, 369–81 industry efforts/regulation with, 373–77 research on American outbreaks for, 369–73, 370t, 372t spinach, 354–62, 377–81 safety agricultural practices’ role in, 101–16 aqueous antimicrobial treatment for, 169–84 economics of, 399–417 government/industry response to, 406–8 irradiation for, 191–201 manure and compost in, 143–60 organic v. conventional in, 83–96 thermal treatment for, 241–59 water and water testing in, 129–39 seasonal availability of, 5 survival of pathogens on preharvest, 28–30 hypotheses from recent outbreaks, 28–29 washing acidified sodium chlorite, 171t, 178–79 chlorine and chlorine compounds, 173–75 chlorine dioxide, 171t, 177–78 electrolyzed water, 175–76, 175t factors influencing, 170–73, 171t hydrogen peroxide, 171t, 182–83 organic acids, 183–84 ozone, 171t, 181–82 peroxyacetic acid, 171t, 179–81 produce safety with, 169–84 Produce industry efforts/regulation, 373–77 Produce Safety Assurance Pyramid, 110–11, 110f Product liability contaminated fresh produce and, 385–96 legal responsibility with, 385 defining products and defects for, 388 Mazetti v. Armour & Company, 386–87

441

modern rule of strict liability in, 387–89 negligence’s role in, 391–93 politics of regulations with, 392–93 statutory law/standards of care with, 391–92 origins of, 386–87 from contract to tort, 386 strict liability in tainted food cases, 386–87 proof of causation for, 389 proving defects for, 388–89 spreading blame to reduce, 393–95 allocation of fault with, 393–94 consumer responsibility and contributory negligence with, 394–95 contribution with, 393–94 indemnification with, 393–94 sustainability/local food movement’s effect on, 395–96 traceability with, 389–91 Winterbottom v. Wright, 386–87 Product recall, 316 Pseudomonas fluorescens, bacterial protective culture with, 211t Pulsed field gel electrophoresis (PFGE), 13, 27 Quince, antimicrobial edible films used on, 234t Quorum sensing signaling molecules, control of pathogens with, 217 Rabbits, E. coli contaminated, 17t–18t Radura logo, 193f Raspberries Cyclospora cayetanensis contaminated, 44, 44t foodborne illness outbreaks with, 105t, 400t imports of, 404, 404f Norovirus contaminated, 44t Rats, E. coli contaminated, 17t Ready-to-eat (RTE), 143, 263 Recall and traceback, 110f, 114 Record-keeping logs and forms, 316–17 Research needs, 421–25 containment, 421, 423–24 environmental risk factors, 421 eradication, 421, 424–25

442

Index

Research needs (continued) internalization, of pathogens, 421 interventions, 421 prevention/microbial ecology, 422–23 RF treatment, 257–59 Rhizosphere, manure-amended soil with, 150 Romaine lettuce E. coli contaminated, 7t, 65f foodborne illness outbreaks with, 105t, 400t modified atmosphere packaging of, 280t retail market shares of, 413t Root tip pruning, 63–64 RTE. See Ready-to-eat S. Baildon, tomatoes contaminated with, 8t S. Braenderup, tomatoes contaminated with, 8t S. enteritidis almonds contaminated with, 8t–9t, 12 antimicrobial edible films’ effects on, 232t inactivation of during compost of, 157t S. Javiana, tomatoes contaminated with, 8t S. Litchfield, cantaloupe contaminated with, 8t S. Montevideo, tomatoes contaminated with, 8t, 69–70 S. Newport, tomatoes contaminated with, 8t S. Poona, cantaloupe contaminated with, 7t S. Saphra, cantaloupe contaminated with, 7t S. Typhimurium, tomatoes contaminated with, 8t Salad Bowl of America, 5 Salmonella bacterial protective culture in control of, 211t cantaloupe contaminated with, 5, 6, 10, 12, 68 foodborne outbreaks associated with, 105t inactivation of during compost of, 157t incidence in watersheds of, 21–23, 22t incidence on produce of, 11–13, 18t, 19 irradiation’s influence on, 195–96 irrigation water contamination with, 136–37 jalapeño peppers contaminated with, 9t, 10 leafy greens contaminated with, 72–73, 370t lettuce contaminated with, 24t, 72, 370t

location on produce of, 206 manure/animal source of, 146t manure with, 147 number of U.S. outbreaks of, 9 oranges contaminated with, 69 phages in control of, 207 process water contamination with, 138 shedding by animals of, 20–21 soil with, 24t spinach contaminated with, 24t, 73 sprouts contaminated with, 70–71 survival in manure, compost, 87–89, 88t thermal resistance of, 244t thermal treatment of cantaloupe to eliminate, 246, 246f–248f tomatoes contaminated with, xvii, 5, 6, 10, 12, 69–70 transport in environment of, 23–26, 24t water with, 24t Salmonella enterica animal sources of, 18t, 19 cattle, 18t dairy cattle, 18t deer, 18t goats, 18t pigs, 18t poultry, 18t sheep, 18t wild birds, 18t wild tortoises, 18t sprouts contaminated with, 10 Sandwich method, 338, 338f, 339f Sanitation, 307–20 employee training in GMP, 309–10 facility and equipment sanitation for, 313–15 chemical handling procedures, 314–15 designated cleaning crew, 314 internal inspections, 313–14 Master Sanitation Plan, 313 pest control, 315 sanitation standard operation procedures, 314 monitoring for, 315–17 record-keeping logs and forms, 316–17 traceback and product recall, 316 process controls for, 310–13 finished product specifications, 312–13 GAP, 311

Index

hazard analysis and critical control point, 310–11 raw product specifications, 311 washing and water sanitation, 311–12 risk assessment for, 307–9 chemical hazards in, 308–9 microbiological hazards in, 308 physical hazards in, 308 verification for, 317–19 microbiological testing in, 319 supplier approval programs in, 318–19 third-party audits in, 318 worker hygiene for, 309–10 Sanitation standard operation procedures (SSOP), 314 Seven R’s of response, 126–28 Shedding, E. coli and Salmonella in, 20–21 Sheep Campylobacter contaminated, 18t E. coli contaminated, 17t Salmonella enterica contaminated, 18t Shigatoxin-positive E. coli (STEC), 16–19, 17t–18t Shigella foodborne outbreaks associated with, 105t number of U.S. outbreaks of, 9 Shigella flexneri, tomatoes contaminated with, 9t Shigella sonnei, carrots contaminated with, 9t, 10 Siderophores, control of pathogens with, 217–18 Single Nucleotide Polymorphism (SNP), 27 Snow peas, foodborne illness outbreaks with, 105t, 400t SNP. See Single Nucleotide Polymorphism Sodium hypochlorite, oxidation capacity of, 171t Soil E. coli with, 24t, 148–51 manure-amended, 149–52 length of time in soil influencing pathogens in, 151 microbial activity of soil influencing pathogens in, 150 rates/ways of application influencing pathogens in, 151 rhizosphere’s presence influencing pathogens in, 150

443

soil temperature’s effect on pathogens in, 149–50 survival of pathogens in, 151–52 type of animal waste influencing pathogens in, 150–51 pathogen survival in, 87–91, 88t, 89t Salmonella with, 24t Soil drenching, 63 Sorters’ hands, preharvest contamination by, 49, 49t, 50f Spinach consumption/imports of, 402t, 403f E. coli contaminated, xvii, 7t, 15, 24t, 73–74, 331, 351 public policy on, 377–81 spinach recall of 2006, public response to, 351, 358, 359f, 359t foodborne illness outbreaks with, 105t, 400t retail market shares of, 413t Salmonella contaminated, 73 Salmonella with, 24t Spinach recall of 2006, public response, 351–66 brief history of, 351–53 economics of, 399, 412–16, 412f, 413t, 414f, 415f public policy on, 377–81 study methods exploring, 353 Computer Assisted Telephone Interviews for, 353 sample, 353 survey instrument, 353 study results exploring, 354–62 awareness and interest with recall, 354–56, 355f, 356t concerns of spinach safety affect other produce, 360 conclusions reached with, 262–366 demographics of spinach eaters after recall, 262, 263t knowledge about details of recall, 356–58, 357f, 357t knowledge about E. coli and symptoms, 358, 359f, 359t many think spinach safer now, 361, 361f many unsure recall was over, 360–61 most will eat spinach again, 262t, 361–262

444

Index

Spinach recall of 2006 (continued) questions asked about recall, 356t after recall, 262t, 360–62, 361f before recall, 354 during recall, 354–60, 355f, 356t, 357f, 357t, 359f, 359t some ate spinach during recall, 359–60 some thought washing spinach made it safe, 360 SPR. See Surface plasmon resonance Sprouts alfalfa, 339–40 E. coli and S. enterica contaminated, 10 E. coli contaminated, 70–71 hot-water treatment for, 251 IMB for detection of pathogens in, 341–42, 342f internalization of pathogens in, 70–71 manure contamination of, 145t–146t Salmonella contaminated, 70–71 Squash, foodborne illness outbreaks with, 105t, 400t SSOP. See Sanitation standard operation procedures Standards of care, 391–92 Staphylococcus aureus antimicrobial edible films’ effects on, 232t bacterial protective culture in control of, 211t number of U.S. outbreaks of, 9 strawberries contaminated with, 44t STEC. See Shigatoxin-positive E. coli Strawberries antimicrobial edible films used on, 233t, 234t Campylobacter jejuni contaminated, 44t consumption/imports of, 402t Hepatitis A contaminated, 44t Staphylococcus aureus contaminated, 44t Surface plasmon resonance (SPR), 332 Surface water, foodborne pathogens in, 131–33 Sustainability, 395–96 Thermal treatment, 241–59 fundamentals of, 242–48, 244t–245t, 245f–248f case study on, 246, 246f–248f heat transfer, 242–43

thermal inactivation kinetics, 243–46, 244t–245t, 245f hot-water, 249–57, 253t, 254t apples, 254 beans, 250 broccoli, 250 cantaloupe, 251–54, 253t, 254t commercial application considerations for, 257 fruits, 251–56, 253t, 254t grapes, 255 green onions, 251 lettuce, 249–50 mangos, 256 oranges, 254–55 pears, 256 pineapple, 256 sprouting seeds, 256–57 sprouts, 251 tomatoes, 255–56 vegetables, 249–51 watercress, 251 microwave/RF/infrared, 257–59 Third-party audit, 323 Third-party audit programs, 318, 321–29 current issues with, 322–25 definitions, 322–23 development and implementation of food safety plan, 323–24 documentation for, 325 history of, 321–22 internal audits and, 325–29, 326f, 328f audit day, 327 corrective action reports, 327–28, 328f scheduling of, 326–27 training record, 326f verification with, 325 plant sanitation verification by, 318 preparing for, 323 training for, 324–25 Third-party audit standard, 323 Thyme, antimicrobial edible films with, 228–29 Time-resolved fluorescence approach (TRF), immunomagnetic beads with, 340–41, 341f Time-resolved fluorescence of lanthanide cations, 340 TIR. See Total internal reflection

Index

Tomatoes consumption/imports of, 402t foodborne illness outbreaks with, 105t, 106f, 400t hot-water treatment for, 255–56 internalization of pathogens in, 69–70 manure contamination of, 146t modified atmosphere packaging of, 280t S. Baildon contaminated, 8t S. Braenderup contaminated, 8t S. Javiana contaminated, 8t S. Montevideo contaminated, 8t, 69–70 S. Newport contaminated, 8t S. Typhimurium contaminated, 8t Salmonella contaminated, xvii, 5, 6, 10, 12, 69–70 Shigella flexneri contaminated, 9t Total internal reflection (TIR), 332 Total management commitment, 110, 110f Total plate count (TPC), 253 TPC. See Total plate count Traceback, 316 Training record, 326f TRF. See Time-resolved fluorescence approach Turkeys, E. coli contaminated, 17t U. S. Army-Natick Soldier RD&E Center, Department of Defense Combat Feeding Directorate (NSRDEC/ CFD), 264 UFPA. See United Fresh Produce Association United Fresh Produce Association (UFPA), 307 USDA Agricultural Marketing Service study, 14, 14t V. cholerae, manure/animal source of, 146t Vacuum infiltration, 63 Vegetables. See specific vegetables Washing, produce acidified sodium chlorite, 171t, 178–79 chlorine and chlorine compounds, 173–75 chlorine dioxide, 171t, 177–78 consumer handling practices of, 298 electrolyzed water, 175–76, 175t factors influencing, 170–73, 171t

445

microbial type, level and attachment, 172–73 produce surface properties, 173 washing conditions, 170–71 washing conditions—antimicrobial concentration, 170 washing conditions—pH, 171 washing conditions—sanitizer flow hydrodynamics, 171 washing conditions—temperature, 170 washing conditions—time, 170 water quality, 171–72, 171t water quality—organic loading, 171–72, 171t water quality—water pH and chemistry, 172 hydrogen peroxide, 171t, 182–83 organic acids, 183–84 ozone, 171t, 181–82 peroxyacetic acid, 171t, 179–81 produce safety with, 169–84 Water acidic electrolyzed, 172, 175–76 E. coli with, 24t electrolyzed, 175–76, 175t electrolyzed neutral, 176 foodborne pathogens in, 130–34, 133t groundwater, 130–31 pathogen survival in water, 133–34, 133t surface water, 131–33 hot-water treatment, 249–57, 253t, 254t irrigation, 136–38 fecal contamination in, 136–38 IMB for detection of pathogens in, 342–44, 343f preharvest produce contamination from, 45f, 46–47 standards for quality of, 138 microbial water quality, 110f, 111 process, 138–39 produce safety with, 129–39 quality of aqueous antimicrobial treatment, 171–72, 171t Salmonella with, 24t sanitation, 311–12 testing, 134–36, 134t definitions, 134t standards/criteria for indicators in, 135–36

446

Index

Watercress, hot-water treatment for, 251 Watermelon, consumption/imports of, 402t Watersheds fecal indicators of contamination in, 26–28 incidence of pathogens in, 21–23, 22t Waxes, 295 Whey protein isolate (WPI), human pathogen effected by, 232t White grape juice, thermal resistance of, 244t Wild birds, Salmonella enterica contaminated, 18t Wild tortoises, Salmonella enterica contaminated, 18t Winterbottom v. Wright, 386–87 Worker health/hygiene, 110f, 112

WPI. See Whey protein isolate X-ray, 192t Yam starch, human pathogen effected by, 232t Yeast, biological control of pathogens with, 213–15 Yersinia, incidence on produce of, 11 Yersinia enterocolitica growing at refrigerator temperatures, 107 number of U.S. outbreaks of, 9 Yersinia pseudotuberculosis carrots contaminated with, 9t, 10 lettuce contaminated with, 9t manure/animal source of, 146t Zebu, E. coli contaminated, 17t

Journals of the Institute of Food Technologists

Discounted Rate for IFT Members!

Available Free Online!

Available Free Online!

Journal of Food Science

Comprehensive Reviews in Food Science and Food Safety

Journal of Food Science Education

Edited by: Daryl B. Lund

Edited by: Manfred Kroger

Edited by: Grady Chism

Print ISSN: 0022-1147 Online ISSN: 1750-3841 Frequency: 9 times per year

Online ISSN: 1541-4337 Frequency: Quarterly

Online ISSN: 1541-4329 Frequency: Quarterly

2007 Impact Factor: 2.211

The Journal of Food Science Education serves as an international forum aimed at all those committed to the improvement of food science education, including primary, secondary, undergraduate and graduate, continuing, and workplace education.

2007 Impact Factor: 1.255 The goal of the Journal of Food Science is to offer scientists, researchers, and other food professionals the opportunity to share knowledge of scientific advancements in the myriad disciplines affecting their work, through a respected peer-reviewed publication. For more information, or to subscribe, please visit www.interscience.wiley.com/ journal/jfds

Comprehensive Reviews in Food Science and Food Safety provides coverage of a narrowly defined topic on any food science or food safety aspect, including nutrition, engineering, microbiology, sensory evaluation, physiology, genetics, economics, regulations, and history. For more information, or to read this journal FREE online, please visit www.interscience.wiley.com/ journal/crfs

For more information, or to read this journal FREE online, please visit www.interscience.wiley.com/ journal/jfse

Join the IFT! The Institute of Food Technologists is a nonprofit scientific society with 22,000 members working in food science, food technology, and related professions in industry, academia, and government. For information on joining THE society for food science and technology, please visit www.ift.org

www.wiley.com/go/IFT 1