Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment (Food Microbiology and Food Safety)

Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment

Food Microbiology and Food Safety Series Food Microbiology and Food Safety publishes valuable, practical, and timely resources for professionals and researchers working on microbiological topics associated with foods, as well as food safety issues and problems. Series Editor Michael P. Doyle, Regents Professor and Director of the Center for Food Safety, University of Georgia, Griffith, GA, USA Editorial Board Francis F. Busta, Director, National Center for Food Protection and Defense, University of Minnesota, Minneapolis, MN, USA Bruce R. Cords, Vice President, Environment, Food Safety & Public Health, Ecolab Inc., St. Paul, MN, USA Catherine W. Donnelly, Professor of Nutrition and Food Science, University of Vermont, Burlington, VT, USA Paul A. Hall, President, AIV Microbiology and Food Safety Consultants, LLC, Hawthorn Woods, IL, USA Ailsa D. Hocking, Chief Research Scientist, CSIRO—Food Science Australia, North Ryde, Australia Thomas J. Montville, Professor of Food Microbiology, Rutgers University, New Brunswick, NJ, USA R. Bruce Tompkin, Formerly Vice President-Product Safety, ConAgra Refrigerated Prepared Foods, Downers Grove, IL, USA

For other titles published in this series, go to www.springer.com/series/7131

Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment Jeffrey L. Kornacki Editor Kornacki Microbiology Solutions, Inc., P. O. Box 163, McFarland, WI 53558, USA

123

Editor Jeffrey L. Kornacki Kornacki Microbiology Solutions, Inc. P. O. Box 163 McFarland, WI 53558 USA [email protected]

ISBN 978-1-4419-5517-3 e-ISBN 978-1-4419-5518-0 DOI 10.1007/978-1-4419-5518-0 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2010921492 © Springer Science+Business Media, LLC 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

Preface

This book is borne out of many experiences with many people in hundreds of food processing facilities mainly in North America. The principles of food safety and food quality microbiology range from simple to complex, as does the experience of those in charge of maintaining food safety and quality. Many in our culture assume that the principles are simple, like, “Wash your hands after using the rest room.” However, in my experience, the vast majority of companies that produce foods contaminated with pathogenic bacteria or spoilage organisms are not willfully negligent as some may think. Rather companies often fail for other reasons which could include equipment inadequately designed for appropriate sanitation, poorly constructed facilities, and paradigms that prevent them from recognizing true microbiological risk. Quality assurance/food safety managers become the point persons to deal with food contamination situations. It is a heavy burden that you bear and this book, while useful to anyone wishing to investigate sources of contamination in food processing facilities, is really written with you in mind. This book is written in hopes that it makes your load lighter, your confidence greater, and the food your company produces safer. McFarland, WI

Jeffrey L. Kornacki

v

Acknowledgments

The technical editor of this book, Dr. Kornacki, wishes to express his heartfelt thanks to a number of individuals who were instrumental in forming his perspectives on in-plant microbiological investigations. First I want to thank my father, Thomas Kornacki, a retired criminal investigator of 35 years, whose investigations were quite a bit more dangerous but no less complex in their own way. Thanks for your amazing example of observation, logic, and persistence; you are one of a kind. I hope some of your rigorous approach to problems has been passed down to me. I also want to thank Emeritus Professor Wayne Becker, who awakened me to a fascination with living cells and then Emeritus Professor Robert Deibel, both of whom taught inspired courses that I attended in cell biology and Food Bacteriology, respectively, in the 1970s. These courses created in me a passion to one day become a microbiological investigator and troubleshooter in the food industry. Heartfelt thanks are due my graduate Major Professor, the late Elmer Marth, who very graciously allowed me to pursue graduate degrees under his guidance and leadership. How little I knew then the practical value of his insights, vast contribution to the field of food microbiology, and incredible example of diligence, thoroughness, and analysis. I also owe a great debt of gratitude to Dr. Damien Gabis and Dr. Russell Flowers of Silliker Laboratories whose philosophy, example, and approaches matured me immensely as a microbiological troubleshooter through our many food processing plant visits together in the late 1980s and early 1990s. Thanks are also due to Dave Evanson, Dr. Richard Smittle, and Steve Decker also of Silliker for the same reason. Apart from these individuals, their insights and guidance, the excellent contributions of the authors of this book, and the fine example and encouragement of many other professional colleagues through the years, this book would not have been possible. I would be remiss to omit the fine contribution of the authors that contributed to Chapter 2, many of whom are of national and international reputations. I wish to recognize the fine contributions of Robert Behling (Bacillus cereus), Joseph Eifert (Arcobacter), Marilyn Erickson (Clostridium botulinum), Joshua Gurtler (Cronobacter sakazakii), Erick Line and Bradley Stawick (Campylobacter), Roy Radcliff (Mycobacterium paratuberculosis), Elliot Ryser (Listeria monocytogenes), vii

viii

Acknowledgments

and Drs. Ryser and Yan (Staphylococcus aureus and Clostridium perfringens). I am also very grateful for the fine work of Drs. Moorman, Pruett and Weidman with Chapter 10. May you all be kings! (Proverbs 25:2) Thanks also to Mr. Dwight Clough for his considerable assistance with many of the nuts and bolts of putting this book together. McFarland, WI September 9, 2009

Jeffrey L. Kornacki

Contents

1 Troubleshooting Costs . . . . . . . . . . . . . . . . . . . . . . . . Jeffrey L. Kornacki 2 Selected Pathogens of Concern to Industrial Food Processors: Infectious, Toxigenic, Toxico-Infectious, Selected Emerging Pathogenic Bacteria . . . . . . . . . . . . . . . Robert G. Behling, Joseph Eifert, Marilyn C. Erickson, Joshua B. Gurtler, Jeffrey L. Kornacki, Erick Line, Roy Radcliff, Elliot T. Ryser, Bradley Stawick, and Zhinong Yan

1

5

3 Solving Microbial Spoilage Problems in Processed Foods . . . . . Rocelle Clavero

63

4 Where These Contaminants Are Found . . . . . . . . . . . . . . . Jeffrey L. Kornacki

79

5 What Factors Are Required for Microbes to Grow, Survive, and Die? . . . . . . . . . . . . . . . . . . . . . . . . . . . Jeffrey L. Kornacki

103

6 Where Do I Start (Beginning the Investigation)? . . . . . . . . . . Jeffrey L. Kornacki

117

7 How Do I Sample the Environment and Equipment? . . . . . . . Jeffrey L. Kornacki

125

8 How Many Samples Do I Take? . . . . . . . . . . . . . . . . . . . Jeffrey L. Kornacki

137

9 When Can I Start Up My Factory or Processing Line Again? . . . Jeffrey L. Kornacki

147

10 Value and Methods for Molecular Subtyping of Bacteria . . . . . Mark Moorman, Payton Pruett, and Martin Weidman

157

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

175

ix

Contributors

Robert G. Behling, B.S. Behling Food Safety Associates, Madison, WI, USA, [email protected] Rocelle Clavero, Ph.D. Sara Lee Corporation, Downers Grove, IL, USA, [email protected] Joseph Eifert, Ph.D. Department of Food Science and Technology, Virginia Tech, Blacksburg, VA, USA, [email protected] Marilyn C. Erickson, Ph.D. University of Georgia, Griffin, GA, USA, [email protected] Joshua B. Gurtler, Ph.D. USDA ARS, Wyndmoor, PA, USA, [email protected] Jeffrey L. Kornacki, Ph.D. Kornacki Microbiology Solutions, Inc., McFarland, WI, USA, [email protected] Erick Line, Ph.D. USDA-ARS, Athens, GA, USA, [email protected] Mark Moorman, Ph.D. The Kellogg Company, Battle Creek, MI, USA, [email protected] Payton Pruett, Ph.D. The Kroger Company, Cincinnati, OH, USA, [email protected] Roy Radcliff, Ph.D. Department of Applied Sciences, Marshfield Clinic, Marshfield, WI, USA, [email protected] Elliot T. Ryser, Ph.D. Michigan State University, East Lansing, MI, USA, [email protected] Bradley Stawick, M.S. Stawick Laboratory Management, Memphis, TN, USA, [email protected] Martin Weidman, Ph.D. Department of Food Science, Cornell University, Ithaca, NY, USA, [email protected] Zhinong Yan, Ph.D. Mol Industries, Grand Rapids, MI, USA, [email protected] xi

Chapter 1

Troubleshooting Costs Jeffrey L. Kornacki

Abstract Seventy-six million cases of foodborne disease occur each year in the United States alone. Medical and lost productivity costs of the most common pathogens are estimated to be $5.6–9.4 billion. Product recalls, whether from foodborne illness or spoilage, result in added costs to manufacturers in a variety of ways. These may include expenses associated with lawsuits from real or allegedly stricken individuals and lawsuits from shorted customers. Other costs include those associated with efforts involved in finding the source of the contamination and eliminating it and include time when lines are shut down and therefore non-productive, additional non-routine testing, consultant fees, time and personnel required to overhaul the entire food safety system, lost market share to competitors, and the cost associated with redesign of the factory and redesign or acquisition of more hygienic equipment. The cost associated with an effective quality assurance plan is well worth the effort to prevent the situations described.

1.1 The Cost of Food Contamination 1.1.1 The Extent of Microbial Foodborne Illness in America Many have argued that the US food supply is the safest in the world. Nevertheless, the Centers for Disease Control and Prevention (CDC) have estimated that approximately 5,200 deaths, 325,000 hospitalizations, and 76 million cases of foodborne disease occur each year in the United States (Mead et al., 1999). Known pathogens account for an estimated 14 million illnesses, 60,000 hospitalizations, and 1,800 deaths annually (MMWR, 2003). The Foodborne Disease Outbreak Surveillance System of the CDC indicates that over 1,000 outbreaks occur each year. No one can J.L. Kornacki (B) Kornacki Microbiology Solutions, Inc., McFarland, WI, USA e-mail: [email protected]

J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_1, 

1

2

J.L. Kornacki

estimate the human cost in terms of suffering resulting from illness and death of loved ones. Other more quantifiable costs have been estimated (Buzby and Roberts, 1996).

1.1.2 Overview of Costs Associated with Selected Foodborne Pathogens It is difficult to assess the costs associated with foodborne illness. Buzby and Roberts (1996) estimated the costs of seven of these pathogens (Campylobacter jejuni, Clostridium perfringens, Escherichia coli O157:H7, Listeria monocytogenes, Salmonella, Staphylococcus aureus, and Toxoplasma gondii) to be $5.6–9.4 billion (in 1993 dollars) in medical charges and lost productivity alone. Selected foodborne pathogenic bacteria are discussed individually in Chapter 2. Product recalls, whether from foodborne illness or spoilage, result in added costs to manufacturers in a variety of ways. These may include expenses associated with lawsuits from real or allegedly stricken individuals and lawsuits from shorted customers. Other costs include those associated with efforts involved in finding the source of the contamination and eliminating it and include time when lines are shut down and therefore non-productive, additional non-routine testing, consultant fees, time and personnel required to overhaul the entire food safety system, lost market share to competitors, and the cost associated with redesign of the factory and redesign or acquisition of more hygienic equipment. This author is aware of one situation in the United States where a manufacturer in the late 1990s spent $11 million to remodel a portion of their food manufacturing factory and redesign and install some of the processing equipment in the aftermath of a national food infection and recall. Settlements outside of the court room can also be in millions of dollars. Serious damage to a company’s brand name can and has occurred even in instances when the company issuing the recall is not at fault (e.g., in cases where a product has been contaminated by a supplier’s ingredient). Some companies have resorted to changing their name as a result. Damage to a company’s reputation and profit margin can also result in loss of jobs and salary reductions. In some instances, a recall from a competitor has even resulted in market losses due to brand confusion. Food safety is in everyone’s best interest.

1.1.3 Costs Associated with Spoilage and Foods with Microbial Indicators of “Unacceptable Quality” Many of the same costs listed above apply to spoiled food products. Spoiled foods are not necessarily hazardous but are organoleptically unacceptable. Consequently, these foods may be perceived as dangerous resulting in expenses associated with product recalls, lawsuits from allegedly stricken individuals and shorted customers,

1

Troubleshooting Costs

3

efforts to find and eliminate the source(s) of the contamination, non-productive lines, additional testing, consultant fees, lost market share to competitors, and the costs associated with redesign of the facility and/or equipment. Food spoilage is discussed in Chapter 3. There are even economic consequences to otherwise acceptable foods, wherein levels of selected quality indicators (e.g., coliforms) have been exceeded. These situations may result in rejection of product by suppliers. There are frequent disputes between companies over what is the correct number of non-pathogenic contaminants in a sample. Effective sampling is not as straightforward as taking a sample and testing it due to the potential for non-homogeneous microbial populations in the product and there are added expenses to taking and testing the proper number of samples. Sampling plans are addressed in Chapter 8. The cost associated with an effective quality assurance plan is well worth the effort to prevent the situations just described. The savings in terms of maintenance and repair and increased revenues resulting from higher quality foods due to effective microbiological control efforts in processing facilities are often overlooked. Consequently, there is great need to understand how to locate and control microbial contaminants in processed foods and in the environments in which they are produced. That is what the main thrust of this book is about.

References Buzby JC, Roberts T (1996) ERS updates US foodborne disease costs for seven pathogens. ABI/INFORM Global. 3:20–25. http://www.ers.usda.gov/publications/foodreview/sep1996/ sept96e.pdf. Accessed 17 July 2008 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 (Online) 5(5):607–625 http://www.cdc.gov/ncidod/eid/vol5no5/mead.htm. Accessed 17 July 2008 MMWR (2003) Preliminary FoodNet data on the incidence of foodborne illnesses–selected sites, United States, 2002. 52(15):340–343. http://www.cdc.gov/mmwr/preview/mmwrhtml/ mm5215a4.htm. Accessed 17 July 2008

Chapter 2

Selected Pathogens of Concern to Industrial Food Processors: Infectious, Toxigenic, Toxico-Infectious, Selected Emerging Pathogenic Bacteria Robert G. Behling, Joseph Eifert, Marilyn C. Erickson, Joshua B. Gurtler, Jeffrey L. Kornacki, Erick Line, Roy Radcliff, Elliot T. Ryser, Bradley Stawick, and Zhinong Yan Abstract This chapter, written by several contributing authors, is devoted to discussing selected microbes of contemporary importance. Microbes from three categories are described by the following: (1) infectious invasive agents like Salmonella, Listeria monocytogenes, and Campylobacter; (2) toxigenic pathogens such as Staphylococcus aureus, Bacillus cereus, and Clostridium botulinum; and (3) toxico-infectious agents like enterohemorrhagic Escherichia coli and Clostridium perfringens. In addition, emerging pathogens, like Cronobacter (Enterobacter) sakazakii, Arcobacter spp., and Mycobacterium avium subspecies paratuberculosis are also described. In most cases, the discussion includes a description of the organism itself, economic impact of the organism (due to disease, loss of market share, etc.), disease syndromes/infectious process, infectious dose, reservoirs (where the organism originates in the food processing chain), foods associated with the organism, and the occurrence of the organism in food processing environments.

2.1 Introduction This chapter will not address all the pathogenic microbes that are of concern in all foods or all food processing environments. However, selected pathogens will be described which illustrate typical organism types (i.e., infectious, toxigenic, toxicoinfectious) of common concern in food manufacturing environments. A few selected emerging foodborne pathogens will also be discussed. Detailed reviews and descriptions of foodborne pathogens can be found in a number of references (Doyle, 1989; Jay et al., 2005; Doyle et al., 1997). The later part of this chapter will address selected emerging microbial pathogens of concern. R.G. Behling (B) Behling Food Safety Associates, Madison, WI, USA e-mail: [email protected] J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_2, 

5

6

R.G. Behling et al.

Post-process contamination from the factory environment is a very common (and in this authors opinion – the most common) means by which commercially processed foods are contaminated (Kornacki, 2000; Allan et al., 2004; Reij and Den Aantrekker, 2004). Examples of post process contamination from the food processing environment are illustrated in Table 2.1. In this book we refer to “commercially processed” foods as those which have been modified from the raw state into a ready-to-eat format in an industrial manufacturing environment. Environmental contamination can come from ingredients used in processing, whether directly or indirectly, worker’s hands, shoes, walls, floors, and a myriad of other sources. This chapter is devoted to discussing selected microbes of contemporary importance. These microbes fit into three categories which include (1) infectious invasive agents like Salmonella (Section 2.2), Listeria monocytogenes (Section 2.3), Campylobacter (Section 2.4), and enteroinvasive Escherichia coli, (2) toxigenic pathogens like Staphylococcus aureus (Section 2.5), Bacillus cereus (Section 2.6), and Clostridium botulinum (Section 2.7). (The growth and production of pre-formed toxin in foods is a particular concern with enterotoxin producing strains of Staphylococcus, B. cereus, and C. botulinum.) (3) toxico-infectious agents like enterotoxigenic and enterohemorrhagic E. coli (Section 2.8) and Clostridium perfringens (Section 2.9) are described. In addition emerging pathogens, like Cronobacter (Enterobacter) sakazakii (Section 2.11), Arcobacter spp. (Section 2.10), and Mycobacterium avium subspecies paratuberculosis (Section 2.12), are also described. Infectious vs. toxigenic bacterial pathogens In general, infectious pathogens may enter the body and invade or colonize host tissues. This requires some time (e.g., usually greater than 8 h for onset of illness). Toxigenic pathogens create food “poisoning” situations by producing an enterotoxin in the food. Incubation times for onset of disease from toxigenic microbes are often shorter than for invasive pathogens and can be as little as 1 h, as in the case of staphylococcal enterotoxin-induced illness. The short incubation time in comparison to the infectious pathogens results because the agent of illness, the toxin, is pre-formed in the food and ingested. Illness is not contingent upon the organism migrating to the intestinal tract implanting and growing. Selected examples of invasive and infectious foodborne pathogens and their importance in various foods follow.

2.2 Salmonella, an Infectious Invasive Agent Salmonella spp. are an example of an “invasive” infectious pathogen and are second only to the thermophilic Campylobacter spp. (e.g., jejuni, coli) in the number of foodborne disease cases per year attributed to bacteria (Table 2.2).

C. jejuni S. enteritidis

S. ealing

S. berta S. typhimurium S. Napoli S. eastbourne L. monocytogenes L. monocytogenes C. botulinum S. aureus E. coli O157:H7 Y. enterocolitica S. typhi S. aureus S. aureus E. coli O157:H7 S. Enteritidis PT4

S. m˝unchen S. typhimurium E. coli O157:H7 B. cereus Y. enterocolitica L. monocytogenes

Tuna salad Ice cream

Infant formulae

Soft cheese Cooked sliced ham Chocolate Chocolate Butter Hot dogs Canned salmon Lasagne Different foods Chocolate milk Canned meat Crabmeat Canned mushrooms Flavored yogurt Pastry

Yeasts Pasteurized milk Pasteurized milk Pasteurized milk Pasteurized milk Mexican type cheese

Adapted from Reij and Aantrekker (2004); and ICMSF (2002).

Pathogen

Product

Probably chicken handled in same kitchen Pasteurized ice cream mix in tanker truck previously used for transporting raw liquid eggs Contamination from the processing environment, insulation material of the drying tower Cheese ripening in buckets previously used for chicken carcasses Cooked ham placed into containers previously used for curing Possibly contaminated water used in double-walled pipes, tanks Contamination from the processing environment Contamination from the processing environment Contamination from the processing environment Contamination from the processing environment, cooling water Growth of S. aureus in the processing equipment, improper cleaning Contaminated meat grinder and equipment at retail level Probably during manual mixing of pasteurized milk and chocolate Use of non-potable water for can cooling Contamination during manual picking of cooked meat Possible growth of S. aureus in the brine bath before canning Pump previously used for raw milk Equipment previously used for raw eggs or insufficiently cleaned piping and nozzles used for cream Contamination from the processing environment Possibly cross-connection between raw and pasteurized milk Contamination from pipes and rubber seals of the bottling line Filling equipment Post-process contamination Contamination from the processing environment

Comment

Table 2.1 Examples of outbreaks attributed to environmental contamination

2 Selected Pathogens of Concern to Industrial Food Processors 7

8

R.G. Behling et al. Table 2.2 Estimated annual foodborne disease from selected pathogens

Bacterium

Number of total illness

Total illness (%)

Hospitalized (%)

Deaths (%)

Number of deaths

Campylobacter Salmonella (non-typhoidal) Listeria monocytogenes E. coli O157:H7 Clostridium perfringens Staphylococcus Shigella

2M 1.3 M 2,493 62,458 248,520 185,060 89,648

14.2 9.7 0.0 0.5 1.8 1.3 0.6

17.3 25.6 3.8 3.0 0.1 2.9 2.0

5.5 30.6 27.6 2.9 0.4 0.1 0.8

99 553 499 52 7 2 14

Adapted from Mead et al. (1999). M = million.

2.2.1 Salmonella: The Organism Salmonella species are gram-negative, rod-shaped, usually motile, members of the taxonomic family, Enterobacteriaceae. Despite great advances in molecular genetic approaches to identification and characterization these organisms are still serologically defined, i.e., by their somatic (O) and (usually) flagellar (H) and sometimes capsular (Vi) antigens. Over approximately 2,400 different serotypes are known to exist. The nomenclature of this microbe has gone through a number of changes resulting in some confusion. In this author’s opinion it is easiest to refer to the serotype designation (e.g., Salmonella serotype Typhimurium) as opposed to other nomenclatural approaches (e.g., Salmonella enterica serovar Typhimurium).

2.2.2 Cost Costs that are difficult to measure include pain and suffering, death, and loss of a company’s reputation. Other costs may include lost market share, lost jobs or reduced wages, lawsuits from shorted customers, the price to remanufacture products that have been destroyed, the cost to recondition contaminated product (if possible and allowed), and lawsuits from stricken individuals or class actions (see Chapter 1). The United States Department of Agriculture (USDA) estimated that 696,000–3,800,000 cases of non-typhoid, foodborne Salmonellosis occurs annually with an estimated cost of 0.9–12.2 billion dollars (Buzby and Roberts, 1996).

2.2.3 Disease Syndromes Salmonella can cause a number of disease syndromes including typhoid fever from Salmonella typhi (rarely found in foods produced in the United States). However,

2

Selected Pathogens of Concern to Industrial Food Processors

9

other strains of Salmonella cause gastroenteritis, bacteremia, and enteric or paratyphoid fever (Hanes, 2003). Onset times typically range from 18 to 36 h (IAFP, 1999). Symptoms include abdominal pain, diarrhea, occasionally with mucous or blood. Nausea and vomiting often occur but are rarely severe or protracted. A fever of 38–39◦ C is common, often after a chill. In many instances the disease resolves within 48 h. However, it can last with diarrhea and low-grade fever for 10–14 days. In severe cases dehydration may lead to hypotension, cramps, oliguria, and uremia. Symptoms are often more severe in infants and adults over 60 years of age. Fatalities rarely exceed 1% of the affected population and are generally limited to infants, elderly, and debilitated individuals (Hanes, 2003). Nevertheless, Salmonella infection accounts for more foodborne deaths (31%) in the United States than any other foodborne pathogen (Mead et al., 1999). Furthermore, multi-drug-resistant Salmonella DT104 has been associated with double the hospitalization rate and ten times the case fatality rate of other foodborne Salmonella (Hanes, 2003). The presence of viable salmonellae in the gastrointestinal tract indicates that the organism survived a variety of non-specific host defenses including lactoperoxidase in saliva, stomach acidity, mucous secretions from intestinal goblet cells, and sloughing of luminal epithelial cells. In addition, they must survive non-specific phagocytic cells, immune responses associated with specific T and B lymphocytes, Peyer’s patches, and complement inactivation (D’Aoust, 1991). Once they have survived these conditions, they attach to intestinal tissues and mesenteric lymph follicles resulting in enterocolitis. Endotoxin is produced, leukocytes move into the infected tissues, and increased mucous secretion occurs. Mucosal inflammation results from the release of prostaglandins by leukocytes which also activates adenyl cyclase in intestinal epithelial cells causing increased fluid secretion into the intestinal lumen and resulting in diarrhea. Septicemia and other chronic conditions result when host defenses fail to keep these invasive Salmonella in check (D’Aoust, 1991).

2.2.4 Infectious Dose The infectious dose appears to be very low as evidenced by some foods implicated in foodborne disease with only a few cells recovered. An example of this occurred in a 1994 frozen dessert-associated outbreak wherein the level of Salmonella serotype Enteritidis was reported to have a most probable number (MPN) range of 4 cells per 1,000 g to 46 cells per 100 g with a median of 93 cells in 1,000 g (Vought and Tatini, 1998). The 95% confidence interval for these MPN values was from < 1 cell per 1,000 g to 2.4 cells/g. The number of Salmonella serotype Enteritidis cells per serving was estimated at 25 cells. In this study, the infective dose appeared to be less than 28 cells. Evidence from other studies indicates that from 1 to 10 cells may constitute an infectious dose in some circumstances (D’Aoust et al., 1985 and Kapperud et al., 1990).

10

R.G. Behling et al.

2.2.5 Reservoirs Salmonellae are widespread in the natural environment and a number of these are host specific (e.g., Salmonella serotype Pullorum in chickens, Salmonella serotype Cholera-Suis in pigs). In many countries poultry remain the dominant reservoir, although pork, beef, and mutton have served as vehicles of infection. The eggborne pandemic of Salmonella serotype Enteritidis phage type 4 in Europe and phage type 8 in North America illustrates the importance of poultry products as vehicles of human salmonellosis.

2.2.6 Foods Associated with Human Salmonella spp. Infection Salmonella spp. have a long history of food contamination and have caused illness from ingestion of a wide variety of foods. These organisms have been a particular concern with foods of animal origin (e.g., meat, poultry, eggs, and dairy products). Dry foods and fruit and vegetableborne outbreaks have also occurred. One multistate cantaloupeborne outbreak affected more than 25,000 across 30 states (Ries et al., 1990). Major outbreaks have occurred with chocolate, milk powder, potato salad, egg salad, raw milk, mustard dressing, salad base, cheddar cheese, liver pate, aspic glaze, pasteurized milk, egg drink, cuttlefish, cooked eggs, cantaloupes, fruit soup, mayonnaise, paprika chips, ice cream, and alfalfa sprouts (D’Aoust, 1997). In the 1994 frozen dessert-associated outbreak mentioned above, an estimated 224,000 people were infected with Salmonella serotype Enteritidis. The source of the organism was traced to contaminated post-pasteurized ice cream mix which had been shipped in tank trucks previously used to transport raw eggs (Hennessy et al., 1996).

2.2.7 Dry Foods Dry products are not often associated with microbial contamination problems. However, salmonellae have been a particular concern with some dry foods and dry food production environments. Control of these organisms is a priority among industries that produce dried foods such as dry milk, infant formula, chocolate, dry soup mixes, and rendered animal proteins (an ingredient in animal feed and pet food). Human outbreaks of disease have been reported with dry milk, chocolate, and even paprika potato chips, dry cereal, and peanut butter (D’Aoust et al., 1975; Greenwood and Hopper, 1983; Kapperud et al., 1990; Lehmacher et al., 1995; Weissman et al., 1977; CDC, 2007, 2009). It is important to note that simply because a microbe cannot grow in a low water activity food, it may still survive for some time. Factors that influence microbial growth survival and death are discussed in Chapter 5.

2

Selected Pathogens of Concern to Industrial Food Processors

11

2.2.8 Food Processing Environments A wide variety of food factory environments may be contaminated with this microbe due to their widespread occurrence in the natural environment and likely presence in some raw ingredients which may enter factory environments. In the author’s experience, birds, which are frequent carriers of this organism, may find roosts on factory roofs or roof-associated structures (e.g., air intakes for air handling units). Most food processing facilities have flat roofs and many are not adequately sloped to drains resulting in collection of standing water. Standing water on roof tops will permit the growth of salmonellae to high numbers. Entry of salmonellae in the factory environment may occur through inappropriately sealed roof top-associated penetrations. Other routes of entry are described in Chapter 4. In the author’s experience, it is rare that the post-cook side of a factory is contaminated with more than one serotype of Salmonella. Most often the Salmonella serotype found on the finished side of the facility is its “signature” organism or “house-bug.” It appears that each environment selects for the strain which has likely adapted to its environment. However, exceptions to this observation have been noted and the presence of multiple serotypes suggests multiple sources for the microbe. The author has observed sporadic detection of a specific serotype of Salmonella in some dry product processing environments for 10 or more years.

2.3 L. monocytogenes, an Infectious Invasive Agent The very widespread (some say ubiquitous) distribution of this organism in the natural environment coupled with its resistance to freezing, growth in the presence of 10% salt, survival in concentrated brine solutions, and its ability to grow at 1–45◦ C (optimum at 35–37◦ C) makes control of this organism in the processing environment challenging. Eradication of this organism from ready-to-eat meat and poultry processing environments is unlikely given current technology (Tompkin et al., 1999). Hence, implementation of rigorous controls is essential to prevent processed food contamination.

2.3.1 The Organism L. monocytogenes is a gram-positive, short (0.4–0.5×0.5–2 μm) non-sporeforming, rod-shaped, microaerobic bacterium that exhibits tumbling motility. The organism appears translucent with a “characteristic” blue-green sheen when observed under oblique lighting. However, some technicians are better than others at visualizing this phenomenon. It is typically weakly β-hemolytic on horse blood agar and exhibits a characteristic CAMP reaction on sheep blood agar when streaked perpendicularly to S. aureus (enhanced hemolysis) and Rhodococcus equi (hemolysis not enhanced). Other than Listeria seeligeri, the remaining Listeria spp.

12

R.G. Behling et al. Table 2.3 Key reactions of Listeria

Camp reaction (enhanced hemolysis) Species

Staphylococcus streak

Rhodococcus equi streak

Rhamnose

Xylose

monocytogenes ivanovii innocua welshimeri seeligeri grayi b

+ – – – + –

–a + – – – –

+ – Variable Variable – Variable

– + – + + –

a Rare strains may show a region of enhanced hemolysis near both the Staphylococcus and the Rhodococcus streaks b This species is mannitol positive

yield different CAMP reactions (Table 2.3). Typical L. monocytogenes isolates ferment rhamnose, dextrose, esculin, and maltose but not xylose and mannitol (Datta, 2003). The CAMP reaction and key biochemical reactions have been traditionally used to confirm cultures. However, numerous Listeria identification test kits are now available that greatly expedite this process without the use of the CAMP reaction.

2.3.2 Cost Approximately 2,518 cases of foodborne listeriosis, including ∼500 fatalities, occur annually in the United States at an estimated cost of $2.3 billion, making listeriosis the second most costly foodborne illness after salmonellosis which some have estimated at 2.33 billion dollars (Mead et al., 1999; Buzby and Roberts, 1996). Consequently, foodborne listeriosis has been targeted by many public heath programs, most notably Healthy People 2010 – a comprehensive nationwide health promotion and disease prevention program developed by the Department of Health and Human Services to reduce bacterial infections and enhance life expectancy/quality.

2.3.3 Disease Syndromes Two types of listeriosis are recognized – (a) an invasive form that can be lifethreatening in newborn infants, the elderly, and immunocompromised adults and (b) a less common self-limiting gastrointestinal illness. In the gastrointestinal form, flu-like symptoms (e.g., diarrhea, vomiting, fever) may occur 18–24 h after ingestion of the contaminated food. In contrast, invasive listeriosis has an onset time of 3 to as long as 70 days after which adults typically experience septicemia, meningitis, or endocarditis, whereas unborn fetuses develop abscesses in their liver, lungs,

2

Selected Pathogens of Concern to Industrial Food Processors

13

and other organs that often result in spontaneous abortion and stillbirth. Surviving children may be seriously ill with meningitis and neurological impairment.

2.3.4 The Infectious Process Once the bacterium enters the host’s monocytes, macrophages, or polymorphonuclear leukocytes, it is bloodborne (septicemic) and can grow. Its intracellular presence with phagocytic cells also permits access to the brain and probably transplacental migration to the fetus in pregnant women. The pathogenesis of L. monocytogenes centers on its ability to survive and multiply in phagocytic host cells (FDA “Bad Bug Book” http://vm.cfsan.fda.gov/∼mow/chap6.html).

2.3.5 Infectious Dose The infectious dose of the organism, as with other pathogens, will be related to the virulence of the particular strain and the host’s susceptibility. In fact there is an estimated 2,584-fold greater risk of Listeriosis among transplant patients than in healthy individuals under the age of 65 (ILSI, 2005). Controversy exists about the infectious dose. It is clear that most Listeriosis results from ingestion of very high numbers of the organism with 82.9% of cases attributed to ingestion of foods with greater than 1 × 106 cfu, it remains unknown if there is a minimum level that can cause illness (FDA, USDA, 2003; http://www.fda.gov/Food/ScienceResearch/ResearchAreas/RiskAssessment SafetyAssessment/ucm183966.htm). However, the ICMSF has reported that “Epidemiologic data indicate that foods involved in listeriosis outbreaks are those in which the organism has multiplied and in general have contained levels well in excess of 100 cfu/g and proposed that the concentration of L. monocytogenes in frankfurters not exceed 100 cfu/g at the time of consumption” (ICMSF Volume 7, 2002a, p. 294). Nevertheless because a single microbe has the potential to cause illness (likely in a very debilitated host). The US government has tolerance of negative in 25 g or 0.4 cfu/g. In addition the 2003 FDA/USDA Listeria risk assessment (http://www.fda.gov/ Food/ScienceResearch/ResearchAreas/RiskAssessmentSafetyAssessment/ucm1839 66.htm) indicated that up to 10 million-fold differences in risk between various foods. The highest risks were found for deli meats and frankfurters with very low risks found for cultured milk products, process cheese, hard cheese, ice cream, and other frozen dairy products. A consequence of this risk assessment has been a proposed tolerance in selected ready-to-eat foods of <100 cfu/g L. monocytogenes (FDA, 2008a, b). Essentially these draft guidelines, if approved, will be for foods in which L. monocytogenes cannot grow due to low pH (≤ 4.4), low water activity (≤ 0.92), or formulated to prevent L. monocytogenes from growing. The documents recognize that frozen

14

R.G. Behling et al.

foods will prevent Listeria growth and would be acceptable if they contain less than 100 cfu/g of L. monocytogenes if consumed in the frozen state. Hence, the proposed tolerance level may apply to ice cream that is consumed in the frozen state but may not apply to frozen items which must be thawed prior to consumption. Nevertheless non-exempt foods with less than 100 cfu/g made under insanitary conditions would still, likely, be subject to regulatory consequences.

2.3.6 Reservoirs and Implicated Foods Listeria is widespread in the natural environment having been found in soil, water, sewage, decaying vegetation, humans, domestic animals (including pets), raw agricultural commodities, food processing environments, and the home (Ryser and Marth, 1999). The microbe has been found in a wide variety of foods including meats, poultry, dairy products, vegetables, and seafoods. In fact, invasive outbreaks of Listeriosis have been reported from consumption of a variety of products including raw milk, sour milk, cream, cottage cheese, pasteurized milk, cheese (blue-mold, hard, Brie de Meaux, Vacherin Mont d’Or, Mexican-style, Pont l’Eveque, raw milk cheese), butter, pork, pork tongue, delicatessen turkey meat, Pâte, processed meats, pork tongue in aspic (jelly), Rillettes, mousse, raw vegetables, coleslaw, shellfish, shrimp, and raw eggs (Norton and Braden, 2007). Since 1998, over 130 recalls involving more than 80 million of pounds of readyto-eat meat and poultry products were issued due to Listeria contamination with three of these recalls related to major outbreaks of listeriosis. Virtually all of these recalls have been attributed to post-processing contamination at the manufacturing facility.

2.3.7 Food Processing Environments The extremely widespread nature of the organism in the environment, its psychrotrophic growth, resistance to other stress as compared to other non-spore-forming pathogens (e.g., freezing, salt, heat) makes it a particularly difficult microbe to control in a wide variety of food processing facilities. A review of the incidence and control of Listeria in food processing facilities can be found in Kornacki (2007).

2.4 Campylobacter, an Infectious Invasive Agent Campylobacter species are enteric pathogens and are considered one of the leading foodborne disease agents in the United States causing an estimated 2.1–2.4 million cases of gastroenteritis annually (Altekruse et al., 1999).

2

Selected Pathogens of Concern to Industrial Food Processors

15

2.4.1 The Organism Campylobacter species are gram-negative, spiral-shaped rods and typically motile. The Campylobacter genus consists of 14 species. The most common foodborne species are Campylobacter jejuni and Campylobacter coli. Members of the genus are susceptible to environmental stresses and are considered to be relatively fragile. Because of these concerns, this organism can be difficult to isolate in the laboratory.

2.4.2 Costs Like any foodborne disease agent, it can be difficult to measure the exact costs related to an outbreak, especially those related to pain and suffering, losses related to the reputation of the manufacturer, market share loss, but also lost jobs, and lawsuits. According to the United States Department of Agriculture, campylobacteriosis costs in the United States are estimated to be 1.2–6.6 billion dollars annually (Buzby and Roberts, 1996).

2.4.3 Disease Syndromes The most common type of illness caused by Campylobacter spp. is gastroenteritis referred to as campylobacteriosis. Enteric symptoms are caused by a thermolabile toxin (CJT) (Ray, 1996). This toxin is similar to cholera toxin and the LT toxin of enterotoxigenic E. coli (Smith, 1995). Symptoms of campylobacteriosis include diarrhea, abdominal pain or cramps, headache, muscle pain, and fever (Jay et al., 2005). Diarrhea may also be bloody. The onset of symptoms typically occur within 2–5 days of exposure and illness usually lasts 7–10 days, with relapses occurring in 25% of cases (US FDA, 2009). Infections are usually self-limiting, and treatments typically include fluid and electrolyte replacement. Antibiotic treatments may be used in severe cases (Altekruse et al., 1999). Most common groups afflicted include children less than 5 years of age and young adults 15–29 years. The fatality rate is approximately 0.1%. Over the last several years, it has been shown that males are afflicted more often than females with campylobacteriosis (Franco and Williams, 1994; CDC, 2004). A serious neurological complication called Guillain–Barré syndrome (GBS) can occur in a small percentage of patients following infection with some C. jejuni strains. GBS is an acute inflammation of peripheral nerves. GBS may cause fever, pain, and weakness and can also lead to paralysis (Keener et al., 2004). C. jejuni serovar O:19 and serotypes associated with Guillain–Barré syndrome are considered by the International Commission on Microbiological Specifications for Foods (ICMSF) under the category “Severe Hazard-Restricted Populations” (ICMSF, 2002a).

16

R.G. Behling et al.

It is estimated that about a quarter of GBS patients experienced a recent C. jejuni infection in the months preceding onset of GBS symptoms. Most GBS patients recover fully; however, about 20% continue to suffer varying degrees of disability (Hughes and Cornblath, 2005), and 3–8% die (Smith, 1995). There is an increasing rate of antibiotic-resistant Campylobacter. The rate is highest in the developing world. Resistance to ciprofloxacin, azithromycin, and fluoroquinolone has been noted (Altekruse et al., 1999).

2.4.4 Infectious Dose The infectious dose of Campylobacter has been shown to be low. Levels at or below 500 organisms have been shown to cause illness (Keener et al., 2004; Smith, 1995). C. jejuni accounts for approximately 99% of cases (FDA), even though other species can cause human illness.

2.4.5 Reservoirs The primary ecological niche for campylobacters is the intestinal tract of warmblooded animals. Campylobacters replicate almost exclusively within these hosts (Ketley, 1997). The reservoir for infection comprises a wide variety of both wild and domestic animals (Skelly and Weinstein, 2003). C. jejuni is commonly found in the intestinal tract of birds, cattle, and sheep; whereas, C. coli is most often associated with pigs (Doyle, 1944). Domestic pets such as dogs and cats can serve as important reservoirs for campylobacter with reported isolation rates as high as 66% in healthy cats and 34% in healthy dogs (Moreno et al., 1993). Other species such as rodents, flies, and insects may also harbor campylobacters and may serve as vectors of disease for other hosts. Many avian species have been associated with high rates of colonization of C. jejuni including both wild and domestic birds. Campylobacters are frequently isolated from the natural environment typically as a result of fecal contamination from colonized animal hosts. Campylobacters have been isolated from streams, rivers, lakes, sea water, and estuaries that have been tainted with fecal contamination from wild or domestic animals (Knill et al., 1982).

2.4.6 Foods Associated with Campylobacter Campylobacter is most frequently associated with foods of animal origin. Any raw meat from an animal bred for consumption may be contaminated with campylobacters (Butzler, 2004). Raw poultry meat is usually contaminated with Campylobacter spp. and there is sufficient epidemiological evidence to suggest that poultry meat serves as a primary source of Campylobacter infection in humans (Sahin et al., 2002). The prevalence of Campylobacter on raw meat products from other food

2

Selected Pathogens of Concern to Industrial Food Processors

17

animals tends to be lower than that for poultry (Murphy et al., 2006). Beef, pork, lamb, seafood, and shellfish have all been implicated in cases of campylobacteriosis (Kramer et al., 2000; Wilson and Moore, 1996). Campylobacters are not usually associated with vegetables, although they have been detected in spinach, lettuce, radish, green onions, parsley, and potatoes (Park and Sanders, 1992) and some cases of human infection from consumption of contaminated vegetables have been described (Jacobs-Reitsma, 2000). The epidemiology of human campylobacter infection is generally more often associated with numerous, sporadic small-scale infections than large-scale outbreaks. For example, in a study among university students in Georgia, 70% of campylobacter infections could be attributed to eating undercooked chicken (Tauxe, 1992). Poor kitchen hygiene may also play a role: cross-contamination events during handling of contaminated fresh chicken parts have been shown to be a risk factor for infection and illness (Luber et al., 2006). When large outbreaks occur they are most often waterborne disease outbreaks (Sacks et al., 1986; Millson et al., 1991) or are associated with consumption of raw milk or milk that has been contaminated post-pasteurization (Jones et al., 1981; Morgan et al., 1994; Riordan et al., 1993).

2.4.7 Campylobacter and Poultry The strong association between campylobacter and poultry warrants further discussion of specific concerns in the poultry production and processing environments. C. jejuni has been isolated from the reproductive tracts of broiler breeder hens (Hiett et al., 2002) and there is some recent evidence supporting vertical transmission (Cox et al., 2002); the significance of this source for the contamination of chicken flocks is a point of conjecture (Callicott et al., 2006). C. jejuni is not usually isolated from the production environment during the first 2 weeks after the chicks are placed. However, by the third or fourth week of production most flocks are contaminated and the pathogen generally spreads rapidly through most members of the flock (Stern and Line, 2000). The intestinal tract of the chickens may harbor up to 107 cfu C. jejuni g–1 with no apparent harm to the host (Stern et al., 1988). The stresses associated with cooping and transport of chickens from the farm to the processing facility and holding the birds prior to processing have been demonstrated to increase Campylobacter populations on the birds (Stern et al., 1995). During processing, the poultry carcasses may become contaminated with intestinal contents and as a result most raw poultry products are contaminated with C. jejuni (Stern, 1992). To significantly reduce the load of campylobacters entering the poultry processing facility, intervention on the farm during production is required. Carryover of the same strain of Campylobacter from flock to subsequent flock reared in the same house is thought to be a relatively infrequent event (Shreeve et al., 2002). Potential on-farm intervention strategies employing increased biosecurity measures, litter treatment, acidified feed, probiotics, and competitive exclusion products have

18

R.G. Behling et al.

met with mixed success (Wagenaar et al., 2006). Likewise vaccines have been developed for Campylobacter but have not proven completely efficacious. Campylobacter has been demonstrated to be able to form biofilms on a variety of surfaces (Joshua et al., 2006) which increase its chances of survival in the poultry production environment (Trachoo et al., 2002) and emphasizes the need for proper sanitization of water lines. Novel intervention strategies employing bacteriocins or bacteriophage may be useful in the future; however, they are not yet commercially available (Joerger, 2003; Stern et al., 2006). Effective intervention on the farm will likely require a multifaceted approach and it is unlikely that there will be any one “silver bullet” approach to eliminate Campylobacter. In the poultry processing facility there are many critical control points which have been identified to reduce contamination. These include washer and product temperature controls, chemical interventions (including chlorine and trisodium phosphate among others), water replacements and counter-flow technology in the scalder and chiller, and equipment maintenance (White et al., 1997). In some European countries the concept of “scheduled processing” or keeping colonized and non-colonized flock separate during processing is seen as a promising control strategy (Wagenaar et al., 2006), but this may not be applicable to processing conditions in the United States where prevalence of Campylobacter in broiler flocks is very high. Freezing of contaminated poultry can also be utilized to significantly reduce Campylobacter populations as was recently demonstrated in Iceland (Georgsson et al., 2006); however, this is not necessarily a cost-effective strategy in the US market. Unlike other foodborne pathogens, Campylobacter does not grow effectively in the environment or under normal food storage conditions (Park, 2002). There are a number of techniques to control Campylobacter spp. in foods including physical treatments such as heat, cold, dehydration, hydrostatic pressure, and irradiation (Alter and Scherer, 2006). Heat treatment processes designed to kill Salmonella and Listeria will eradicate Campylobacter spp. in similar food matrices as well (Moore and Madden, 2000). A terminal pasteurization step (e.g., irradiation or heat pasteurization) applied under controlled conditions at the processing plant may be the best means currently available for reducing campylobacteriosis (Stern and Line, 2000). Campylobacter are also much less tolerant to osmotic stress than other foodborne pathogens (Doyle and Roman, 1982). Salmonella and Listeria will grow in sodium chloride concentrations of 4.5 and 10%, respectively, whereas Campylobacter strains are not able to grow in the presence of 2% sodium chloride (Alter and Scherer, 2006). Because campylobacters are microaerobic, they are injured by oxidative stress and have an inherent sensitivity toward oxygen (Park, 2002). High hydrostatic pressure can also be used to reduce populations of campylobacters. Solomon and Hoover (2004) demonstrated that C. jejuni populations in inoculated milk or chicken puree could be reduced by 2–3 log units at 300–325 MPa, while treatment at 400 MPa completely inactivated the pathogen. Campylobacters are also sensitive to pH with a pH value above 9.0 or below 4.0 leading to rapid decreases in populations (Gill and Harris, 1983). Campylobacters are susceptible to most routinely used disinfectants including chlorine (Wang et al., 1983). However, the effectiveness of chlorinated water is reduced when the organisms are attached

2

Selected Pathogens of Concern to Industrial Food Processors

19

to organic matter such as chicken carcasses (Alter and Scherer, 2006). Trachoo and Frank (2002) demonstrated that chlorine was the most effective sanitizer for inactivation of Campylobacter in biofilms. Campylobacter may not be the best choice for monitoring in processing plant environmental samples as they are typically more fragile than other foodborne pathogens or indicator organisms and should not survive a thorough cleaning and sanitization regimen appropriate for removal of more fastidious microorganisms.

2.5 Staphylococcus, a Toxigenic Pathogen S. aureus is a common bacterial pathogen causing staphylococcal food poisoning (SFP) – a leading cause of foodborne intoxication worldwide – and accounts for an estimated 14% of all foodborne illnesses in the United States. SFP is not attributed to ingestion of live bacterial cells but rather acquired from ingesting one or more heatstable pre-formed staphylococcal enterotoxins (SEs) in foods contaminated with enterotoxin producing strains of staphylococci, principally, S. aureus. This type of food poisoning is classified as an intoxication since it does not require growth of the bacterium in the host. Indeed, numerous outbreaks have been caused by foods in which the organism has been killed but the heat-stable toxin remained. SEs are unique because they are not destroyed by heating including canning.

2.5.1 The Organism Staphylococci belong to the family Micrococcaceae. They are gram-positive spherical bacteria about 1 μm in diameter that appear as grape-like clusters under the microscope. The grape-like configuration of staphylococci helps to distinguish staphylococci from streptococci that usually form chains because they divide in one plane only. Staphylococci are catalase-positive, oxidase-negative, facultative anaerobes that grow by aerobic respiration or fermentatively with the principal end product being lactic acid. The catalase test is important in distinguishing streptococci (catalase negative) from staphylococci, which are strong catalase producers. In 1884, Rosenbach described the two pigmented colony types of staphylococci and proposed the appropriate nomenclature: S. aureus (yellow) and Staphylococcus albus (white). The latter species was named Staphylococcus epidermidis in 1908. Thirty-one Staphylococcus spp. are currently recognized of which 15 are human pathogens. Among these, S. aureus and S. epidermidis are the most significant in their interactions with humans. S. aureus forms a fairly large yellow colony on rich media. It is often βhemolytic on blood agar, producing a strong zone of lysis. S. aureus grows in a temperature range from 15 to 45◦ C and at NaCl concentrations as high as 16%. It does not survive legal milk pasteurization. Nearly all strains of S. aureus produce the enzyme coagulase (an enzyme closely associated with toxin

20

R.G. Behling et al.

producing strains). However, some coagulase negative staphyloccal strains have been isolated that also product SET. S. aureus should always be considered as a potential pathogen with multi-antibiotic-resistant strains of this organism also a leading cause of post-operative infections in hospitals. Unlike S. aureus, S. epidermidis produces relatively small white colony and is non-hemolytic with nearly all strains lacking the enzyme coagulase. S. epidermidis is generally regarded as being non-pathogenic; however, a few strains have been linked to infections in hospital environments. Multiple antibiotic resistance characterized by resistance to methicillin is increasingly common in both S. aureus and S. epidermidis. Methicillin-resistant S. aureus (MRSA) causes outbreaks in hospitals and is frequently endemic in hospital environments.

2.5.2 Staphylococcal Enterotoxin The current classification of staphylococcal enterotoxins (SEs) is based on antigenicity with each enterotoxin designated by letter in order of discovery. Twelve SEs have been thus far identified and include SE A, B, C1 , C2 , C3 , D, E, F, G, H, I, and J. Initially, the designation “SEF” was used to refer to an exotoxin commonly produced by isolates of S. aureus associated with toxin shock syndrome (TSS). However, this toxin was later designated TSS toxin 1 when it was confirmed that SEF was not emetic. The relative incidence of SEs produced by isolates of S. aureus varies. In general, SEA is most common having been implicated in more than 80% of all outbreaks of staphylococcal food poisoning. SED is the second most common and has been most frequently associated with egg and fish products. Few outbreaks have been traced to SEE (Jablonski and Bohach, 2001). Production of SEs is favored at optimal growth temperature, aw , and pH with S. aureus producing less or no SE under suboptimal growth conditions. SEs can be produced at 10–46.6◦ C with 40–45◦ C being best. The minimum aw for SE production is 0.90 whereas S. aureus can grow at aw values as low as 0.84. SE production is favored at pH 5.2–9.0 (optimum 6.5–7.5), whereas the range for the growth is from 4.3 to 9.4. The SEs are quite heat resistant with SEB retaining biological activity after 16 h of heating at 60◦ C/pH 7.3. Furthermore, no change in serological reactivity was found in SEC after 30 min of heating at 60◦ C. Heating of SEA at 80◦ C for 3 min or at 100◦ C for 1 min led to a loss in serological reactivity. Retorted canned mushrooms imported from China in 1989 were shown to be contaminated with heat-stable staphylococcal enterotoxin (SET); (http://findarticles.com/p/articles/mi_m1370/is_n7_v23/ai_8017475; http://www. cdc.gov/mmwR/preview/mmwrhtml/00001410.htm, Brunner and Wong, 1992). SET has been reported to have a D250◦ F value of about 20 min (David et al., 1996). SEs are also resistant to gamma irradiation with about 33% of SEA remaining active in a meat slurry after exposure to 8 kGy. All SEs except SEB are resistant to pepsin at pH 2.0 and therefore survive passage through the stomach.

2

Selected Pathogens of Concern to Industrial Food Processors

21

2.5.3 Cost S. aureus in food causes an estimated 1,513,000 cases of illnesses and 1,210 deaths annually in the United States. Costs associated with S. aureus infections and intoxications are estimated at $6.8 billion from all sources of which $1.2 billion in expenses is attributed to foodborne sources.

2.5.4 Disease Syndromes In humans, S aureus causes various suppurative (pus-forming) infections including superficial skin lesions, boils, styes, and furunculosis as well as more serious infections such as pneumonia, mastitis, phlebitis, meningitis, and urinary tract infections; and deep-seated infections, such as osteomyelitis and endocarditis. S. aureus is a major cause of hospital-acquired post-operative infections and infections associated with indwelling catheters and other medical devices. S. aureus causes food poisoning by releasing enterotoxins into the food and toxic shock syndrome by release of superantigens into the bloodstream. The onset of symptoms in staphylococcal food poisoning is usually rapid, typically within 1–6 h, and is influenced by individual susceptibility to the toxin, amount of contaminated food consumed, amount of toxin in the food ingested, and general health status. The most common symptoms are nausea, vomiting, abdominal cramping, diarrhea, sweating, headache, prostration, and sometimes a fall in body temperature. Some individuals may not exhibit all of these symptoms. In more severe cases, headache, muscle cramping, and transient changes in blood pressure and pulse rate may occur. Most individuals recover on their own within 2 days, but symptoms may persist for 3 or more days in severe cases. The usual treatment for healthy persons consists of bed resting and fluid replacement to counteract accompanying dehydration. Non-hospital-acquired infections can usually be treated with penicillinase-resistant β-lactams. However, hospital-acquired infections are most often caused by antibiotic-resistant strains and can only be treated with vancomycin.

2.5.5 Toxic Dose Many factors contribute to the development of SFP and the degree of severity including susceptibility of the individual to SE, the total amount of food/toxin ingested, the type of toxin, and the overall health of the infected person with children and elderly individuals being more susceptible. SEB causes more severe symptoms than SEA as is now considered a potential bioterrorism threat if inhaled. Despite these differences, a basal level of approximately 1 ng of SE per gram of contaminated food is sufficient to induce symptoms of SFP. This toxin level is reached when S. aureus populations exceed 105 cfu/g in contaminated food.

22

R.G. Behling et al.

2.5.6 Reservoirs Staphylococci are commonly found in air, dust, sewage, water, milk, and food as well as on food contact surfaces and equipment. Humans and animals are the primary reservoirs for staphylococci with these organisms isolated from the nasal passage, throat, hair, and skin of more than 50% of healthy individuals. Higher isolation rates are common among those who associate with or who come in contact with sick individuals and hospital environments. Equipment and environmental surfaces can also serve as sources of S. aureus, although food handlers are the primary source of contamination in outbreaks of SFP. SFP is caused by ingesting enterotoxins produced in food by certain strains of S. aureus, usually because the food has not been kept sufficiently hot (≥60◦ C, 140◦ F) or cold (≤7.2◦ C, 45◦ F). Most SFP outbreaks are caused by food contamination during processing, preparation, and packaging. S. aureus is commonly found in the nose, mouth, and throat of humans and transmission to foods may occur via purulent discharges such as from an infected finger or even from normal skin since 30–50% of all healthy individuals harbor the organism and 15–35% are persistent carriers. This is one reason why individuals with open sores or infected cuts should not handle food. S. aureus competes poorly with the native microflora in most raw foods; however, products containing higher levels of salt provide S. aureus with a competitive edge. Consequently, products most frequently incriminated in SFP outbreaks are cooked or otherwise processed high-protein foods that have come in contact with worker’s hands and then were either served after being temperature abused or served after improper heating/refrigerated storage. Foods most frequently implicated in outbreaks include meat and meat products (particularly ham due to the high salt content); poultry and egg products; salads such as egg, tuna, chicken, potato, and macaroni; bakery goods such as cream-filled pastries, cream pies, and chocolate eclairs; sandwich fillings; and milk and dairy products. Foods that require considerable handling during preparation and that are kept at slightly elevated temperatures after preparation are most frequently involved in staphylococcal food poisoning. Hence, adherence to stringent hand washing and sanitation practices in food preparation areas along with proper storage temperatures is critical in minimizing contamination along with subsequent growth of S. aureus to potentially toxic levels.

2.6 B. cereus, a Toxigenic Pathogen 2.6.1 The Organism B. cereus is a gram-positive, spore-forming rod-shaped microorganism. It is closely related to Bacillus anthracis, a serious human and animal pathogen and to Bacillus thuringiensis, an insect pathogen (Helgason et al., 2000; Erickson and Kornacki, 2003). Together with B. cereus variety mycoides, these four form the B. cereus group

2

Selected Pathogens of Concern to Industrial Food Processors

23

of organisms (Bennett and Belay, 2001). The spores of B. cereus can survive most cooking processes. Mainly prolific as an aerobic vegetative cell, survival in anaerobic conditions is also possible. Some psychrotrophic strains are known (Bennett and Belay, 2001). Germination of spores to high populations of viable cells may produce a toxin in foods or in human intestines, causing gastrointestinal illness. B. cereus enterotoxin has also been associated with endophthalmitis (eye infection) (Wong, 2001).

2.6.2 Factors Related to Cost The CDC has estimated that 27,360 cases of B. cereus-induced foodborne illness occur each year in the United States (Mead et al., 1999). However, the number of reported incidents of B. cereus foodborne illness ranges from 6 to 50 incidents per year and most of these data are from cases associated with outbreaks. Active surveillance for B. cereus-induced illness is not done in the United States. The illness is self-limiting and usually not severe. Furthermore, not all state public health laboratories routinely test for B. cereus. People may thus become ill and not consult a doctor due to the relatively mild symptoms with the result being under reporting of the actual number of illnesses (MMWR, 1994). Thus, many sporadic cases go undetected. Consequently, the financial costs are difficult to estimate.

2.6.3 Symptoms B. cereus strains can produce two types of foodborne illness: diarrheal and emetic. The diarrheal illness is often associated with meat products, soups, potatoes and other starchy vegetables, puddings, and sauces. Onset times may occur 8–16 h after ingestion of food containing the microorganisms and/or toxin. Abdominal pain, diarrhea, and possibly nausea and vomiting may ensue. Illness usually lasts for 12–14 h and complications are rare. Tripartite hemolysin BL has been identified as a diarrheal toxin while cereulide is known as an emetic toxin. These two are the only specifically identified B. cereus toxins to date (Schoeni and Wong, 2005). The emetic illness may result in diarrhea and abdominal cramps but is most often characterized by nausea and vomiting. The onset of symptoms may occur only 1–5 h after ingestion and symptoms may continue for 6–24 h. Rice dishes and pasta products held at improper temperatures and allowed to cool slowly are often associated with this type of illness. Cereulide is known to survive autoclave treatments (Lui, 2001) which are similar to some retort processes. In fact, a document entitled, B. Cereus at http://www.nzfsa.govt.nz/science/data-sheets/bacillus-cereus.pdf cites a report of Jenson and Moir (1997) who indicate that the emetic toxin of B. cereus can survive 90 min at 126◦ C.

24

R.G. Behling et al.

2.6.4 Infectious Dose B. cereus can produce both heat-stable and heat-labile toxins. However, a nonbiological assay does not yet exist for the emetic toxin nor has the FDA recommended use of the commercially available kits for detection of the diarrheal toxin until further validations are done (Rhodehammel and Harmon, 2001). The number of viable cells generated by the species is the best way to estimate their toxicity. B. cereus intoxication usually requires high cell numbers (>105 cfu/g) in healthy adults. Methods for recovery of the microbes can be found in the FDA Bacteriological Manual and the Compendium of Methods for the Microbiological Examination of Foods, among others.

2.6.5 Reservoirs B. cereus is widely distributed in soil, vegetation, and a variety of foods. It is present in the intestinal flora of about 10% of healthy adults and can be found in dairy products, meats, spices, dried products, and cereals, particularly rice (Hammack et al., 1990).

2.6.6 Foods Associated with B. cereus Fried rice is a common source of illness caused by B. cereus. It is frequently present in uncooked rice, their heat-resistant spores can survive and germinate after cooking and a heat-stable enterotoxin may be produced that can survive further heating such as stir frying. B. cereus food poisoning associated with fried rice was the cause of two outbreaks at child day care centers in Virginia in 1993 (MMWR, 1994). Toxin production is enhanced by the presence of protein such as eggs or meat. Foods with high fat content may also have a protective effect. Therefore, the enterotoxin may be present in the food or it may be produced after ingestion within the small intestine.

2.6.7 Food Processing Environments B. cereus is ubiquitous in soil and raw vegetables and therefore should be expected to be present in the environments of many food production facilities. The production of stress-resistant spores may make this microorganism difficult to control in a factory environment. Kornacki isolated the organism from a food service refrigerator where it was the apparent source of a foodborne illness and from a salad production environment during the course of a food factory risk assessment (Kornacki, 2009, Personal Communication). The first recorded outbreak of foodborne disease from the consumption of raw, sprouted seeds was in 1973 and this was from soy, mustard, and cress grown in home-sprouting packs which were contaminated with

2

Selected Pathogens of Concern to Industrial Food Processors

25

B. cereus (Food Safety Australia, November, 2000). As stated earlier, most human gastrointestinal disease attributed to B. cereus intoxication has been traced to improper preparation and storage of foods. Delay in the cooling process due to voluminous containers is often a factor. It is advisable to hasten the cooling process by distribution to smaller containers and immediate refrigeration. For further information a recent review article can be referenced (Schoeni and Wong, 2005).

2.7 C. botulinum, a Toxigenic Pathogen 2.7.1 The Organism C. botulinum, a gram-positive, anaerobic, rod-shaped bacterium, consists of four physiological diverse groups (groups I–IV) that share the common feature of producing the extremely potent botulinum neurotoxins. These neurotoxins are in turn differentiated on the basis of their serological reaction and are classified as types A–G. Foodborne botulism is an intoxication involving the consumption of food containing botulinal toxin produced during the growth of these organisms in food. Groups responsible for foodborne botulism in humans include group I (all type A strains and proteolytic strains of types B and F) and group II (all type E strains and nonproteolytic strains of types B and F) whereas groups III (strains C and D) and IV (strain G) affect primarily non-human animal hosts.

2.7.2 Disease Incidence and Syndrome Foodborne botulism is a severe but rare disease. In the United States, there were 444 foodborne botulism outbreaks reported from 1950 to 1996 (CDC, 1998). Higher incidences have been reported in countries (Poland and Russia) where economic conditions have contributed to an increased reliance on home bottling/canning of foods. In the United States, it has been estimated that the cost per case of botulism is approximately $30 million compared to L. monocytogenes or Salmonella with an average cost per case of $10,000–12,000 (Setlow and Johnson, 1997). Botulinal intoxication can range from a mild illness, that may be disregarded or misdiagnosed, to a serious disease that can be fatal within 24 h. Rapidity of onset and severity of disease depend on the rate and amount of toxin absorbed with roughly half the annual cases of foodborne botulism being attributed to type A strains and types B and E responsible for the remaining cases. Lethal doses of type A botulinal neurotoxin for humans (1 μg/kg) have been estimated from primate data; however, type A is considered more lethal than types B and E (Shapiro et al., 1998). Relative ease of passage through the intestinal wall affects toxicities. Following absorption from the gastrointestinal tract, the water-soluble neurotoxic proteins (zinc-containing endopeptidases) are carried by the bloodstream and irreversibly bound to peripheral nerve endings where the release of acetylcholine is

26

R.G. Behling et al.

blocked. Signs and symptoms of botulism develop 12–72 h after consumption of the toxin-containing food and include nausea, vomiting, fatigue, dizziness, headache, dryness of skin, mouth and throat, constipation, paralysis of muscles, double vision, and difficulty breathing. Paralysis of the respiratory muscles can result in death if not treated, although the mortality rate is less than 10%. Treatment includes administration of equine antitoxin and supportive care with up to 95% of patients requiring hospitalization and 62% of patients needing mechanical ventilation. Recovery may take weeks to months (Shapiro et al., 1998).

2.7.3 Reservoirs and Prevalence in Foods C. botulinum is widely dispersed in the environment (soils, sediments, and the gastrointestinal tracts of animals) by virtue of their ability to produce resistant endospores; however, the spore load as well as the predominant type varies with the geographic region. In the United States, the spores of type A are found most commonly west of the Rocky Mountains and the neurotoxin of this type accounts for 85% of foodborne outbreaks west of the Mississippi. In the Eastern United States, type B spores are most prevalent and consequently there has been a 60% incidence in type B outbreaks east of the Mississippi River. Implicated foods in the United States are vegetables, particularly “low-acid” vegetables such as beans, peppers, carrots, and corn; however, where once outbreaks were most commonly associated with home-preserved foods, non-preserved foods and public eating places have become more often involved. In Canada and Alaska, most foodborne outbreaks have resulted from type E toxin and have been associated with native and Eskimo foods. Similarly in Russia and Japan, neurotoxin type E contaminated pickled and home-preserved fish have been the leading vehicles because the principal habitat of type E spores is freshwater and brackish marine habitats. In European countries such as Poland, France, Germany, Hungary, Portugal, the former Czechoslovakia, and Belgium, the foods most often implicated have been home-preserved meats such as ham, fermented sausages, and canned products, and the predominant type has been B. Whereas European type B strains are predominantly nonproteolytic, American type B strains have been predominantly proteolytic.

2.7.4 Physiological Characteristics of C. botulinum Due to the presence of exogenous proteases, protein breakdown products, in addition to sugars, may be used for growth by proteolytic C. botulinum. Under these conditions, off-odors are produced concurrently with protein breakdown that fortunately serve to alert the consumer that the product is spoiled. In contrast, spoilage odors may not be present in foods where growth of non-proteolytic strains of C. botulinum has occurred (Lynt et al., 1975).

2

Selected Pathogens of Concern to Industrial Food Processors

27

In addition to growth substrates, proteolytic and nonproteolytic strains of C. botulinum also differ in their minimum and optimal growth temperatures. In the case of proteolytic strains, the optimal temperature for growth is 37◦ C, with growth occurring between 10 and 48◦ C. Nonproteolytic strains, however, have a lower optimal growth temperature (30◦ C) but more importantly may grow at temperatures as low as 3◦ C (Graham et al., 1997). Consequently, there has been considerable concern raised that nonproteolytic organisms may grow and produce toxin in refrigerated foods that receive minimal processing and have extended shelf lives. Tolerance to pH, salt, and water activity by the vegetative cells also differ between the proteolytic and nonproteolytic strains. Conditions that favor growth of proteolytic strains include low acid (pH above 4.6), low salt (below 10%), and relatively high moisture (aw above 0.94). Growth of nonproteolytic strains, on the other hand, requires higher pH (above 5.0) and moisture (aw above 0.97), but lower salt (above 5%) contents (Kim and Foegeding, 1993). In addition to the differences in tolerance of vegetative cells to environmental conditions, differences in spore resistance exist between the C. botulinum groups. D100◦ C values for spores from proteolytic and nonproteolytic strains are approximately 25 min and less than 0.1 min, respectively. Consequently, the high heat resistance by spores of proteolytic strains represents a major concern in the processing of low-acid canned foods. Botulinum toxin synthesis and activation is a complex process that is highly regulated by nutritional and environmental conditions. In general, toxin production onsets during late log and early stationary phases; however, toxin production varies among strains (Bradshaw et al., 2004).

2.7.5 Detection of C. botulinum Neurotoxins To date, the mouse bioassay remains the only validated method for food analysis of botulinal neurotoxins. In this assay, the toxin (minimum detection limit, 0.03 ng) is detected by injection of a food extract into mice, which are then observed for characteristic symptoms of botulism and ultimately death over a 48-h period. In vitro assays, such as the standard enzyme-linked immunosorbent assay, often lack specificity and exhibit poor sensitivity compared to the mouse bioassay (Sharma and Whiting, 2005).

2.7.6 Control Treatments Control of C. botulinum in foods may be exerted at three levels: inactivation of C. botulinum spores or inhibition of germination; inhibition of C. botulinum growth and toxin formation; and inactivation of C. botulinum neurotoxin. Thermal processing is the most common method used to inactivate spores of C. botulinum with the most heat-resistant spores (group I) having D121◦ C values of 0.1–0.2 min and

28

R.G. Behling et al.

therefore serving as the target organism. It is important to be aware that thermal destruction of C. botulinum spores does not follow first-order kinetics, indicating that some spores are more heat resistant than others (Peleg and Cole, 2000). Moreover, heat resistance of spores is greater at higher pH values (Mafart et al., 2001) and fat contents (Molin and Snygg, 1967) and lower sodium chloride concentrations (Juneja and Eblen, 1995) and water activities (Murrell and Scott, 1966). Spores are of particular concern in the commercial stabilization of canned low-acid foods, hence the canning industry has adopted a 12-D process as the minimum thermoprocess (Hauschild, 1989). For the less heat-resistant spores of group II strains, moderate temperatures (40–50◦ C) may be combined with high pressures of up to 827 mPa (Reddy et al., 1999). Lower heat treatments (pasteurization) in combination with other control measures, such as refrigeration, are used for perishable vacuum-packaged foods. Avoidance of conditions or compositions that increase spore germination may also be considered in control of C. botulinum in foods. For example, germination is similar in aerobic and anaerobic conditions and will occur from 1 to 40◦ C but not at 50◦ C (Plowman and Peck, 2002). Furthermore, formulating products with ingredients that lack L-alanine and L-lactate (essential germinants) or sodium bicarbonate and sodium thioglycollate (accelerants of germination) is desirable (Plowman and Peck, 2002). Such a case was demonstrated in potato and broccoli purees whose lag times for growth of C. botulinum were much longer than seen with mushroom purees (Braconnier et al., 2003). Targeting the primary factors that control growth of C. botulinum, temperature, pH, water activity, redox potential, and oxygen level have been used for control of this pathogen in foods. The critical level of oxygen that will permit growth of proteolytic C. botulinum is 1–2% but this level depends on other conditions such as aw , pH, or redox potential (Kim and Foegeding, 1993; Johnson, 1999). When control by these parameters is inadequate, inhibitory substances (nitrites, sorbates, parabens, nisin, phenolic antioxidants, polyphosphates, ascorbates, EDTA, metabisulfite, n-monoalkyl maleates and fumarates, and lactate salts) may be added to the food system (Kim and Foegeding, 1993). Efficacy of these antimicrobials added to or found naturally in foods, however, may be reduced by fat (Glass and Johnson, 2004). Growth of competitive and growth-promoting microorganisms in foods, on the other hand, has a very significant effect on the fate of C. botulinum. While acid-tolerant molds can provide an environment that enhances the growth of C. botulinum, lactic acid bacteria can inhibit growth of C. botulinum largely not only by reducing the pH but also by the production of bacteriocins (Rodgers et al., 2003). Ideally, an appropriate degree of protection against growth and toxin production by C. botulinum would employ multiple barriers. Irregardless, challenging foods with spores of C. botulinum to determine whether toxin is produced in optimal conditions or during temperature abuse is desirable to evaluate the safety of that food. Although only minute amounts of botulinal neurotoxin need be present to observe a toxic response, protective measures for inactivation of the agents could ensure that they do not reach those toxic doses. In addition to high sensitivity to temperatures above 55◦ C, botulinal neurotoxins are sensitive to protease activity and oxidizing agents (i.e., chlorine) (Siegel, 1993).

2

Selected Pathogens of Concern to Industrial Food Processors

29

2.8 Enterohemorrhagic E. coli, a Toxico-infectious Pathogen E. coli O157:H7 and other enterohemorrhagic E. coli produce a toxin(s) after it implants in the colon and colonizes it resulting in illness. Pre-formed toxins have not been shown to occur in foods or cause human disease. Hence this organism is considered to be “toxico-infectious” agent in this chapter, as opposed to an invasive pathogen (such as Salmonella spp.). However, some evidence for an invasive mechanism has been reported (Doyle et al., 1997). It is a difficult organism to manage from a public health standpoint, because of its low infectious dose which may be, in part, related to its substantial acid tolerance and ability to survive low pHs sometimes found in the stomach. The CDC estimates that E. coli O157:H7 sickens over 73,000 individuals per year. Despite the relatively low number of illnesses compared to other foodborne microbes, there are an estimated 2,100 hospitalizations and 61 deaths (CDC), putting it fourth in the number of annual foodborne illness related deaths below non-typhoidal Salmonella (553 deaths), L. monocytogenes (499 deaths), and Campylobacter (99 deaths) (CDC; Mead et al, 1999).

2.8.1 The Organism E. coli is a gram-negative, non-spore-forming short rod-shaped bacterium capable of growth and gas production at 45.5◦ C (except when testing water, shellfish, and shellfish harvest water, which use 44.5◦ C) in lactose-containing medium which also exhibit the characteristic biotype 1 (+,+,–,–) or biotype II (–,+,–,–) standing for positive “+” or negative “–” reactions on Indole, Methyl Red, Voges-Proskauer, and Citrate (IMViC) reactions, respectively (Kornacki and Johnson, 2001). Most E. coli strains are harmless inhabitants of the gastrointestinal tract of man and animals. However, several foodborne pathogenic strains of E. coli are known to exist (Kornacki and Marth, 1982, Doyle et al., 1997b). In 1982 a particularly severe strain, with the “157” O-antigen and the “7” H-antigen, was isolated from clinical samples of individuals with gastrointestinal illness associated with the consumption of undercooked hamburgers from two fast food restaurants in Oregon and Michigan in which over 700 persons in four states were infected including 51 cases of HUS and four deaths (Besser et al., 1999; Feng, 1995). There was also an outbreak associated with consumption of raw milk that year (Mortimore and Wallace, 1998). This organism is distinguished from other E. coli strains by their inability to ferment sorbitol and their lack of production of β-glucuronidase (Besser et al., 1999). Early evidence with one outbreak-associated strain suggested that they may not be able to grow at 45.5◦ C (Doyle and Schoeni, 1994). Palumbo et al. (1995) reported that 18 of 23 strains were capable of growth at 45◦ C in EC broth in a circulating water bath. It is not clear whether those strains of E. coli O157:H7 that are capable of growth at 45◦ C would be capable of growth in typical coliform waterbath held at 45.5 ± 0.2◦ C, however. Nevertheless, another strain assayed with a temperature gradient

30

R.G. Behling et al.

incubator was not recoverable at such temperatures and the authors concluded that “The temperature range for growth of E. coli 0157:H7 is inconsistent with that of other fecal coliforms, suggesting that this pathogen is excluded with standard enumeration procedures used for foods and water.” (Raghubeer and Matches, 1990). These data indicate that the 45.5◦ C incubation temperature used in the identification of generic E. coli cannot be expected to be a reliable means to isolate E. coli O157:H7. Furthermore Tuttle et al. (1999) found that the MPN of E. coli (performed at 45.5◦ C) did not show any correlation with the presence of E. coli O157:H7. The authors speculated that this may be a result of the heterogeneity of distribution of E. coli O157:H7 in ground beef.

2.8.2 Cost Buzby and Roberts (1996) estimated that illness due to E. coli O157:H7 costs $300–700 million annually in the United States. However, the non-calculable cost associated with human suffering can be staggering given the seriousness of some forms of the disease its potential sequelae (e.g., hemolytic uremic syndrome – HUS).

2.8.3 Disease Syndromes Some individuals can be infected but remain asymptomatic (Doyle et al., 1997). However, human illness from E. coli O157:H7 can result in nonbloody diarrhea and hemorrhagic colitis. In 3–5% of cases hemolytic uremic syndrome (HUS) may result. 2.8.3.1 Onset Time Onset times of 3–4 days have been reported (Besser et al., 1999). The ICMSF indicated a range of 3–9 days (ICMSF, 2002b). However, the CDC (2004) in their document entitled, “Diagnosis and management of foodborne illnesses: A primer for physicians. Second Revision” has indicated that a 1- to 8-day incubation period is possible. This document was produced collaboratively by the CDC and the American Medical Association, the American Nurses Association, the FDA’s Center for Food Safety and Applied Nutrition, and the US Department of Agriculture and also appears in another publication authored by all the above (http://www.amaassn.org/ama1/pub/upload/mm/36/2004_food_introclin.pdf). Elsewhere CDC (2000) stated a 1- to 10-day incubation period may occur. The average onset time is 4 days (ICMSF, 2002b). This time frame appears to be consistent with gastrointestinal transport times of solid food and data from animal studies. It is difficult to imagine a shorter onset time than one entire 24 h day with a contaminated solid food because some hours, perhaps 10–12 h in some instances, can be expected

2

Selected Pathogens of Concern to Industrial Food Processors

31

for the microbe to traverse the relatively harsh environments of the stomach (perhaps 4–6 h for large particles), and small intestine (perhaps another 6 h), where growth would not be expected to occur, before it reaches the colon where attachment, proliferation, colonization, toxin production, and effacement of colonic microvilli can begin. 2.8.3.2 Attachment and Colonization The ability of the organisms to adhere to intestinal epithelial cells and colonize the gut is “undoubtedly one of the key determinants of virulence” (Patton and Patton, 1998). Ritchie et al. (2003) showed that infant rabbits were useful models for investigation of the intestinal stage of enterohemorrhagic E. coli pathogenesis. Data from this model indicated that diarrhea and inflammation in the colon were dependent on colonization (Ritchie et al., 2003). Interestingly they found that colonization without persistent diarrhea resulted when a Stx2 non-producing isogenic mutant strain was used. Infant rabbits developed severe diarrhea 2–3 days post-intragastric inoculation with a high inoculum (5 × 108 cfu) of this organism (per 90 g body weight). They also developed diarrhea when given 100 μg of Stx2 /kg on days zero and one, but not in controls given heat-inactivated Stx2 . These data suggest that attachment, colonization, and toxin production are required for disease symptoms. E. coli O157:H7 levels associated with intestinal colonization in animal models appear to be very high in photomicrographs where colonization of the entire crypt length of the cecum and colon of gnotobiotic pigs was noted at the fourth day (Francis et al., 1986). Ritchie et al. (2003) recovered 1 × 108 /g E. coli O157:H7/g of colonized intestine. It should be noted that growth of E. coli O157:H7 from about 10 cells/ml (1–10 cells/g) is considered to be a high level in ground beef product (ICMSF, 2002b) to the late log/early stationary phase occurring in 17.5 h under ideal aerated conditions in nutritious medium (Brain–Heart Infusion) at 37◦ C (Palumbo et al., 1995). This is also consistent with what one would expect for the time to produce one visible colony from a single cell on nutritious solid media under optimum laboratory conditions. Furthermore, McIngvale et al. (2002) found that the highest amount of toxin mRNA was detected from late log-phase and early stationary-phase cells corresponding to 108 –109 cfu/ml. Thus it seems logical that colonization of the human colon should therefore take longer than 17 h once the microbe has implanted on colonic epithelial cells after passage from mouth to the colon, given a contamination level of 10 cells or less per gram of product. If the results from the infant rabbit model can be extrapolated to humans then inflammation would not be expected to occur until after 17.5 h in humans after the organism embedded in large particles of a solid food has traversed the mouth, stomach, and small intestine to the colon; a process that may take an additional 10–12 h thus resulting in hypothetical minimal onset time of 27.5–29.5 h in this example. Furthermore genes involved in the attaching effacing lesion were shown to be regulated by a quorum (cell population)-based sensing mechanism (Sperandio et al., 1996). Data derived from a study of E. coli O157:H7 growth in BHI by Palumbo

32

R.G. Behling et al.

et al. (1995) indicated that toxin production did not occur at a detectable level until about the mid-log phase of growth (about 8 h) when incubated at 37◦ C under ideal aerated conditions in the laboratory. The amount of toxin produced correlated to the population of E. coli O157:H7, with the maximum amount of toxin being detected when the population reached the stationary growth phase (about 17.5 h in the example given). E. coli O157:H7 growth (and therefore colonization) in the suboptimal (e.g., anaerobic, variable nutrition, presence of competitors) environment of the human colon is apt to be much slower than under optimal controlled conditions in a laboratory. Not surprisingly, Tamplin (2002) demonstrated that the rate of inoculated E. coli O157:H7 growth (in beef) decreased as the ratio of background flora to E. coli increased. It seems possible, through unlikely, that shorter onset times shorter than 24 h may occur with liquid foods or small particles that could be expected to pass through the gastrointestinal intestinal tract more quickly. However, this author has not seen examples of shorter onset times in the literature. Onset times shorter than 24 h should be rare indeed. The reported onset time between the onset of HUS or TTP-like illness is 4–15 days (Griffin et al., 1988). HUS and TTP-like illness (the role of E. coli in TTP has been questioned in recent times-editor) resulting from E. coli O157:H7 need not be preceded by diarrhea, although that would be more common (Griffin et al., 1988).

2.8.3.3 Hemorrhagic Colitis Symptoms of hemorrhagic colitis are characterized by a prodromal phase which includes crampy abdominal pain, followed 1–2 days later by nonbloody diarrhea which progresses within 1–2 days to bloody diarrhea lasting 4–9 days (Doyle et al., 1997; Griffin et al., 1988).

2.8.3.4 HUS The prodrome, described briefly above, is bloody diarrhea. This can then be followed by acute nephropathy, seizures, coma, and death. The disease is characterized by hemolytic anemia, thrombocytopenia, and renal failure. Among those who have been colonized with E. coli O157:H7 approximately 3–5% develop HUS (Tuttle et al., 1999; ICMSF, 2002b). A mortality rate of 4% can still be expected in HUS patients, even in the presence of meticulous care (Besser et al., 1999).

2.8.3.5 TPP Thrombotic Thrombocytopenic Purpura-like illness is similar to HUS but includes fever and central nervous system disorder (ICMSF, 2002b). TTP tends to be diagnosed in adults and some feel that post-diarrheal TTP is likely to be the same disorder as HUS (Besser et al., 1999).

2

Selected Pathogens of Concern to Industrial Food Processors

33

2.8.4 Pathogenic Mechanisms The pathogenic mechanisms of E. coli O157:H7 are not fully elucidated. The virulence mechanism associated with hemorrhagic colitis is a combination of attaching and effacing adherence to the colon. These strains produce one or both of two toxins (Stx1 and Stx2 ) related to the toxins produced by Shigella dysenteriae, known as Shiga toxins which act by inhibiting protein synthesis. Stx2 -producing E. coli O157:H7 are more common and they are often associated with severe cases of bloody diarrhea than those which produce Stx1 or a combination of Stx1 and Stx2 (Ritchie et al., 2003). It has been proposed that intestinal fluid secretion and therefore diarrhea results from the selective killing of the villus tips colonic epithelial cells by Stx (Eslava et al., 2003). Damage to the underlying tissue and vasculature, perhaps by exotoxin- and endotoxin-related mechanisms, results in bloody diarrhea (Doyle et al., 1997). The Stx’s then enter into the blood stream where they may damage kidney glomeruli (a characteristic of HUS) and cause other problems (Doyle et al., 1997). A set of genes called the locus of enterocyte effacement (LEE) are considered a key pathogenicity “island” for E. coli O157:H7. These genes include, among others, the eae gene which encodes for intimin and a gene called tir (stands for translocated intimin receptor) which plays a critical role in the attaching/effacing lesions (Ritchie et al., 2003). The Tir protein is translocated into the enterocyte which provides a receptor for intimin which induces the production of actin filaments in the enterocyte that produce a cup-like attachment structure on the enterocyte for E. coli O157:H7 cells. The active portions of the Stx toxins enter the cell, are transported through the endoplasmic reticulum, and inhibit the function of the 28S rRNA thus inhibiting protein synthesis (Paton and Paton, 1998).

2.8.5 Infectious Dose Ground beef patties with less than 700 organisms per uncooked patty have been associated with illness (CDC, 2001; Tuttle et al., 1999). In one study levels of 0.3–15 cells/g of ground beef were found in positive lots implicated in an outbreak (Doyle et al., 1997). Mead and Griffin (1998) reported doses as low as 50 cells can be infectious and in one reference the FDA has indicated that as few as 10 cells may be adequate to cause illness (FDA, 2001; http://www.cfsan. fda.gov/∼mow/chap15.html). Consequently the meat industry has taken strenuous actions to control this in the beef supply and in the processing environment.

2.8.6 Reservoirs The main reservoir for this organism appears to be the gastrointestinal tract of cattle. Approximately 1% of healthy cattle have the organisms. However, it has been found

34

R.G. Behling et al.

in other ruminants and in dogs, horses, and birds as well (Besser et al., 1999; Meng et al., 2001).

2.8.7 Foods Associated with E. coli O157:H7 More disease outbreaks have been caused in the United States by undercooked ground beef than by any other food vector. However, ground beef is not the only source. Between 1982 and 1994 outbreaks were reported from consumption of contaminated ground beef (32.4%), vegetables and salad bars (5.9%), roast beef (2.9%), raw milk (2.9%), and apple cider (2.9%) (Doyle et al., 1997). In addition FDA reports that outbreaks have also occurred from alfalfa sprouts, unpasteurized fruit juices, lettuce, game meat, and cheese curds (FDA, 2001; http://www.cfsan.fda.gov/∼mow/chap15.html). Raw milk was the vehicle in a school outbreak in Canada. Salami, sandwiches, ranch dressing have also been implicated (Besser et al., 1999). Furthermore radish sprouts were implicated in several Japanese outbreaks including one in Sakkai City where 6,000 school children were affected (Besser et al., 1999) and raw spinach has also been implicated in the United States (Grant et al., 2008).

2.8.8 Unique Acid Tolerance in Foods E. coli O157:H7 survived with only a 2 log10 reduction in 2 months in fermented sausage (pH 4.5) held at 4◦ C (Doyle et al., 1997). It also survived in mayonnaise (pH 3.6–3.9) for 5–7 weeks at 5◦ C and 1–3 weeks at 20◦ C when inoculated at high levels (Doyle et al., 1997). The organism was also shown to survive in apple cider (pH 3.6–4.0) for 10–31 days at 8◦ C and for 2–3 days at 25◦ C (Doyle et al., 1997).

2.8.9 Other Sources of Infection Contaminated food and water remain the most common source of transmission of E. coli O157:H7. Unchlorinated water and swimming water have both been shown to be sources of outbreaks (Besser et al., 1999). However, between 1982 and 1994 person-to-person outbreaks accounted for 13.2% of outbreaks (Doyle et al., 1997).

2.8.10 Food Processing Environments Despite the collection of hundreds of environmental swabs, there has been great difficulty finding growth niches for this organism in the meat industry (Pruett, 2004, Personal Communication; Freier, 2004, Personal Communication). This may be a consequence of the lack of psychrotrophic (e.g., capable of growth at 7◦ C) adapted

2

Selected Pathogens of Concern to Industrial Food Processors

35

strains in the refrigerated environments of meat processing factories which strive to maintain temperatures of 40◦ F (4.4◦ C) in many areas or perhaps competation with other strains of E. coli in these environments. Many strains were shown to be capable of growth at 10◦ C and a few were found able to grow at 8◦ C (Palumbo et al., 1995) which suggests the importance of refrigeration of the product and processing environments. Contamination of beef surfaces may occur during slaughter and processing and the organism may be transferred from the surface to the interior of the product during grinding (Tuttle et al., 1999). Some actions taken by the industry to reduce the prevalence of E. coli O157:H7 on carcasses include refrigeration, cleaning and sanitization in the factory, organic acid carcass rinses, carcass steam chamber treatments, and specialized automated steam trimmers which promote destruction of microbes on cut surfaces, among others.

2.9 C. perfringens, a Toxico-Infectious Agent C. perfringens, as well as C. botulinum and B. cereus, are gram-positive and anaerobic spore-forming bacteria known to cause food poisoning. According to the Center for Disease Control and Prevention (CDC), it ranks as the third most foodborne bacterial common disease in the United States. Sometimes it is called the “food service germ” because foods served and left for long periods at room temperature have been associated with this illness. Strains of this bacterium produce a protein toxin, named C. perfringens enterotoxin (CPE), which is considered as the virulence factor of this food poisoning organism in the context of food poisoning. Almost all C. perfringens-mediated foodborne illness in the United States and other developed countries involves the “Type A” toxin. C. perfringens is widely distributed in the environment and frequently occurs in the intestines of humans and many domestic animals. Its spores are able to survive normal cooking and pasteurization temperatures, after which they can then germinate and multiply during slow cooling, or storage at room temperatures and/or during inadequate re-warming (Jay, 2000).

2.9.1 The Organism C. perfringens is a gram-positive, anaerobic, spore-forming rod-shaped bacterium. It is considered an anaerobic, because it does not grow on agar plates continuously exposed to air. However, unlike most other anaerobes, such as C. botulinum, this bacterium tolerates moderate exposure to air. C. perfringens is a mesophilic bacterium. The lowest temperature for growth is around 20◦ C and the highest is around 50◦ C. The optimum growth temperatures are between 37 and 45◦ C. The organism’s generation time at 45◦ C under optimal conditions can be as rapid as 7 min allowing C. perfringens to quickly multiply in foods where it may form discrete microscopic colonies of high population. Hence

36

R.G. Behling et al.

quantitative results from sampling of such foods may differ widely from sample to sample. Many strains grow over the pH range of 5.5–8.0, but not below 5.0 or above 8.5. The required water activity (aw ) for growth and germination of spores lies between 0.97 and 0.95 with sucrose or NaCl, but can be as low as 0.93 with glycerol. Sporulation requires higher aw than for growth. Growth is inhibited by about 5% NaCl (McClane, 2001). The organism’s ability to form heat-resistant spores also contributes to its ability to cause food poisoning. Spores of some C. perfringens strains can survive boiling under some conditions. However, heat resistance differs among C. perfringens strains. For example, a D100 ◦ C value of 17.6 min for strain NCTC 8238 and 0.31 min for strain ATCC 362 have been reported. It appears that other stress factors, such as high pH, may cross-induce heat resistance of some C. perfringens. In addition, the literature reports some rare strains of C. perfringens isolated from canned food which are also more heat resistant than C. botulinum (Bradshaw et al., 1977; Adams, 1973) a fact that may have some implications for canning of foods that would permit the growth of these organisms.

2.9.2 C. perfringens Enterotoxin The virulence factor of C. perfringens food poisoning is an enterotoxin that induces fluid secretion and electrolyte losses from the GI tract of human and animals. It is a sporulation-specific protein toxin. The CPE is synthesized during the late stage of sporulation. The toxin production peak occurs just before lysis of cell’s sporangium, and the CPE is released along with spores (McClane, 2000). C. perfringens is classified as five types (A through E) based on four toxins (alpha, beta, epsilon, and iota) produced. Almost all C. perfringens foodborne illness in the United States and other western countries are attributed to type A food poisoning. Necrotic enteritis (known as Darmbrand or Pig-Bel), caused by type C poisoning, is rare in industrial countries. The overwhelming association between type A isolates of C. perfringens and type A food poisoning may be due to its wide distribution in the environment. Unlike the organism, the enterotoxin CPE is not heat stable. Its biological activity can be destroyed by heating for 5 min at 60◦ C. The toxin is also very sensitive to pH extremes.

2.9.3 Cost CDC has indicated that C. perfringens type A food poisoning ranks as the third most commonly reported bacterial cause of foodborne disease in the United States. Outbreaks of type A intoxication that occurred between 1992 and 1997 involved 248,520 cases, as estimated by USDA. The annual costs of illness from C. perfringens infection were estimated at $100 million.

2

Selected Pathogens of Concern to Industrial Food Processors

37

2.9.4 Disease Syndromes Symptoms of C. perfringens type A food poisoning appear between 6 and 24 h (usually 8–12 h) after eating contaminated food and then resolve spontaneously within the next 12–24 h. The symptoms consist of the sudden onset of acute abdominal pain followed by diarrhea. Nausea is common, but fever and vomiting are usually absent. Death rates from C. perfringens type A food poisoning are low, however, fatalities do occur in elderly or in debilitated persons. The illness occurs when people swallow these bacteria or their spores which then multiply and produce toxin in the small intestine (hence its classification as a toxico-infection in this chapter). The diagnosis is confirmed by a laboratory test on a fecal specimen, with an outbreak being confirmed by tests on suspect foods.

2.9.5 Infectious Dose C. perfringens type A food poisoning starts when bacteria are ingested with contaminated food. Most of the bacteria (vegetative cells) are killed by gastric acid in the stomach. However, if the food has sufficiently high numbers (>106 –107 cfu vegetative cells/g food), some of these bacteria may pass through into small intestine. Once present in small intestine, vegetative cells will multiply and later undergo sporulation during which CPE responsible for C. perfringens type A food poisoning is produced.

2.9.6 Reservoirs C. perfringens is ubiquitous and found in soil (at levels of 103 –104 cfu/g), decaying vegetation, dust, foods (>50% of raw or frozen meat contains some C. perfringens), the intestinal tract of human and other vertebrates (e.g., human feces usually contain 104 –106 cfu/g). They are also commonly recovered from infected sites but usually as a component of a polymicrobial flora, which makes their role in pathogenesis difficult to establish. Meats, meat products, and gravy are the most frequently implicated with food poisoning caused by C. perfringens. The heating procedures of such foods may be inadequate to destroy the heat-resistant endospores, and when the foods are cooled and rewarmed, the spores germinate and grow. Consequently, the USDA has published guidelines with the goal of ensuring that cooling of such products not permit more than 1 log10 cfu growth of C. perfringens populations (USDA, 1999).

38

R.G. Behling et al.

2.10 Arcobacter, an Emerging Pathogen 2.10.1 Characteristics of the Organism Arcobacter is a member of the Epsilobacteria group which also includes Campylobacter and Helicobacter spp. The genus Arcobacter contains four different species. Arcobacter nitrofigilis has been isolated from plant roots, and Arcobacter cryaerophilus, Arcobacter skirrowii, and Arcobacter butzleri have been isolated from animals. A. butzleri, A. cryaerophilus, and A. skirrowii have been reported to cause human and animal illnesses, whereas A. butzleri has been isolated most frequently from cases of human enteritis (Kiehlbauch et al., 1991; Lerner et al., 1994). A. butzleri is a gram-negative, curved (vibrio-like), non-spore-forming rod. It is approximately 0.2–0.9 μm wide and 1–3 μm long. It is motile due to a single polar unsheathed flagellum. It is capable of growth at a range of temperatures from 15 to 35◦ C, with its optimum range from 25 to 30◦ C. A. butzleri does not grow at temperatures higher than 35◦ C, unlike C. jejuni subspecies jejuni, which requires a higher temperature ranging from 37 to 42◦ C. Arcobacter and Campylobacter species have been found at similar locations on broiler carcasses and show a close genetic relationship (Vandamme et al., 1991). Additionally, Arcobacter has an ability to grow under aerobic conditions in which Campylobacter is inhibited from growth, but capable of survival. A. butzleri is microaerophilic upon initial isolation, but can be grown aerobically after sub-culturing. The organism does not ferment sugars, but instead uses pyruvate as an energy source. It has few biochemical properties that may aid in its identification; however, a multiplex PCR has proven useful for the detection of all Arcobacter species (Wesley and Baetz, 1999).

2.10.2 Nature of the Disease Marinescu et al. (1996) indicated that the biotypes and serotypes of A. butzleri isolates from poultry products were similar with those isolated from humans with diarrheal illness. This was the first report that implied a link between human illness and contaminated food products. In humans, A. butzleri and A. cryaerophilus have been isolated from stool samples of patients with acute diarrhea. However, the significance of Arcobacter spp. as a cause for human diarrhea is still unknown. This is probably due to the fact that clinical samples are not routinely tested for Arcobacter spp. as is commonly done for Campylobacter spp. or Salmonella spp. (Lehner et al., 2005). Patients infected with A. butzleri can be asymptomatic, but the most common symptom is acute watery diarrhea lasting for 3–15 days, often accompanied by abdominal pain and nausea. Limited information regarding the pathogenicity and epidemiology of A. butzleri is available.

2

Selected Pathogens of Concern to Industrial Food Processors

39

2.10.3 Food and Environmental Sources Previous studies indicate that Arcobacter spp. are present on many retail poultry carcasses and other meat products (Atabay and Corry, 1997; Collins et al., 1996; Festy et al., 1993; Lammerding, 1996; Marinescu et al., 1996; Schoeder-Tucker et al., 1996; Wesley, 1996). Arcobacter spp. have been isolated more frequently from poultry than from red meat (Wesley, 1996; Corry and Atabay, 2001) suggesting that poultry may be a significant reservoir. Most reports of Arcobacter in poultry meat have identified A. butzleri, but A. cryaerophilus and A. skirrowii have also been reported (Atabay et al., 1998). Lehner et al. (2005) summarized recent studies on the prevalence of Arcobacter isolated from retail raw meat products. Currently, there are no standard isolation protocols or methods for Arcobacter spp., therefore the true occurrence of this pathogen in food is largely unknown. Water may play an important role in the transmission of these organisms and drinking water has been cited as a major risk factor in acquiring diarrheal illness associated with Arcobacter spp. (Lehner et al., 2005). A. butzleri is sensitive to chlorine, indicating that disinfection practices normally used in drinking water treatment would be adequate for the control of arcobacters. While control of Arcobacter on the farm may reduce contamination at the processing and retail levels, the habitat of Arcobacter species in living birds is still unknown (Houf et al., 2002). Gude et al. (2005) reported on a study of sources of Arcobacter spp. in chicken rearing and processing. They concluded that Arcobacter species were not present in samples examined from live birds or their immediate environment, but A. butzleri was widely distributed throughout the abattoir environment and on poultry carcasses, usually in low numbers. In another study, with chickens, Eifert et al. (2003) reported that A. butzleri was capable of surviving in litter with or without birds and could therefore be problematic in operations where “built-up” litter is used. If A. butzleri is an environmental pathogen, then the implementation of on-farm Best Management Practices might play a substantial role in reducing its prevalence in commercial poultry. The role of Arcobacter in human disease is unclear. Human exposure to this potential pathogen can occur from several sources in addition to raw meat products. Efforts to reduce or eliminate Arcobacter from the human food chain should be encouraged. Further studies on the ecology and epidemiology of Arcobacter spp. are necessary (Lehner et al., 2005).

2.11 Cronobacter (Enterobacter sakazakii), an Emerging Pathogen 2.11.1 Introduction, Background, and Bacterial Characteristics E. sakazakii is a gram-negative, non-sporulating, rod-shaped, opportunistic bacterium that has been historically distinguished from E. cloacae based on its

40

R.G. Behling et al.

ability to produce yellow pigment, and, as of 1980, comprised a distinct species within the genus Enterobacter (in the Family Enterobacteriaceae) family Enterobacteriaceae based on DNA–DNA hybridization and phenotypic characteristics (Farmer et al., 1980). The pathogen has been reported as growing between the temperatures of 4 and 47◦ C (Nazarowec-White and Farber, 1997b; Farmer et al., 1980) and has been implicated in sporadic cases of neonatal sepsis and meningitis, associated with necrotizing enterocolitis in infants and, in rare instances, has been implicated in infections in immunocompromised adults (Gurtler et al., 2005). Over 80 cases of E. sakazakii-related illness have been reported (Iversen and Forsythe, 2003; FAO/WHO, 2006). Non-pigmented isolates of E. sakazakii have been reported, and pigmented isolates have been known to lose pigment production following multiple transfers (Farmer et al., 1980). E. sakazakii produces the enzyme α-glucosidase, which has been utilized in several differential media for the presumptive detection of the pathogen (Iversen et al., 2004b; Oh and Kang, 2004; Leuschner et al., 2004; Restaino et al., 2006), although falsepositives have been documented. The US Food and Drug Administration method for isolating and enumerating E. sakazakii from dehydrated powdered infant formula involves rehydration with water and pre-enrichment at 36◦ C overnight, enrichment in Enterobacteriaceae enrichment broth and incubating at 36◦ C overnight, streaking onto Violet Red Bile Glucose agar and incubating at 36◦ C overnight, picking five presumptive positive colonies that are streaked onto Tryptic Soy agar and incubating at 36◦ C overnight, and finally yellow pigmented colonies are confirmed by means of the API 20E biochemical identification kit (USFDA, 2002a, 2002b).

2.11.2 E. sakazakii Reservoirs and Presence in Food and the Environment E. sakazakii is not known to have a primary reservoir and appears to be extremely widespread in nature. The pathogen has been isolated from a stirring spoon and a dish brush used to prepare infant formula (Muytjens et al., 1983), tires of a forklift in an infant formula production factory, a leaky water pipe, sugars, and gums (Olson, 2006), water (Cruz et al., 2004; Mosso et al., 1994), dust (Cruz et al., 2004), an unopened non-fat dried milk (Farmer et al., 1980), dried infant foods, milk powders, cheese products, herbs, and spices (Iversen et al., 2004a), infant weaning foods (Jung and Park, 2006), vacuum cleaner bags in homes, factories that process powdered milk, chocolate, cereal, potato flour, and pasta (Kandhai et al., 2004a), fermented bread (Gassem, 1999), a fermented beverage (Gassem, 2002), lettuce (Soriano et al., 2001); mung bean sprouts (Robertson et al., 2002), alfalfa sprouts (Cruz et al., 2004), rice starch, rice flour, and eggs (Kornacki, 1998), rice (Cottyn et al., 2001), beer mugs (Schindler and Metz, 1990), sour tea (Tamura et al., 1995), cheese, minced beef, sausage meat, and vegetables (Leclercq et al., 2002), ground meat (Nazarowec-White and Farber, 1997a), the Mexican fruit fly Anastrepha ludens (Kuzina et al., 2001), the stable fly larvae Stomoxys calcitrans (Hamilton et al., 2003), a physician’s stethoscope and an uninoculated bottle of

2

Selected Pathogens of Concern to Industrial Food Processors

41

bacterial culture medium (Farmer et al., 1980), grass silage (Van Os et al., 1996), hospital air (Masaki et al., 2001), clinical materials (Janicka et al., 1999; Tuncer and Ozsan, 1988), rats (Gakuya et al., 2001), soil (Neelam et al., 1987), rhizosphere (Emilani et al., 2001), sediment and wetlands (Espeland and Wetzel, 2001), crude oil (Assadi and Mathur, 1991), and cutting fluids (Sulliman et al., 1988). E. sakazakii has also demonstrated the ability to attach to bottles and enteral feeding tubes, which are used to feed infants in neonatal intensive care wards (Zogaj et al., 2003). Reports have detailed the unusually high desiccation resistance of E. sakazakii (Breeuwer et al., 2003; 2004; Edelson-Mammel et al., 2005; Caubilla-Barron and Forsythe, 2006; Gurtler and Beuchat, 2007) which may contribute, in part, to its reported presence in powdered infant formulas (Biering et al., 1989; Block et al., 2002; Clark et al., 1990; Himelright et al., 2002; Iversen et al., 2004a; Muytjens et al., 1983; Muytjens et al., 1988; Simmons et al., 1989; Smeets et al., 1998; Van Acker et al., 2001) and powdered infant formula manufacturing environs (Olson, 2006).

2.11.3 Pathogenicity and Infectious Dose The FAO/WHO has categorized E. sakazakii and Salmonella as the only two “category-A” pathogens in powdered infant formula based on their contamination risk and pathogenicity (2006). Historically, neonates and young infants have shown a greater propensity for contracting sporadic E. sakazakii-associated illnesses, most likely due to their immunocompromised state (Urmenyi and Franklin, 1961; Jöker et al., 1965; Monroe and Tift, 1979; Kleiman et al., 1981; Muytjens et al., 1983; Muytjens, 1985; Muytjens and Kollee, 1990; Arseni et al., 1985; Biering et al., 1989; Clark et al., 1990; Willis and Robinson, 1988; Simmons et al., 1989; Noriega et al., 1990; Gallagher and Ball, 1991; Van Acker et al., 2001; Burdette and Santos, 2000; Bar-Oz et al., 2001; Block et al., 2002; Himelright et al., 2002; Weir, 2002; Ministry of Health, New Zealand, 2005; Coignard and Valliant, 2004; Coignard et al., 2006). Infants less than 60 days old appear to be at the greatest risk of infection (FAO/WHO, 2004). Although the unusually high nutritional needs of premature neonates require supplementing their diets (traditionally accomplished with powdered infant formulas), studies are currently underway to develop and market sterile liquid infant formulas that would meet the nutritional needs of this group of infants (Olson, 2006). The true incidence of E. sakazakii-associated illnesses is not known, although infections have been estimated at 1.2 cases per 100,000 infants per year, and 8.7–9.4 cases per 100,000 low and very low birth weight infants per year (Braden, 2006; Stoll et al., 2004). Mortality rates for E. sakazakii-associated neonatal meningitis have been estimated to be between 40 and 80% (Lai, 2001; FAO/WHO, 2006), although 94% of survivors have been reported to experience long-term neurological impairment (Drudy et al., 2006). Powdered infant formulas have been shown to contain heat-stable endotoxins, which, when consumed, may increase the chances of intestinal E. sakazakii invasion (Townsend et al., 2006). In an international case

42

R.G. Behling et al.

study of 46 infants with invasive E. sakazakii-associated infections, 92% of patients, for whom feeding information was available, had consumed reconstituted powdered infant formulas (Bowen and Braden, 2006). An infectious dose for E. sakazakii has not yet been established and estimates range from 1 up to 1,000 cfu (Havelaar and Zwietering, 2004; Iversen and Forsythe, 2003). Non-primates are currently being examined as potential models for human pathogenicity (Lenati et al., 2006; FAO/WHO, 2006).

2.11.4 Regulation Previous USFDA microbiological guidelines for powdered infant formula were ≤10,000 cfu/g for aerobic plate count, ≤3.05 MPN cfu/g for coliforms (including the “so-called” fecal coliforms) and S. aureus, and ≤100 cfu/g for B. cereus, along with “negative” in 60 × 25 g samples for Salmonella, and negative for L. monocytogenes in their proposed guidelines of 1996 (USFDA, 1996). However, the United States FDA (2006) has “tentatively” dropped consideration of a requirement of testing for microbes other than Salmonella (“negative” in n = 60 × 25 g) and E. sakazakii (n = 30 × 25 g samples) in their 2006 proposed guidelines. The Codex Alimentarius Commission (1979) established the microbiological criteria for powdered infant formula for mesophilic aerobic bacteria (n = 5, c = 2, m = 1,000, and M = 10,000), coliforms (n = 5, c = 1, m < 3 MPN/g, and M = 20 cfu/g), and Salmonella negative in 60 × 25 g samples. (In these sampling schemes “n” is the number of samples taken and tested per lot, “c” is the number of samples allowed to be greater than “m” but less than “M,” where “M” is represents a number associated with automatic lot rejection, see Chapter 8). Powdered infant formula performance standards (in cfu/g) for Canada have been set for the aerobic plate count (n = 5, c = 2, m = 1,000/g, and M = 10,000/g), E. coli (n = 5, c = 1, m < 1.8/g and M = 10/g), Salmonella (negative in 20 samples per lot), S. aureus (n = 10, c = 1, m = 10 cfu/g, and M = 100 cfu/g), B. cereus (n = 10, c = 1, m = 100 cfu/g, and M = 10,000/g), and C. perfringens (n = 10, c = 1, m = 100 cfu/g, and M = 1,000 cfu/g) (Health Canada, 2006). The European Food Safety Authority (2004) has recommended a performance objective of the absence of E. sakazakii or Salmonella in 1, 10, or 100 kg of powdered infant formula and follow-up formulas.

2.11.5 Food Industry Concerns Numerous studies have confirmed the presence of E. sakazakii in commercially produced powdered infant formula. Contamination incidence in international surveys have ranged from 2.4 to 14% (Muytjens et al., 1988; Iversen and Forsythe, 2004) while FDA field surveys have reported a 6.6% incidence (Zink, 2003). Contamination levels in formulas that test positive for the pathogen, however, are usually < 1.0 cfu/100 g of powdered infant formula as determined by the MPN

2

Selected Pathogens of Concern to Industrial Food Processors

43

method. Between the years of 2002 and 2005, at least seven voluntary recalls of powdered infant formula were issued due to possible contamination with E. sakazakii (IBFAN, 2005). Contamination of powdered infant formula with E. sakazakii that occurs in manufacturing plants is expected to be a post-processing phenomenon such as might occur in the dry-mixing of finished product or during filling and packaging. Numerous ingredients added to powdered infant formulas may serve as potential sources for introducing pathogens into a processing environment, including lactose, whey protein concentrate, vegetable oil, vitamin and mineral pre-mixes, soy protein isolate, sucrose, corn syrup solids, and corn maltodextrin. Contamination appears to take place after the bactericidal heat treatment step of the hydrated dry powder base, which is a process that involves a sufficiently high temperature to destroy the pathogen. The presence of E. sakazakii in the processing environment, in processing equipment that may come into direct contact with the product, and in ingredients that may be mixed into the dry base powder are the three major factors outlined by the FAO/WHO as contributing to recontamination of powdered infant formulas with the pathogen (2006). Integrated Enterobacteriaceae and E. sakazakii testing programs of environmental samples, product contact surfaces, and finished products, detailed by the ICMSF (2002a), have been recommended by the FAO/WHO (2006) for infant formula manufacturers. The FDA is currently proposing the development of analytical techniques and a standard reporting tool to guide E. sakazakii investigations along with an accompanying questionnaire (Guzewich, 2006). One report from a powdered infant formula processing representative stated that proactive measures are being taken to enhance the bacteriological safety of powdered infant formulas including modifications to HACCP plans, training and awareness programs for employees, environmental monitoring and air sampling programs, plant sanitation, GMP/procedural analysis, and state-of-the-art E. sakazakii testing (Olson, 2006). Nevertheless the widespread nature of the organism, its ability to adapt to dry processing environments, and the statistical improbability of finding it in bulk dry ingredients (see Chapter 8) make control of Cronobacter spp. very difficult in the processing environment.

2.12 M. avium subspecies paratuberculosis, an Emerging Pathogen 2.12.1 The Organism Mycobacteria are gram-positive, rod-shaped bacteria. Their unique cell wall, comprised of complex lipids, causes them to stain “acid fast.” This cell wall composition results in cell clumping and also inhibits their ability to absorb nutrients, which could account for their slow growth compared to other human pathogens (Sung and Collins, 1998). Pathogenic members of the genus Mycobacteria include Mycobacterium tuberculosis, Mycobacterium bovis, and M. leprae. Several members are opportunistic pathogens for humans and include Mycobacterium

44

R.G. Behling et al.

kansasii, Mycobacterium scrofulaceum, Mycobacterium avium-intracellulare, Mycobacterium fortuitum, Mycobacterium marinum, Mycobacterium ulcerans, and Mycobacterium smegmatis. Bacillus Calmette-Guerin is an attenuated form of M. bovis and is used in some countries as a vaccine against M. tuberculosis. Mycobacteria do not produce typical exotoxins or endotoxins. Disease processes generally result in delayed-type hypersensitivity reactions in response to Mycobacterial proteins.

2.12.2 Disease Mycobacterium paratuberculosis (MAP) is the etiologic agent of Johne’s disease in cattle and other ruminants. Johne’s disease is a chronic, progressive, and severe gastrointestinal illness (Harris and Barletta, 2001). It is characterized by chronic or intermittent diarrhea, emaciation, and death (Stabel, 1998) and has been of worldwide significance for many decades (Doyle, 1956; Harris and Barletta, 2001). The disease affects at least 10% of cows in 22% of US herds (Wells et al., 1999). However, some estimates have ranged as high as 21–54% of herds in the United States and Canada (Hermon-Taylor, 2001). This organism causes chronic intestinal inflammation in both large and small ruminants, monogastrics (e.g., dog and pigs) and at least four types of non-human primates (Hermon-Taylor, 2001). There is evidence that MAP is associated with Crohn’s disease (CD) in humans. Specifically, MAP has been cultured from intestinal tissues, breast milk, and the blood of CD patients (Chiodini et al., 1984; 1986; McFadden et al., 1987; Mishina et al., 1996; Naser et al., 2000; Hermon-Taylor, 2001; Bull et al., 2003; Naser et al., 2004). However, its role in Crohn’s disease in humans is controversial (Stabel, 1998). Despite the aforementioned studies supporting an association of MAP and CD, other investigations have been unable to show any association or substantiate evidence from these previous studies (Ellingson et al., 2003; Baksh et al., 2004; Freeman and Noble, 2005; Lozano-Leon et al., 2006). The disease syndrome produced by MAP in cattle has a similar pathology as Crohn’s disease in humans. In addition, isolates of MAP that have been recovered from human Crohn’s lesions have induced Johne’s disease in infant goats (van Kruiningen et al., 1986) and young chickens (van Kruningen et al., 1991). Due to the conflicting scientific reports, the association of MAP with Crohn’s disease is a highly debated topic.

2.12.3 Costs The economic impact of Johne’s disease to the US cattle industry has been estimated at 1.5 billion dollars per year (Jones, 1989). In one study 3% of 350 beef cattle surveyed at three slaughter facilities showed evidence of MAP infection (by fecal

2

Selected Pathogens of Concern to Industrial Food Processors

45

culture or ileocecal lymph nodes, Rossiter and Henning, 2001). The prevalence of MAP in some herds has been estimated as high as 34% (Collins et al., 1994).

2.12.4 Reservoirs MAP multiplies mainly in the lymphatic system and intestinal tract of infected species (Kennedy et al., 2001). “Vast numbers” of the organism were reported in feces (Doyle, 1956). In fact, fecal contamination appears to be the principal means by which this organism is transmitted through the environment (Collins, 2001; Boor, 2001). Chiodini and Hermon-Taylor (1993) indicated that millions of cells may be shed in feces of clinically or sub-clinically infected cattle. Chiodini (1989) reported that up to 108 cells of MAP per gram could be shed during the clinical phase of infection. This is an important factor to consider regarding potential fecal–oral contamination. Some believe that this organism may be transferred to the human population from infected animals (e.g., through raw or inadequately pasteurized milk from infected dairy cattle). The potential for other routes of human exposure and potential infection (e.g., meats, water, and the environment) has largely been ignored. Hermon-Taylor (2001) has stated that “raw and processed meats are also at risk,” despite attention largely focused on the role that milk contamination may play in MAP exposure in humans and the lack of published studies on MAP survival in meats (Collins, 2001).

2.12.5 Food Processing Issues 2.12.5.1 Heat Resistance Milkborne transmission of tuberculosis by Mycobacterium bovis was common before pasteurization of fluid dairy products, widespread refrigeration and other quality enhancements became commonplace after World War II (Bryan, 1983). Early milk pasteurization parameters were based on destruction of this organism until it was realized that the ricketsial pathogen, Coxiella burnetii was more heat resistant. The heat resistance of MAP is considered high for a non-spore-forming organism (Stabel et al., 2001) and Sung and Collins (1998) showed that its heat resistance in milk was greater than that of Salmonella, L. monocytogenes, and C. burnetii. Furthermore, MAP has been isolated from pasteurized milk in the UK (Simmons, 2001). More recently Ellingson et al. (2005) showed that 2.8% of 702 pints of previously unopened pasteurized retail whole milk collected from California, Minnesota, and Wisconsin over a 12-month period contained viable MAP. MAP is considered to be an obligate intracellular pathogen; however, special laboratory medias can be used to recover the microbe when they are supplemented with Mycobactin J. Consequently, the likelihood that the microbe forms microbial growth niches in factory environments seems small, though uninvestigated. Hence the possibility of post-pasteurization contamination may be reduced compared to other organisms that

46

R.G. Behling et al.

are known to form growth niches on the processing plant environment, lending some credence to those who think the microbe may survive pasteurization. Nevertheless, controversy, exists regarding whether or not the organism survives milk pasteurization despite a number of studies dealing with this issue (see Stabel, 1998 for a discussion of some of these studies). This may be due to many factors including differences in methodologies used to heat-treat milk, those used to recover the organism, the ability of the microbe to form clumps, and the low numbers (e.g., 5–8 cfu/ml) of MAP shed in raw milk of infected (clinically or asymptomatic) cows (Stable et al., 2001). Hence the risk of infection from milk may be low due to low numbers likely to be present in raw and hence pasteurized milk. Clark et al. (2005) tested 101 cheese samples in a limited study taken over a 6-month period from Northern and Southern Wisconsin and Minnesota and found no evidence of viable MAP. They found no evidence of viable MAP in 0.17 g per sample (1 ml of a 1/6 dilution of 5 g product). However, 9.2% of the samples (6 of 65) contained hspX and IS900 genetic elements consistent with the present of MAP DNA. The presence of MAP in cooked or smoked meat products has not been investigated. Fecal contamination of meats from the hide and intestinal tract of cattle during slaughter is a well-known phenomenon. Furthermore, MAP has been shown to survive for up to 152–146 days in naturally contaminated feces under different conditions (Lovell et al., 1942). Collins et al. (2001) stated that “. . . as with any organism found in feces, post mortem contamination of the carcass and products made from the carcass, in particular ground beef, is possible.” Hence, occasional contamination of carcasses from MAP in feces will occur. Furthermore, infected animals (beef or cows) may become septicemic (Collins, 2001). Animal tissues from slaughtered cattle with Johne’s disease could, therefore, have very high numbers of MAP per gram of meat tissue. It is presently unknown whether or not MAP survives meat cooking or smoke house treatments. Chapter 5 deals with the factors that influence microbial heat resistance, among other things. 2.12.5.2 Reason for Lack of Information About MAP Until relatively recently, the scientific community has tended to ignore MAP due to the extended time and effort needed for culturing (e.g., 2 months or more to recover visible colonies) by traditional approaches. This fact resulted in reliance upon harsh sample decontamination regimes used with traditional methods to eliminate many other organisms that grow faster. These decontamination regimes also have negative effects on MAP recovery (Johansen et al., 2006) and may cause cells to become viable but non-culturable. This is inferred from the fact that the environment of granulomas, in which M. tuberculosis cells can exist in a dormant state in recovering tuberculosis patients, is believed to be very harsh. The environment within granulomas is characterized by low oxygen, high CO2, acidic pH, and the presence of aliphatic organic acids. It was originally thought that M. tuberculosis would die quickly in the harsh environment within a granuloma, but it was later found that it can survive many years (Cunningham and Spreaddbury, 1998 quoting Wayne and

2

Selected Pathogens of Concern to Industrial Food Processors

47

Salkin, 1956). Yanmin et al. (2000) postulated that “persisting M. tuberculosis may exist in some physiological form in which limited metabolism, presumably with little or no cell turnover, accounting for tolerance to conventional antibiotic treatment.” Current assays were borrowed from clinical and veterinary microbiology assays developed to detect high numbers of cells in infected tissue or fecal samples. Hence, there is a need to adapt methods more suitable to recover low numbers of stressed or injured MAP likely to be found in food, feed, or some environmental samples. Modern molecular techniques such as PCR show some promise but results from PCR methodology, which may take but a few hours or days, have been criticized because of uncertainty whether or not positive sample assays detected living cells or merely DNA from dead cells (Stabel, 1998). The issue becomes a bit more confused when one considers that MAP was also detected in some milk samples by cultural methods, but not by PCR (Millar et al., 1996). These findings leave open at least three questions: (1) Were failures to recover live cells a consequence of PCR detection of dead cells? (2) Did failures to recover live cells from PCR-positive samples occur because living cells were in a viable but non-culturable state? (3) Were failures to recover live cells from PCR-positive samples and vice versa an artifact of sampling when low numbers of cells are present? A diagnostic test for MAP that can be used on food will most likely have to be rapid to allow testing within each product’s shelf life, differentiate live organisms from those inactivated during processing and extremely sensitive so that it can detect MAP with no enrichment. For these reasons, immunomagnetic capture coupled with a very sensitive PCR-based assay is currently attracting a lot of research interest.

2.12.6 Some Research Needs Suggested research needs related to the role of MAP in foods include the following: 1. A better understanding of the role of MAP in Crohn’s disease. 2. More sensitive and rapid techniques for recovery of viable MAP in foods, feeds, and the environment. 3. Better understanding of the microbial ecology of the microbe both in the food manufacturing environment, the natural environment, and other foods including meats and poultry. 4. Development of a suitable surrogate microorganism capable of more rapid growth and/or detection. Use of a light emitting strain of MAP has also been proposed for such work (Boor, 2001). 5. Development of a suitable thermal surrogate organism, which could be used to inexpensively validate thermal processes with techniques available to a typical commercial laboratory. 6. Means and development of processing systems related to keeping this organism out of the pasteurized milk supply.

48

R.G. Behling et al.

References Adams (1973) Inactivation of Clostridium perfringens type A spores at ultrahigh temperatures. Appl Microbiol 26(3):282–287 Allan JT, Yan Z, Genzlinger LL, Kornacki. JL (2004) Temperature and biological soil effects on the survival of selected foodborne pathogens on a mortar surface. J Food Prot 67(12): 2661–2665 Altekruse SF, Stern NJ, Fields PI, Swerdlow DL (1999) Campylobacter jejuni – An emerging foodborne pathogen. Emerg Infect Dis 5:28–35 Alter T, Scherer K (2006) Stress response of Campylobacter spp. and its role in food processing. J Vet Med 53:351–357 Arseni A, Malamou-Lada H, Kostalos C, Sta-I-Kou A, Koustia H (1985) A fatal case of sepsis in a premature newborn baby associated with Enterobacter sakazakii bacteremia. Acta Microbiol Hell 29:402–407 Assadi MM, Mathur RP (1991) Application of an HPLC system in the analysis of biodegraded crude oil compounds. J Liq Chromatogr 14:3623–3629 Atabay HI, Corry JEL (1997) The prevalence of campylobacters and arcobacters in broiler chickens. J Appl Microbiol 83:619–626 Atabay HI, Corry JEL, On SLW (1998) Diversity and prevalence of Arcobacter spp. in broiler chickens. J Appl Microbiol 84:1007–1016 Baksh FK, Finkelstein SD, Ariyanayagam-Baksh SM, et al (2004) Absence of Mycobacterium avium subsp. paratuberculosis in the microdissected granulomas of Crohn’s disease. Mod Pathol 10:1289–1294 Bar-Oz B, Preminger A, Peleg O, Block C, Arad I (2001) Enterobacter sakazakii infection in the newborn. Acta-Paediatrica 90:356–358 Bennett R, Belay N (2001) Bacillus cereus, Chapter 32. In: Downes FB, Ito K (eds) Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC Besser RE, Griffin PM, Slutsker L (1999) Escherichia coli O157:H7 gastroenteritis and the hemolytic uremic syndrome: An emerging infectious disease. Annu Rev Med 50:355–367 Biering G, Karlsson S, Clark NVC, Jonsdottir KE, Ludvigsson P, Steingrimsson O (1989) Three cases of neonatal meningitis caused by Enterobacter sakazakii in powdered milk. J Clin Microbiol 27:2054–2056 Block C, Peleg O, Minster N, Bar-Oz B, Simhon A, Arad I, Shapiro M (2002) Cluster of Neonatal infections in Jerusalem due to unusual biochemical variant of Enterobacter sakazakii. Eur J Clin Microbiol Infect Dis 21:613–616 Boor K (2001) Molecular approaches for monitoring Mycobacterium paratuberculosis. Talk presented at the Food Microbiology Research Conference XVIII. Chicago, 5 Nov 2001 Bowen AB, Braden CR (2006) Invasive Enterobacter sakazakii disease in infants. Emerg Infect Dis 12:1185–1189 Braconnier A, Broussolle V, Dargaignaratz C, Nguyen-The C, Carlin F (2003) Growth and germination of proteolytic Clostridium botulinum in vegetable-based media. J Food Prot 66:833–839 Braden C (2006) Clinical and epidemiological significance of E. sakazakii: Symposium on Enterobacter sakazakii. Int Assoc Food Prot Ann Meeting, Calgary Alberta, 13–16 Aug 2006 Bradshaw M, Dineen SS, Maks ND, Johnson EA (2004) Regulation of neurotoxin complex expression in Clostridium botulinum strains 62A, Hall A-hyper, and NCTC 2916. Anaerobe 10:321–333 Bradshaw JG, Peeler JT, Twedt RM (1977) Thermal inactivation of ileal loop-reactive Clostridium perfringens type A strains in phosphate buffer and beef gravy. Appl Environ Microbiol 34(3):280–284 Breeuwer P, Lardeau A, Peterz M, Joosten HM (2003) Desiccation and heat tolerance of Enterobacter sakazakii. J Appl Microbiol 95:967–973

2

Selected Pathogens of Concern to Industrial Food Processors

49

Breeuwer P, Michot L, Joosten H (2004) Genetic basis of dry stress resistance of Enterobacter sakazakii, Abstract #T75, Program and Abstract Book, p. 156. 91st Ann Mtg, Int Assn Food Prot, Phoenix, 8–11 August 2004 Brunner KG, Wong ACL (1992) Staphylococcus aureus growth and enterotoxin production in mushrooms. J Food Sci 57(3):700–703 Bryan FL (1983) Epidermiology of milk-borne diseases. J Food Prot 46:637–649 Bull TJ, McMinn EJ, Sidi-Boumedine K, et al (2003) Detection and verification of Mycobacterium avium subsp. paratuberculosis in fresh ileocolonic mucosal biopsy specimens from individuals with and without Crohn’s Disease. J Clin Microbiol 41:2915–2923 Burdette JH, Santos C (2000) Enterobacter sakazakii brain abscess in the neonate: The importance of neuroradiologic imaging. Pediatr Radiol 30:33–34 Butzler JP (2004) Campylobacter, from obscurity to celebrity. Clin Microbiol Infect 10:868–876 Buzby JC, Roberts T (1996) ERS Updates US Foodborne Disease Costs for Seven Pathogens. Food Rev 19:20–25 CDC (1994) Bacillus cereus food poisoning associated with fried rice at two child day care centersVirginia, 1993. MMWR 43:177–178 CDC (1998) Botulism in the United States, 1899–1996. In: Handbook for Epidemiologists, Clinicians, and Laboratory Workers. CDC, Atlanta CDC (2000) Surveillance summaries. March 17. MMWR 49(No. SS–1) CDC (2001) Diagnosis and Management of Foodborne Illness: A Primer for Physicians. January 26, 50(RR02): 1–69 CDC (2004) Diagnosis and management of foodborne illnesses: A primer for physicians, 2nd rev. MMWR April 16. 53(RR 04):1–33. http://www.cdc.gov/mmwr/preview/ mmwrhtml/rr5304a1.htm. Accessed 24 April 2009 CDC (2007) Multi-state outbreak of Salmonella serotype Tennessee infections associated with peanut butter – United States, 2006–2007. June 1. MMWR 56(21):521–524 CDC (2009) Multistate outbreak of Salmonella infections associated with peanut butter and peanut butter-containing products-United States, 2008–2009. January 29. MMWR 58(Early Release):1–6 Callicott KA, Friethriksdottir V, Reiersen J, Lowman R, Bisaillon JR, Gunnarsson E, Berndtson E, Hiett KL, Needleman DS, Stern NJ (2006) Lack of evidence for vertical transmission of Campylobacter spp. in chickens. Appl Environ Microbiol 72:5794–5798 Caubilla-Barron J, Forsythe SJ (2006) P-019. Long-term persistence and recovery of Enterobacter sakazakii and other Enterobacteriaceae from powdered infant milk formula. Am Soc Microbiol Gen Mtg. Orlando, 21–25 May 2006 Chiodini RJ (1989) Crohn’s disease and the mycobacterioses: A review and comparison of two disease entities. Clin Microbiol Rev 2:90–117 Chiodini RJ, Herman-Taylor J (1993) The thermal resistance of Mycobacterium paratuberculosis in raw milk under conditions simulating pasteurization. J Vet Diagn Invest 5:629–631 Chiodini RJ, Kruiningen HJ, Thayer WR, et al (1986) Spheroplastic phase of mycobacteria isolated from patients with Crohn’s disease. J Clin Microbiol 24:357–363 Chiodini RJ, Van Kruingen HJ, Merkal RS, Thayer WR, Jr., Coutu JA (1984) Characteristics of an unclassified Mycobacterium species isolated from patients with Crohn’s disease. J Clin Microbiol 20:966–971 Clark DL, Anderson JL, Koziczkowski JJ, Ellingson JL (2005) Prevalence of Mycobacterium avium subspecies paratuberculosis in retail cheese curds from Wisconsin and Minnesota. Poster P4-43. IAFP Annual Meeting, Phoenix, AZ. J Food Prot 68(Supplement A):135 Clark NC, Hill BC, O’Hara CM, Steingrimsson O, Cooksey RC (1990) Epidemiologic typing of Enterobacter sakazakii in two neonatal nosocomial outbreaks. Diag Microbiol Infect Dis 13:467–472 Codex Alimentarius Commission (CAC) (1979) Recommended international code of hygienic practice for foods for infants and children. CAC/ RCP 21-1979

50

R.G. Behling et al.

Coignard B, Valliant V (2004) Infections à Enterobacter sakazakii assssociées à la consommation d’une préparation en poudre pour nourrissons. France, octobre à decembre 2004. Rapport d’investigation. (Enterobacter sakazakii infections associated with the consumption of powdered infant formula. France, October to December 2004. Investigation report.) Editions InVS, Saint-Maurice, France, 2006. 88p. Also available at http://www.invs.sante.fr/display/?doc = publications/2006/infections_e_sakazakii/index.html. Accessed 4 May 2009 Coignard B, Vaillant V, Vincent J-P, Lefléche A, Mariani-Kurkdjian P, Bernet C, L’Hériteau F, Sénéchal H, Grimont P, Bingen E, Desenclos JC (2006) Infections sévères à Enterobacter sakazakii chez des nouveau-nés ayant consommé une préparation en poudre pour nourrissons, France, octobre-décembre 2004. (Enterobacter sakazakii infections associated with the consumption of powdered infant formula. France, October to December 2004. Investigation report.) BEH no. 2–3, 10–13 Collins MT (2001) Ecological and physical characteristics of Mycobacterium paratuberculosis. Talk presented at IAFP Symposium S14 – Mycobacterium paratuberculosis: Villain or Bystander? Minneapolis, 5–8 Aug 2001 Collins CI, Wesley IV, Murano EA (1996) Detection of Arcobacter spp. in ground pork by modified plating methods. J Food Prot 59:448–452 Collins MT, Sockett DC, Goodger WJ, Conrad TA, Thomas CB, Carr DJ (1994) Herd prevalence and geographic distribution of, and risk factors for, bovine paratuberculosis in Wisconsin. J Am Vet Med Assoc 204:636–641 Collins MT, Spahr U, Murphy PM (2001) Ecological characteristics of M. paratuberculosis, Chapter 3 . In: Mycobacterium paratuberculosis. Bulletin of the International Dairy Federation No. 362. IDF/FIL, Brussels, Belgium. Corry JEL, Atabay HI (2001) Poultry as a source of Campylobacter and related organisms. J Appl Microbiol 90:96S–114S Cottyn B, Regalado E, Lanoot B, De Cleene M, Mew TW, Swings J (2001) Bacterial populations associated with rice seed in the tropical environment. Phytopathology 91:282–292 Cox NA, Stern NJ, Hiett KL, Berrang ME (2002) Identification of a new source of Campylobacter contamination in poultry: Transmission from breeder hens to broiler chickens. Avian Dis 46:535–541 Cruz AC, Fernandez E, Salinas E, Ramirez P, Montiel C, Eslava CA (2004) Q-051. Characterization of Enterobacter sakazakii isolated from different sources. Am Soc for Microbiol 104th Gen Mtg. New Orleans, 23–27 May 2004 Cunningham AF, Spreadbury CL (1998) Mycobacterial stationary phase induced by low oxygen tension: Cell wall thickening and localization of the 16 kilodalton a-crystalin homolog. J Bacteriol 180:801–808 D’Aoust JY (1991) Pathogenicity of foodborne Salmonella. Int J Food Microbiol 12:17–40 D’Aoust JY (1997) Salmonella species, Chapter 8. In: Doyle MP, Beuchat LR, Montville TJ (eds) Food Microbiology: Fundamentals and Frontiers. American Society for Microbiology Press, Washington, DC, pp. 129–158 D’Aoust JY, Aris BJ, Thisdele P, Durante A, Brisson N, Dragon D, Lachapelle G, Johnston M, Laidley R (1975) Salmonella eastbourne outbreak associated with chocolate. Can Inst Food Sci Technol J 8:181–184 D’Aoust JY, Warburton DW, Sewell AM (1985) Salmonella typhimurium phage type 10 from cheddar cheese implicated in a major Canadian foodborne outbreak. J Food Prot 48:1062–1066 Datta AR (2003) Listeria monocytogenes, Chapter 7. In: Milliotis MD, Bier JW (eds) International Handbook of Foodborne Pathogens. CRC, Danvers, MA, pp. 105–122 David JRD, Graves RH, Carlson VR (1996) Aseptic Processing and Packaging of Food: An Industry Perspective. CRC Press, Boca Raton, FL, p. 185 Doyle LP (1944) A Vibrio associated with swine dysentery. Am J Vet Res 5:3–5 Doyle MP (ed) (1989) Foodborne Bacterial Pathogens. Marcel Dekker, New York Doyle MP, Beuchat LR, Montville TJ (eds) (1997a) Food Microbiology Fundamentals and Frontiers. ASM Press, Washington, DC

2

Selected Pathogens of Concern to Industrial Food Processors

51

Doyle MP, Roman DJ (1982) Response of Campylobacter jejuni to sodium chloride. Appl Environ Microbiol 43:561–565 Doyle MP, Schoeni J (1994) Survival and growth characteristics of Escherichia coli associated with hemorrhagic colitis. Appl Environ Microbiol 48:855–856 Doyle MP, Zhao T, Meng J, Zhao S (1997b) Escherichia coli O157:H7, Chapter 10. In: Doyle MP, Beuchat LR, Montville TJ (eds) Food Microbiology: Fundamentals and Frontiers. Amer Soc Microbiol Press, Washington, DC, pp. 171–191 Doyle TM (1956) Johne’s disease. Vet Rec 68:869–878 Drudy D, Mullane NR, Quinn T, Wall PG, Fanning S (2006) Enterobacter sakazakii: An emerging pathogen in powdered infant formula. Clin Infect Dis 42:996–1002 Edelson-Mammel SG, Porteous M, Buchanan R (2005) Survival of Enterobacter sakazakii in a dehydrated powdered infant formula. J Food Prot 68:1900–1902 Eifert JD, Castle RM, Pierson FW, Hackney CR, Larsen C (2003) Comparison of sampling techniques for detection of Arcobacter butzleri from chickens. Poult Sci 82(12):1898–1902 Ellingson JL, Anderson JL, Kozickowski JJ, Radcliff RP, Sloan SJ, Allen SE, Sullivan NM (2005) Detection of viable Mycobacterium avium subsp. paratuberculosis in retail pasteurized whole milk by two culture methods and PCR. J Food Prot 68(5):966–972 Ellingson JLE, Cheville JC, Brees D, et al (2003) Absence of Mycobacterium avium subspecies paratuberculosis components from Crohn’s disease intestinal biopsy tissues. Clin Med Res 1:217–226 Emilani F, Lajmanovich R, Gonzales SM (2001) Escherichia coli: Biochemical phenotype diversity in fresh waters (Santa Fe Province, Argentina). Rev Argent De Microbiol 33: 65–74 Erickson M, Kornacki JL (2003) Bacillus anthracis: Current knowledge in relation to contamination in food. J Food Prot 66(4):691–699 Eslava C, Villaseca J, Hernandez U, Cravioto A (2003) Escherichia coli, Chapter 8. In: Miliotis M, Bier J (eds) International Handbook of Foodborne Pathogens. Marcel Dekker, New York, pp. 123–135 Espeland EM, Wetzel RG (2001) Complexation, stabilization, and UV photolysis of extracellular and surface-bound glucosidase and alkaline phosphatase: Implications for biofilm microbiota. Microbiol Ecol 42:572–585 European Food Safety Authority (2004) Microbiological risks in infant formulae and follow-on formulae. EFSA J, 113:1–35. http://www.efsa.europa.eu/en/science/biohaz/biohaz_ opinions/691.html. Accessed 28 April 2009 Farmer JJ, Asbury MA, Hickman FW, Brenner DJ, and the Enterobacteriaceae Study Group (1980) Enterobacter sakazakii, new species of Enterobacteriaceae isolated from clinical specimens. Int J Syst Bacteriol 30:569–584 FDA (1996) 21CFR Parts 106 and 107, Current good manufacturing practice, quality control procedures, quality factors, notification requirements, and records and reports for the production of infant formula; proposed rule. Federal Register July 9. 61(132): 36153–36219. http://www.cfsan.fda.gov/∼lrd/fr960709.html. Accessed 5 May 2009 FDA (2001) Foodborne Pathogenic Microorganisms and Natural Toxins Handbook. http://www.cfsan.fda.gov/∼mow/chap15.html. Accessed 24 April 2009 FDA (2002a) Isolation and enumeration of Enterobacter sakazakii from dehydrated powdered infant formula. http://www.cfsan.fda.gov/∼comm/mmesakaz.html. Accessed 5 Nov 2003 FDA (2002b) Questions and answers on method for E. Sakazakii in powdered infant formula. Aug 2002. http://www.cfsan.fda.gov/∼comm/mmesakqa.html. Accessed 21 Oct 2003 FDA (2006) 21CFR Parts 106 and 107, Current good manufacturing practice, quality control procedures, quality factors, notification requirements, and records for the production of infant formula: Reopening of comment period. Federal Register Aug 1 71(147):43392–43398 FDA (2008a) Guidance for industry control of Listeria in refrigerated or frozen readyto-eat foods. Draft Guidance. (Contains Nonbinding Recommendations). February. http://www.cfsan.fda.gov/∼dms/lmrtegui.html. Accessed 23 April 2009

52

R.G. Behling et al.

FDA (2008b) Compliance policy guide guidance for FDA staff Sec. 555.320, Listeria monocytogenes: Draft guidance. http://www.fda.gov/OHRMS/DOCKETS/98fr/FDA-2008-D-0058GDL.pdf. Accessed 23 April 2009 FDA-CFSAN/USDA-FSIS (2003) USDA_FSIS/ FDA-CFSAN/CDC. 2003. Quantitative Assessment of Relative Risk to Public Health from Foodborne Listeria monocytogenes Among Selected Categories of Ready-to-Eat Foods. September. http://www.fda.gov/Food/Science Research/ResearchAreas/RiskAssessmentSafetyAssessment/ucm183966.htm Accessed March 10, 2010. Feng P (1995) Escherichia coli O157:H7: Novel vehicles of infection and emergence of phenotypic variants. Emerg Infect Dis April–June 1(2):1–8 Festy B, Squinazi F, Marin M, Derimay R, Lior H (1993) Poultry meat and water as the possible sources of Arcobacter butzleri associated with human disease in Paris, France. Acta GastroEnterol Belg 56(suppl. 34):35 Food and Agriculture Organization/World Health Organization (2004) Joint FAO/WHO workshop on Enterbacter sakazakii and other microorganisms in powdered infant formula, Geneva, 2–5 Feb 2004. http://www.who.int/foodsafety/micro/jemra/meetings/feb2004/en/. Accessed 28 April 2009 Food and Agriculture Organization/World Health Organization (2006) Joint FAO/WHO workshop on Enterbacter sakazakii and Salmonella in powdered infant formula, FAO headquarters Rome, Italy, 16–20 January, 2006. http://www.ibfanasia.org/Contaminats/FAOWHO%20Report%20E%20sakakazii%20and%20Salmonella%20in%20PIF%20Advance%20 copy.pdf. Accessed 5 May 2009 Food Science Australia (2000) Sprouts and foodborne disease. In: Food Safety and Hygiene – A bulletin for the Australian Food Industry. Food Science Australia. November. Francis DH, Collins JE, Duimstra JR (1986) Infection of gnotobiotic pigs with an Escherichia coli O157:H7 strain associated with an outbreak of hemorrhagic colitis. Infect Immun 51: 953–956 Franco DA, Williams CE (1994) Campylobacter jejuni. In: Hui YH, Gorham JR, Murrell KD, Cliver DO (eds) Foodborne Disease Handbook, vol 1. Marcel Dekker, New York, pp. 71–96 Freeman H, Noble M (2005) Lack of evidence for Mycobacterium avium subspecies paratuberculosis in Crohn’s disease. Inflamm Bowel Dis 11:782–783 Gakuya FM, Kyule MN, Gathura PB, Kariuki S (2001) Antimicrobial resistance of bacterial organisms isolated from rats. East Afr Med J 78:646–649 Gallagher PG, Ball WS (1991) Cerebral infarctions due to CNS infection with Enterobacter sakazakii. Pediatr Radiol 21:135–136 Gassem MAA (1999) Study of the micro-organisms associated with fermented bread (khamir) produced from sorghum in Gizan region, Saudi Arabia. J Appl Microbiol 86:221–225 Gassem MAA (2002) A microbiological study of sobia: A fermented beverage in the Western province of Saudi Arabia. World J Microbiol Biotechnol 18:173–177 Georgsson F, Thornorkelsson AE, Geirsdottir M, Reiersen J, Stern NJ (2006) The influence of freezing and duration of storage on Campylobacter and indicator bacteria in broiler carcasses. Food Microbiol 23:677–683 Gill CO, Harris LM (1983) Limiting conditions of temperature and pH for growth of thermophilic campylobacters on solid media. J Food Prot 46:767–768 Glass KA, Johnson EA (2004) Antagonistic effect of fat on the antibotulinal activity of food preservatives and fatty acids. Food Microbiol 21:675–682 Graham AF, Mason DR, Maxwell FJ, Peck MW (1997) Effect of pH and NaCl on growth from spores of non-proteolytic Clostridium botulinum at chill temperatures. Lett Appl Microbiol 24:95–100 Grant J, Wendelboe AM, Wendel A, Jepson B, Torres P, Smelser C, Rolfs RT (2008) Spinachassociated Escherichia coli O157:H7 outbreak, Utah and New Mexico, 2006. Emerg Infect Dis October 2008 14(10):1633–1636. http://www.cdc.gov/eid/content/14/10/pdfs/1633.pdf. Accessed 5 May 2009

2

Selected Pathogens of Concern to Industrial Food Processors

53

Greenwood MH, Hopper WL (1983) Chocolate bars contaminated with Salmonella napoli: An infectivity study. Br Med J 286:1394 Griffin PM, Ostroff SM, Tauxe RV, Greene KD, Wells JG, Lewis JH, Blake P (1988) Illnesses associated with Escherichia coli O157:H7 infections. Ann Intern Med 109:705–712 Gude A, Hillman TJ, Helps CR, Allen VM, Corry JEL (2005) Ecology of Arcobacter species in chicken rearing and processing. Lett Appl Microbiol 41:82–87 Gurtler JB, Beuchat LR (2007) Survival of Enterobacter sakazakii in powdered infant formula as affected by composition, water activity, and temperature. J Food Prot 70:1579–1586 Gurtler JB, Kornacki JL, Beuchat LR (2005) Enterobacter sakazakii: A coliform of increased concern to infant health. Int J Food Microbiol 104:1–34 Guzewich J (2006) Current approaches to investigating cases of E. sakazakii: Symposium on Enterobacter sakazakii. Int Assoc Food Prot Ann Meeting, Calgary, Alberta, 13–16 Aug 2006 Hamilton JV, Lehane MJ, Braig HR (2003) Isolation of Enterobacter sakazakii from midgut of Stomoxys calcitrans. Emerg Infect Dis 9:1355–1356 Hammack S, Johnson M, French D, Reynolds A, Reynolds S, Shelton J (February 1990) Originally published as a News Release. University of Georgia, Cooperative Extension Service (www.ces.uga.edu) Hanes D (2003) Nontyphoid Salmonella, Chapter 9. In: Miliotis MD, Bier JW (eds) International Handbook of Foodborne Pathogens. Marcel Dekker, New York, pp. 137–149 Harris BA, Barletta RG (2001) Mycobacterium avium subsp. paratuberculosis in veterinary medicine. Clin Microbiol Rev 14:489–512 Hauschild AHW (1989) Clostridium botulinum. In: Doyle MP (ed) Bacterial Foodborne Pathogens. Marcel Dekker, New York, pp. 111–189 Havelaar AH, Zwietering M (2004) On the risk of Enterobacter sakazakii in infant milk formula. Trends Food Sci Technol 15:99–100 Health C (2006) Health products and food branch (HPFB) standards and guidelines for microbiological safety of food – an interpretive summary. http://www.hc-sc.gc.ca/fn-an/res-rech/analymeth/microbio/volume1/intsum-somexp_e.html. Accessed 28 November 2006 Helgason E, Økstad OA, Caugant DA, Johansen HA, Fouet A, Mock M, Hegna I, Kolstø AB (2000) Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis—one species on the basis of genetic evidence. Appl Environ Microbiol 66:2627–2630 Hennessy WH, Hedberg CW, Slutsker L, White KE, Besser-Wiek JM, Moen ME, Feldman J, Coleman WW, Edmonson LM, MacDonald KL, Osterholm MT (1996) A national outbreak of salmonella enteritidis infections from ice cream. New Engl J Med 334:1281–1286 Hermon-Taylor J (2001) Mycobacterium avium subspecies paratuberculosis: The nature of the problem. Food Control 12:331–334 Hiett KL, Cox NA, Buhr RJ, Stern NJ (2002) Genotype analysis of Campylobacter isolated from distinct segments of the reproductive tracts of broiler breeder hens. Curr Microbiol 45:400–404 Himelright I, Harris E, Lorch V, Anderson M, Jones T, Craig A, Kuehnert M, Forster T, Arduino M, Jensen B, Jernigan D (2002) Enterobacter sakazakii infections associated with the use of powdered infant formula – Tennessee. MMWR 51:298–300 Houf K, De Zutter L, Van Hoof J, Vandamme P (2002) Occurrence and distribution of Arcobacter species in poultry processing. J Food Prot 65(8):1233–1239 Hughes RA, Cornblath DR (2005) Guillain-Barre syndrome. Lancet 366:1653–1666 IAFP (1999) Procedures to Investigate Foodborne Disease, 5th edn. Table B (Illnesses Acquired by Ingestion of Contaminated Foods: A Condensed Classification by Symptoms, Incubation Periods, and Types of Agents) ICMSF (2002) Microorganims in Foods, vol 7. Springer, New York ICMSF (2002) E. coli O157:H7 in frozen raw ground beef patties, Chapter 17. In: Tomkin RB et al (eds) Microbiological Testing in Food Safety Management. Kluwer/Plenum Publishers, New York. ILSI (2005) Achieving continuous improvement in reductions in foodborne Listeriosis – A riskbased approach. J Food Prot 68:1932–1994

54

R.G. Behling et al.

International Commission for Microbiological Specifications for Foods (ICMSF) (2002) Microorganisms in Foods, vol 7: Microbiological Testing in Food Safety Management, Chapter 8. Selection of cases and attribute plans. Kluwer Academic/Plenum Publishers, New York International baby food action network (IBFAN) (2005) International code documentation centre legal update, March. http://www.ibfan.org/english/pdfs/icdc0205.pdf. Accessed 11 December 2006 Iversen C, Caubilla-Barron J, Forsythe S (2004a) P-004. Isolation of Enterobacter sakazakii, Enterobacteriaceae and other Microbial Contaminants from Powdered Infant Formula Milk and Related Products. American Society for Microbiology 104th General Meeting, New Orleans Iversen C, Druggan P, Forsythe S (2004b) A selective medium for Enterobacter sakazakii, a preliminary study. Int J Food Microbiol 96:133–139 Iversen C, Forsythe S (2003) Risk profile of Enterobacter sakazakii, an emergent pathogen associated with infant milk formula. Trends Food Sci Technol 14:443–454 Iversen C, Forsythe S (2004) Isolation of Enterobacter sakazakii and other Enterobacteriaceae from powdered formula milk and related products. Food Microbiol 21:771–777 Jablonski LM, Bohach GA (2001) Staphylococcus aureus. In: Doyle MP, Beuchat LR, Montville TJ (eds) Food Microbiology: Fundamentals and Frontiers, 2nd edn. ASM Press, Washington, DC, pp. 411–434 Jacobs-Reitsma WF (2000) Campylobacter in the food supply. In: Nachamkin I, Blaser MJ (eds) Campylobacter, 2nd edn. American Society for Microbiology, Washington, DC, pp. 467–481 Janicka G, Kania I, Ulatowska B, Krusznska E, Wojda M (1999) The occurrence of the Enterobacter genus rods in clinical material and material taken from hospital environment. Wiadomosci Lekaarskie 52:554–558 Jay J, Loessner M, Golden DA (2005) Modern Food Microbiology, 7th edn. Springer, New York Jay JM (2000) Food poisoning caused by gram-positive sporeforming bacteria. In: Modern Food Microbiology, 6th edn. ASPEN, Gaithersburg, MD, pp. 461–484 Jenson I, Moir CJ (1997) Bacillus cereus and other Bacillus species. In: Hocking AD, Arnold G, Jenson I, Newton K, Sutherland P (eds) Foodborne Microorganisms of Public Health Importance, 5th edn. AIFST (NSW Branch), Sydney, Australia, pp. 379–406 Joerger RD (2003) Alternatives to antibiotics: Bacteriocins, antimicrobial peptides and bacteriophages. Poult Sci 82:640–647 Johansen KA, Hugen EE, Payeur JB (2006) Growth of Mycobacterium avium subsp. paratuberculosis in the presence of hexadecylpyridinium chloride, natamycin, and vancomycin. J Food Prot 69:878–883 Johnson EA (1999) Clostridium botulinum Encyclopedia of Food Microbiology. Academic Press, San Diego, pp. 458–463 Jöker RN, Norholm T, Siboni KE (1965) A case of neonatal meningitis caused by a yellow Enterobacter. Danish Med Bull 12:128–130 Jones PH, Willis AT, Robinson DA, Skirrow MB, Josephs DS (1981) Campylobacter enteritis associated with the consumption of free school milk. J Hyg Camb 87:155–162 Jones RL (1989) Review of the economic impact of Johne’s disease in the United States. In: Milner AR, Woods PR (eds) Johne’s Disease: Current Trends in Research, Diagnosis and Management. Commonwealth Scientific and Industrial Research Organization, Melbourne, Australia, pp. 46–50 Joshua GW, Guthrie-Irons C, Karlyshev AV, Wren BW (2006) Biofilm formation in Campylobacter jejuni. Microbiol 152:387–396 Juneja VK, Eblen BS (1995) Influence of sodium chloride on thermal inactivation and recovery of non-proteolytic Clostridium botulinum type B strain KAP B5 spores. J Food Prot 58:813–816 Jung M, Park J (2006) Prevalence and thermal stability of Enterobacter sakazakii from unprocessed ready-to-eat agricultural products and powdered infant formulas. Food Sci Biotech 15:152–157 Kandhai MC, Reij MW, Gorris LGM, Guillaume-Gentil O, van Schothorst M (2004a) Occurrence of Enterobacter sakazakii in food production environments and households. Lancet 363:39–40

2

Selected Pathogens of Concern to Industrial Food Processors

55

Kapperud G, Gustavsen S, Hellesnes I, Hansen AH, Lassen J, Hirn J, Jahkola M, Montenegro MA, Helmuth R (1990) Outbreak of Salmonella typhimurium traced to contaminated chocolate and caused by a strain lacking the 60-megadalton virulence plasmid. J Clin Microbiol 28: 2597–2601 Keener KM, Bashor MP, Curtis PA, Sheldon BW, Kathariou S (2004) Comprehensive review of Campylobacter and poultry processing. Compr Rev Food Sci Food Saf 3:105–116 Kennedy D, HolmstrÖm A, Forshell KP, Vindell E, Suarez Fernandez G (2001) On-farm management of paratuberculosis (Johne’s disease) in dairy herds, Chapter 2. In: Mycobacterium paratuberculosis. Bulletin of the International Dairy Federation No. 362. IDF/FIL, Brussels, Belgium. Ketley JM (1997) Pathogenesis of enteric infection by Campylobacter. Microbiol 143:5–21 Kiehlbauch JA, Brenner DJ, Nicholson MA, Baker CN, Patton CM, Steigerwalt AG, Wachsmuth IK (1991) Campylobacter butzleri sp. nov. isolated from humans and animals with diarrheal illness. J Clin Microbiol 29:376–385 Kim J, Foegeding PM (1993) Principles of control. In: Hauschild AH, Dodds KL (eds) Clostridium botulinum: Ecology and Control in Foods. Marcel Dekker, New York, pp. 121–176 Kleiman MB, Allen SD, Neal P, Reynolds J (1981) Meningoencephalitis and compartmentalization of the cerebral ventricles caused by Enterobacter sakazakii. J Clin Microbiol 14: 352–354 Knill MJ, Suckling WG, Pearson AD (1982) Campylobacters from water. In: Newell DG (ed) Campylobacter: Epidemiology, Pathogenesis, Biochemistry. MTP Press, Lancaster, pp. 281–284 Kornacki JL (1998) Enterobacter sakazakii: Pursuit of a putative pathogen in a dairy powder factory (a case study). American Dairy Sci Assn Ann Mtg, Denver, 1998 Kornacki JL (2000) The nuts and bolts of food safety. Food Testing Anal 40:18–22 Kornacki JL, Johnson J (2001) Enterobacteriaceae, Coliforms, and Escherichia coli as quality and safety indicators, Chapter 8. In: Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC, pp. 69–82 Kornacki JL, Marth EH (1982) Foodborne illness caused by Escherichia coli: A review. J Food Prot 45:1051–1067 Kramer JM, Frost JA, Bolton FJ, Wareing DR (2000) Campylobacter contamination of raw meat and poultry at retail sale: Identification of multiple types and comparison with isolates from human infection. J Food Prot 63:1654–1659 Kuzina LV, Peloquin JJ, Vacek DC, Miller TA (2001) Isolation and identification of bacteria associated with adult laboratory Mexican fruit flies, Anastrepha ludens (Diptera: Tephritidae). Curr Microbiol 42:290–294 Lai KK (2001) Enterobacter sakazakii infections among neonates, infants, children, and adults: Case reports and a review of the literature. Med Baltimore 80:113–122 Lammerding AM (1996) Isolation method for the recovery of Arcobacter butzleri from fresh poultry and poultry products. In: Newell DG, Ketley JH, Feldman RA (eds) Campylobacter VIII. Proceedings of the 8th International Workshop on Campylobacters, Helicobacters and Related Organisms. Plenum Publishing, New York, pp. 329–333 Leclercq A, Wanegue C, Baylac P (2002) Comparison of fecal coliform agar and violet red bile lactose agar for fecal coliform enumeration in foods. Appl Environ Microbiol 68: 1631–1638 Lehmacher A, Bockemuhl J, Aleksik S (1995) A nationwide outbreak of human salmonellosis in Germany due to contaminated paprika, and paprika powdered potato-chips. J Infect Dis 115:501–511 Lehner A, Tasara T, Stephan R (2005) Relevant aspects of Arcobacter spp. as potential foodborne pathogen. Int J Food Microbiol 102:127–135 Lenati R, Lin M, Farber JM, Pagotto F (2006) Development of a non-primate model for Enterobacter sakazakii infection: Symposium on Enterobacter sakazakii. Int Assoc Food Prot Ann Mtg, Calgary Alberta, 13–16 Aug 2006

56

R.G. Behling et al.

Lerner J, Brumberger V, Preac-Mursic V (1994) Severe diarrhea associated with Arcobacter butzleri. Eur J Clin Microbiol Infect Dis 13:660–662 Leuschner RGK, Baird F, Donald B, Cox L (2004) A medium for the presumptive detection of Enterobacter sakazakii in infant formula. Food Microbiol 21:527–533 Lovell R, Levi M, Francis J (1942) Studies on the survival of Johne’s bacilli. J Comp Path 54: 120–129 Lozano-Leon A, Barreiro-de Acosta M, Dominguez-Monuz JE (2006) Letter to the Editor, Absence of Mycobacterium avium subspecies paratuberculosis in Crohn’s disease patients. Inflamm Bowel Dis 12:1190–1191 Luber P, Brynestad S, Topsch D, Scherer K, Bartelt E (2006) Quantification of Campylobacter species cross-contamination during handling of contaminated fresh chicken parts in kitchens. Appl Environ Microbiol 72:66–70 Lui Y (2001) Electronic DNA arrays for determination of pathogenic Bacillus cereus. Thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy in the Department of Bioprocess Technology, School of Biotechnology, Royal Institute of Technology, Stockholm. Lynt RK, Kautler DA, Reach RB, Jr (1975) Botulism in commercially canned foods. J Food Prot 38:546–550 Mafart P, Couvert O, Lequerinel I (2001) Effect of pH on the heat resistance of spores – Comparison of two models. Int J Food Microbiol 63:51–56 Marinescu M, Collilgnon A, Squinazi F, Woodward D, Lior H (1996) Biotypes and serogroups of poultry strains of Arcobacter sp. isolated in France. In: Newall DG, Ketley JH (eds) Campylobacter VIII. Plenum Publishing, New York, pp. 519–520 Masaki H, Asoh N, Tao M, Ikeda H, Degawa S, Matsumoto K, Inokuchi K, Watanabe K, Watanabe H, Oishi K, Nagatake T (2001) Detection of gram-negative bacteria in patients and hospital environments at a room in geriatric wards under the infections control against MRSA. J Jpn Assoc Infec Dis 75:144–150 McClane BA (2000) The action, genetics, and synthesis of Clostridium perfringens enterotoxin. In: Cary JW, Linz JE, Bhatnagar D (eds) Microbial Foodborne Diseases. Technomic Publishing, Lancaster, PA, pp. 247–267 McClane BA (2001) Clostridium perfringens. In: Doyle MP, Beuchat LR, Montville TJ (eds) Food Microbiology: Fundamentals and Frontiers, 2nd edn. ASM Press, Washington, DC, pp. 351–372 McFadden JJ, Butcher PD, Chiodini RJ, et al (1987) Crohn’s disease-isolated mycobacteria are identical to Mycobacterium paratuberculosis, as determined by DNA probes that distinguish between mycobacterial species. J Clin Microbiol 25:796–801 McIngvale SC, Elhanafi D, Drake MA (2002) Optimization of reverse transcriptase PCR to detect viable shiga-toxin producing Escherichia coli. Appl Environ Microbiol 68(2):799–806 Mead PS, Griffin PM (1998) Escherichia coli O157:H7. Lancet 352:1207–1212 Mead PS, Slutsker L, Dietz V, McCraig LF, Bresse JS, Shapiro C, Griffin PM, Tauxe RV (1999) Food-related illness and death in the United States. Emerg Infect Dis Sept–Oct 5(5):607–625 Meng J, Feng P, Doyle MP (2001) Pathogenic Escherichia coli, Chapter 35. In: Downes FP Ito K (eds) Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC Millar D, Ford J, Sanderson J, Withey S, Tizard M, Doran T, Hermon-Taylor J (1996). IS900 PCR to detect Mycobacterium paratuberculosis in retail supplies of whole pasteurized cows’ milk in England and Wales. Appl Environ Microbiol 62:3446–3452 Millson M, Bokhout M, Carlson J, Spielberg L, Akdusm R, Borczyk A, Lior H (1991) An outbreak of Campylobacter jejuni gastro-enteritis linked to meltwater contamination of a municipal well. Can. J. Public Health 82:27–31 Ministry of Health New Zealand (2005) Enterobacter sakazakii. http://www.moh.govt.nz/ moh.nsf/0/526cb4a064b49a31cc257042007c4ef3?OpenDocument. Accessed 28 April 2009 Mishina D, Katsel P, Brown ST, Gilberts EC, Greenstein RJ (1996) On the etiology of Crohn’s disease. Proc Natl Acad Sci USA 93:9816–9820

2

Selected Pathogens of Concern to Industrial Food Processors

57

Molin N, Snygg BG (1967) Effect of lipid materials on heat resistance of bacterial spores. Appl Microbiol 15:1422–1426 Monroe PW, Tift WL (1979) Bacteremia associated with Enterobacter sakazakii (yellow, pigmented Enterobacter cloacae). J Clin Microbiol 10:850–851 Moore JE, Madden RH (2000) The effect of thermal stress on Campylobacter coli. J Appl Microbiol 89:862–899 Moreno GS, Griffiths PL, Connerton IF, Park RWA (1993) Occurrence of campylobacters in small domestic and laboratory animals. J App. Bacteriol 75:49–54 Morgan D, Gunneberg C, Gunnell D, Healing TD, Lamerton S, Soltanpoor N, Lewis DA, White DG (1994) An outbreak of Campylobacter infection associated with the consumption of unpasteurized milk at a large festival in England. Eur J Epidemiol 10:581–585 Mortimore S, Wallace C (1998) HACCP: A Practical Approach. Aspen Publishers, Inc. Gaithersburg, MD Mosso MDLA, de la Rosa MDC, Vivar C, Medina MDR (1994) Heterotrophic bacterial populations in the mineral waters of thermal springs in Spain. J Appl Bacteriol 77:370–381 Murphy C, Carroll C, Jordan KN (2006) Environmental survival mechanisms of the foodborne pathogen Campylobacter jejuni. J Appl Microbiol 100:623–632 Murrell WG, Scott WJ (1966) The heat resistance of bacterial spores at various water activities. J Gen Microbiol 43:411–425 Muytjens H (1985) Enterobacter sakazakii: Identification and clinical significance. Antonie van Leeuwenhoek 51:618 Muytjens HL, Kollee LAA (1990) Enterobacter sakazakii meningitis in neonates: Causative role of formula. Pediatric Infect Dis 9:372–373 Muytjens HL, Roelofs WH, Jaspar GHJ (1988) Quality of powdered substitutes for breast milk with regard to members of the family Enterobacteriaceae. J Clin Microbiol 26: 743–746 Muytjens HL, Zanen HC, Sonderkamp HJ, Kolee LA, Wachsmuth IK, Farmer JJ (1983) Analysis of eight cases of neonatal meningitis and sepsis due to Enterobacter sakazakii. J Clin Microbiol 18:115–120 Naser SA, Ghabrial G, Romero C, et al (2004) Culture of Mycobacterium avium subsp. paratuberculosis from the blood of patients with Crohn’s disease. Lancet 364:1039–1044 Naser SA, Schwartz D, Shafran I (2000) Isolation of Mycobacterium avium subsp. paratuberculosis from breast milk of Crohn’s disease patients. Am J Gastroenterol 95:1094–1095 Nazarowec-White M, Farber JM (1997a) Enterobacter sakazakii: A review. Int J Food Microbiol 34:103–113 Nazarowec-White M, Farber JM (1997b) Incidence, survival, and growth of Enterobacter sakazakii in infant formula. J Food Prot 60:226–230 Neelam M, Nawaz Z, Riazuddin S (1987) Hydrocarbon biodegradation biochemical characterization of bacteria isolated from local soils. Pakistan J Sci Ind Res 30:382–385 Noriega FR, Kotloff KL, Martin MA, Schwalbe RS (1990) Nosocomial bacteremia caused by Enterobacter sakazakii and Leuconostoc mesenteroides resulting from extrinsic contamination of infant formula. Pediatr Infect Dis 9:447–449 Norton DM, Braden CR (2007) Foodborne Listeriosis. In: Ryser ET, Marth EH (eds), Listeria, Listeriosis and Food Safety. CRC Press, Boca Raton, FL, pp. 305–356 Oh SW, Kang DH (2004) Fluorogenic selective and differential medium for isolation of Enterobacter sakazakii. Appl Environ Microbiol 70:5692–5694 Olson KE (2006) Quality control/industry perspectives: Symposium on Enterobacter sakazakii. Int Assoc Food Prot Ann Meeting, Calgary Alberta, 13–16 Aug 2006 Palumbo SA, Call JE, Schultz FJ, Williams AC (1995) Minimum and maximum temperatures for growth and verotoxin production by hemorrhagic strains of Escherichia coli. J Food Prot 58(4):352–356 Park SF (2002) The physiology of Campylobacter species and its relevance to their role as foodborne pathogens. Int J Food Microbiol 74:177–188

58

R.G. Behling et al.

Park CE, Sanders GW (1992) Occurrence of thermotolerant campylobacters in fresh vegetables sold at farmers’ outdoor markets and supermarkets. Can J Microbiol 38:313–316 Paton JC, Paton AW (1998) Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin Microbiol Rev 11:45–479 Peleg M, Cole MB (2000) Estimating the survival of Clostridium botulinum spores during heat treatments. J Food Prot 63:190–195 Plowman J, Peck MW (2002) Use of a novel method to characterize the response of spores of non-proteolytic Clostridium botulinum types B, E, and F to a wide range of germinants and conditions. J Appl Microbiol 92:681–694 Raghuberr EV, Matches JR (1990) Temperature range for growth of Escherichia coli serotype O157:H7 and selected coliforms in E. coli medium. J Clin Microbiol 28:803–805 Ray B (1996) Fundamental Food Microbiology. CRC Press, Boca Raton Reddy NR, Solomon HM, Fingerhut GA, Rhodehamel EJ, Balasubramaniam VM, Palaniappan S (1999) Inactivation of Clostridium botulinum type E spores by high pressure processing. J Food Saf 19:277–288 Reij MW, Den Aantrekker ED (2004) Recontamination as a source of pathogens in processed foods. Int J Food Microbiol 91:1–11 Reis AA, Zaza S, Langkop C, Tauxe RV, Blake PA (1990) A multistate outbreak of Salmonella chester linked to imported cantaloupe. Abstract. Interscience Conference on Antimicrobial Agents and Chemotherapy. p. 238. Restaino L, Frampton EW, Lionberg WC, Becker RJ (2006) A chromogenic plating for the isolation and identification of Enterobacter sakazakii from foods, food ingredients, and environmental sources. J of Food Prot 69:315–322 Rhodehammel EJ, Harmon SM (2001) Bacillus cereus, Chapter 14. In: Bacteriological Analytical Manual Online. http://www.cfsan.fda.gov/∼ebam/bam-14.html. Accessed 24 April 2009 Riordan T, Humphrey TJ, Fowles A (1993) A point source outbreak of Campylobacter infection related to bird-pecked milk. Epidemiol Infect 110:261–265 Ritchie JM, Thorpe CM, Rogers AB, Waldor. MK (2003) Critical roles for stx2, eae and tir in Enterohemorrhagic Escherichia coli-induced diarrhea and intestinal inflammation in infant rabbits. Infect Immun 71(12):1–20. http://iai.asm.org/cgi/content/full/71/12/7129. Robertson LF, Johannessen GS, Gjerde BK, Loncarevic S (2002) Microbiological analysis of seed sprouts in Norway. Int J Food Microbiol 75:119–126 Rodgers S, Peiris P, Casadel G (2003) Inhibition of nonproteolytic Clostridium botulinum with lactic acid bacteria and their bacteriocins at refrigeration temperatures. J Food Prot 66:674–678 Rossiter CA, Henning WR (2001) Isolation of Mycobacterium paratuberculosis (MAP) from thin market cows at slaughter. Abstract. Poult Sci Vol. 80: Suppl. 1. 54th Ann Rec Meat Conf, Vol II. p. 113. Ryser ET, Marth EH (eds) (1999) Listeria, Listeriosis and Food Safety, 2nd edn. Marcel Dekker, New York Sacks JJ, Lieb S, Baldy LM, Berta S, Patton CM, White MC, Bigler WJ, White JJ (1986) Epidemic campylobacteriosis associated with a community water supply. Am J Public Health 76:424–429 Sahin O, Morishita TY, Zhang Q (2002) Campylobacter colonization in poultry: Sources of infection and modes of transmission. Anim Health Res Rev 3:95–105 Schindler PR, Metz H (1990) Coliform bacteria in rinsed beer mugs – identification with the API 20E system and resistance behavior. Offentliche Gesundheitswesen 52:592–597 Schoeder-Tucker L, Wesley IV, Kiehlbauch JA, Larson DJ, Thomas L, Erickson GA (1996) Phenotypic and ribosomal RNA characterization of Arcobacter species isolated from porcine aborted fetuses. J Vet Diagn Invest 8:186–195 Schoeni JL, Wong ACL (2005) Bacillus cereus food poisoning and its toxins. J Food Prot 68(3):636–648 Setlow P, Johnson EA (1997) Spores and their significance. In: Doyle MP, Beuchat LR, Montville TJ (eds) Food Microbiology: Fundamentals and Frontier. ASM Press, Washington, DC, pp. 30–65

2

Selected Pathogens of Concern to Industrial Food Processors

59

Shapiro RL, Hatheway C, Swerdlow DL (1998) Botulism in the United States: A Clinical and Epidemiologic Review. Ann Intern Med 129:221–228 Sharma SK, Whiting RC (2005) Methods for detection of Clostridium botulinum toxin in foods. J Food Prot 68:1256–1263 Shreeve JE, Toszehgy M, Ridley A, Newell DG (2002) The carry-over of Campylobacter isolates between sequential poultry flocks. Avian Dis 46:378–385 Siegel LS (1993) Destruction of botulinum toxins in food and water. In: Hauschild AHW, Dodds KL (eds) Clostridium botulinum, Ecology and Control in Foods. Marcel Dekker, New York, pp. 323–341 Simmons BP, Gelfand MS, Haas M, Metts L, Ferguson J (1989) Enterobacter sakazakii infections in neonates associated with intrinsic contamination of a powdered infant formula. Infect Control Hosp Epidemiol 10:398–401 Simmons NA (2001) Detection of Mycobacterium paratuberculosis in retail milk in the United Kingdom: Analysis and perspectives. Talk presented at IAFP Symposium S14 – Mycobacterium paratuberculosis: Villain or Bystander? Minneapolis, MN, 5–8 Aug 2001 Skelly C, Weinstein P (2003) Pathogen survival trajectories: An eco-environmental approach to the modeling of human campylobacteriosis ecology. Environ Health Persp 111:19–28 Smeets LC, Voss A, Muytjens HL, et al. (1998) Genetische karakterisatie van Enterobacter sakazakii – isolaten van Nederlandse patiënten met neonatale meningitis (Genetic characterization of Enterobacter sakazakii isolated from Dutch patients with neonatal meningitis). Netherlands Tijdschrift voor Medische Microbiologie 6:113–115 Smith JL (1995) Arthritis, Guillain-Barré syndrome, and other sequelae of Campylobacter jejuni enteritis. J Food Prot 58:1153–1170 Soloman EB, Hoover DG (2004) Inactivation of Campylobacter jejuni by high hydrostatic pressure. Lett Appl Microbiol 38:505–509 Soriano JM, Rico H, Molto JC, Manes J (2001) Incidence of microbial flora in lettuce meat and Spanish potato omelette from restaurants. Food Microbiol 18:159–163 Speriando V, Schineller JB, Baldwin TO (1996) Quorum sensing controls the expression of the type III secretion gene transcription and protein secretion in enterohemorrhagic and enteropathogenic Escherichia coli. Proc Natl Acad Sci USA 96:15196–15201 Stabel JR (1998) Johne’s disease: A hidden threat. J Dairy Sci 81:283–288 Stable J, Pearce L, Chandler R, Hammer P, Kligin N, Cerf O, Collins MT, Heggum C, Murphy P (2001) Destruction by heat of Mycobacterium paratuberculosis in milk and milk products, Chapter 5. In: Mycobacterium paratuberculosis. Bulletin of the International Dairy Federation No. 362. IDF/FIL, Brussels, Belgium. Stern NJ (1992) Reservoirs for Campylobacter jejuni and approaches for intervention in poultry. In: Nachamkin I, Blaser MJ, Tompkins LS (eds) Campylobacter jejuni: Current Status and Future Trends. American Society for Microbiology, Washington, DC, pp. 49–60 Stern NJ, Bailey JS, Blankenship LC, Cox NA, McHan F (1988) Colonization characteristics of Campylobacter jejuni in chick ceca. Avian Dis 32:330–334 Stern NJ, Clavero MR, Bailey JS, Cox NA, Robach MC (1995) Campylobacter spp. in broilers on the farm and after transport. Poult Sci 74:937–941 Stern NJ, Line JE (2000) Campylobacter. In: Lund BM, Baird-Parker TC, Gould GW (eds) The Microbiological Safety and Quality of Food, vol II. Aspen Publishers, Gaithersburg, MD, pp. 1040–1056 Stern NJ, Svetoch EA, Eruslanov BV, Perelygin VV, Mitsevich EV, Mitsevich IP, Pokhilenko VD, Levchuk VP, Svetoch OE, Seal B (2006) Isolation of a Lactobacillus salivarius and purification of its bacteriocin that is inhibitory to Campylobacter jejuni in the chicken gastrointestinal system. Antimicrob Agents Chemother 50:3111–3116 Stoll B, Hansen N, Fanaroff A, Lemons A (2004) Enterobacter sakazakii is a rare cause of septicemia or meningitis in VLBW infants. J Pediatr 144:821–823 Sulliman SMA, Abubakr MI, Mirghani EF (1988) Microbial contamination of cutting fluids and associated hazards. Tribiol Int 30:737–757

60

R.G. Behling et al.

Sung N, Collins MT (1998) Thermal tolerance of Mycobacterium paratuberculosis. Appl Environ Microbiol 64:999–1005 Tamplin ML (2002) Growth of Escherichia coli O157:H7 in raw ground beef stored at 10ºC and the influence of competitive bacterial flora, strain variation, and fat level. J Food Protect 65: 1535–1540 Tamura A, Kato M, Omori M, Nanba A, Miyagawa K, Wang CR, Zhou WH (1995) Flavor components and microorganisms isolated from Suancha (sour tea, Takeutsu-sancha in Japanese). Nippon Kasei Gakkaishi 46:759–764 Tauxe RV (1992) Epidemiology of Campylobacter jejuni infections in the United States and other industrialized nations. In: Nachamkin I, Blaser MJ, Tompkins LS (eds) Campylobacter jejuni: Current Status and Future Trends. American Society for Microbiology, Washington, DC, pp. 9–19 Tompkin RB, Scott VN, Bernard DT, Sveum WH, Gombas KS (1999) Guidelines to prevent post-processing contamination from Listeria monocytogenes. Dairy, Food and Environ Sanit 19(8):551–562 Townsend S, Caubilla-Barron J, Loc-Carrillo C, Forsythe S (2006) The presence of endotoxin in powdered infant formula milk and the influence of endotoxin and Enterobacter sakazakii on bacterial translocation in the infant rat. Food Microbiol 24:67–74 Trachoo N, Frank JF (2002) Effectiveness of chemical sanitizers against Campylobacter jejunicontaining biofilms. J Food Prot 65:1117–1121 Trachoo N, Frank JF, Stern NJ (2002) Survival of Campylobacter jejuni in biofilms isolated from chicken houses. J Food Prot 65:1110–1116 Tuncer I, Ozsan K (1988) Biochemical typing of Enterobacter isolated from several clinical materials. Mikrobiyoloji Bul 22:105–112 Tuttle J, Gomez T, Doyle MP, Wells JG, Zhao T, Tauxe RV, Griffin PM (1999) Lessons from a large outbreak of Escherichia coli O157:H7 infections: Insights into the infectious dose and method of widespread contamination of hamburger patties. Epidemiol Infect 122:185–192 US CDC (21 May 2008) Campylobacter infections. http://www.cdc.gov/nczved/dfbmd/disease_ listing/campylobacter_gi.html. Accessed 23 April 2009 US FDA (9 April 2009) Bad bug book – Campylobacter jejuni, Available online at http://www.cfsan.fda.gov/∼mow/chap4.html. Accessed 23 April 2009 USDA (1999) Appendix B: Compliance Guidelines for Cooling Heat-Treated Meat and Poultry Products (Stabilization) http://www.fsis.usda.gov/OA/fr/95033F-b.htm. Accessed 28 April 2009 Urmenyi AMC, Franklin AW (1961) Neonatal death from pigmented coliform infection. Lancet 1:313–315 Van Acker J, De Smet F, Muyldermans G, Bougatef A, Naessens A, Lauwers S (2001) Outbreak of necrotizing enterocolitis associated with Enterobacter sakazakii in powdered milk formula. J Clin Microbiol 39:293–297 Van Kruiningen HJ, Chiodini RJ, Thayer WR, Coutu JA, Merkal RS, Runnels PL (1986) Experimental disease in infant goats induced by a Mycobacterium isolated from a patient with Crohn’s disease: A preliminary report. Dig Dis Sci 31(12):1351–1360 Van Kruiningen HJ, Ruiz B, Gumprecht LJT (1991) Experimental disease in young chickens induced by a Mycobacterium paratuberculosis isolate from a patient with Crohn’s disease. Can J Vet Res. Revue Canadienne De Recherche Veterinaire. So Can J Vet Res 55(2): 199–202 Van Os M, Van Wikeselaar PG, Spoelstra SF (1996) Formation of biogenic amines in well fermented grass silages. J Agri Sci Cambridge 127:97–107 Vandamme P, Flasen E, Rossau R, Hoste B, Segers P, Tygat R, Delay J (1991) Revision of Campylobacter, Helicobacter and Wolinella taxonomy: Emendation of generic descriptions and proposal for Arcobacter gen.nov. Int J Syst Bact 41:88–103 Vought KJ (1998) Salmonella enteritidis contamination of ice cream associated with a 1994 multistate outbreak. J Food Prot 61(1):5–10

2

Selected Pathogens of Concern to Industrial Food Processors

61

Wagenaar JA, Mevius DJ, Havelaar AH (2006) Campylobacter in primary animal production and control strategies to reduce the burden of human campylobacteriosis. Rev Sci Tech 25:581–594 Wang WL, Powers BW, Leutchefeld NW, Blaser MJ (1983) Effects of disinfectants on Campylobacter jejuni. Appl Environ Microbiol 45:1202–1205 Wayne LG, Salkin D (1956) The bacteriology of restricted tuberculosis pulmonary lesions. I. The effects of interval between reversal of infectiousness and subsequent surgery. Am Rev Tuberc Pulm Dis 74:376–387 Weir E (2002) Powdered infant formula and fatal infection with Enterobacter sakazakii. Can Med Assoc J 166:1570 Weissman JB, Dean RMAD, Williams M, Swanton N, Ali S (1977) An island-wide epidemic of salmonellosis in Trinidad traced to contaminated powdered milk. West Indies Med J 26: 135–143 Wells SJ, Wagner BA, Dargatz DA (1999) Factors associated with M. a. paratuberculosis infection in US Dairy herds. In: Manning E, Collins M (eds) Proceedings of the Sixth International Colloquium on Paratuberculosis. Melbourne, Australia, p. 62 Wesley IV (1996) Arcobacter and Helicobacter – risks for foods and beverages. J Food Prot 59:1127–1132 Wesley IV, Baetz AL (1999) Natural and experimental infections of Arcobacter in poultry. Poult Sci 78:536–545 White PL, Baker AR, James WO (1997) Strategies to control Salmonella and Campylobacter in raw poultry products. Rev Sci Tech 16:525–541 Willis J, Robinson JE (1988) Enterobacter sakazakii meningitis in neonates. Pediatr Infect Dis J 7:196–199 Wilson IG, Moore JE (1996) Presence of Salmonella spp. and Campylobacter spp. in shellfish. Epidemiol Infect 116:147–153 Wong AC (2001) Virulence Factors in B. cereus Endophthalmitis. FRI annual meeting, Madison, WI Yanmin H, Mangan JA, Dhillon J, Sole M, Mitchison DA, Butcher PD, Coates ARM (2000) Detection of mRNA transcripts and active transcription in persistent Mycobacterium tuberculosis induced by exposure to rifampin or pyrazinamide. J Bacteriol 182:6358–6365 Zink D (2003) FDA field survey of powdered formula manufacturing. FDA Food Advisory Committee. Contaminants and natural toxicants subcommittee meeting: Enterobacter sakazakii contamination in powdered infant formula. Washington, DC, 18–19 Mar 2003. See http://www.fda.gov/ohrms/dockets/AC/03/slides/3939s1_Zink.ppt. Accessed 5 May 2009 Zogaj X, Bokranz W, Nimtz M, Romling U (2003) Production of cellulose and curli fimbriae by members of the family Enterobacteriaceae isolated from the human gastrointestinal tract. Infect Immun 71:4151–4158

Chapter 3

Solving Microbial Spoilage Problems in Processed Foods Rocelle Clavero

Abstract This chapter surveys common microbial food spoilage processes. The chapter is organized by food products and includes sections addressing spoilage in meat, poultry, fish; dairy products (milk, butter, cheese); beverage products; bakery products; canned foods; fruit and confectionery products; and emulsions. It addresses the isolation and identification of spoilage organisms and provides several case studies as examples. It introduces various organisms responsible for spoilage including Gram-positive lactic acid bacteria, Gram-negative aerobic bacteria, yeasts, molds, and fungal contaminants. Throughout the chapter, attention is given to when, where, and how spoilage organisms enter the food processing chain. Troubleshooting techniques are suggested. The effect (or lack of effect) of heating, dehydration, pH change, cooling, and sealing on various organisms is explained throughout. The chapter contains four tables that connect specific organisms to various spoilage manifestations in a variety of food products.

3.1 Introduction Various preservation methods involving one or a combination of process technologies, such as dehydration, fermentation, salting, chemical disinfection treatments, canning, refrigeration, freezing, and irradiation have been employed in the manufacture of processed foods. These methods, however, are not generally designed to completely eliminate all microbes. Most processes are only applied to destroy pathogenic microorganisms and/or inhibit growth of common spoilage organisms. Loss of control in any one of these processes can render a product susceptible to microbial deterioration and spoilage resulting in loss of quality.

R. Clavero (B) Sara Lee Corporation, Downers Grove, IL, USA e-mail: [email protected]

J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_3, 

63

64

R. Clavero

Despite modern technological advances, spoilage of processed food commodities continues to occur. Food spoilage is a complex process, and foods are considered spoiled if organoleptic changes make them unacceptable to the consumer. It is widely recognized that food spoilage often occurs due to growth of microorganisms and activities of enzymes they secrete. Other non-microbial spoilage processes, e.g., light oxidation of colors, are beyond the scope of this chapter. Decomposition by-products and metabolites released during growth can result in the loss of a product’s organoleptic attributes – change in appearance (discoloration), the development of off-odors, slime formation, or any other characteristic that makes the food undesirable for consumption. The effective and efficient resolution to spoilage problems in the processed food industry requires a logical sequence that begins with the identification of the problem. Spoilage manifestations must be precisely defined and the investigation should involve collection of relevant information. These problems typically cannot be resolved by merely conducting off-site interviews and conference calls. A thorough investigation must include observations of in-plant practices and review of production, sanitation, and maintenance records. It is through the analyses of factual information and the sequence(s) of events leading to the occurrence of spoilage that will ultimately provide a clear evidence of causality. The microflora that develops during storage that cause spoilage can be predicted based on the knowledge of the origin of the food, the nature of the substrate, and the type of preservation factor such as temperature, atmosphere, aW , and pH used in the manufacture of the product (see Chapter 5 for a detailed factors influencing microbial growth, survival, and death). Ambient storage temperatures (60–80◦ F) will support growth of mesophilic microorganisms. Chilled storage, on the other hand, will restrict proliferation of mesophiles but will allow psychrotrophs to predominate. Metabolic activities, however, are substantially inhibited under frozen conditions. Similarly, aerobic conditions will promote growth of microorganisms that require oxygen for their metabolic functions. Limited amounts or complete absence of oxygen such as in vacuum packed and modified atmosphere packaged (MAP), however, will slow down or inhibit growth of aerobes but may allow growth of anaerobic microorganisms. Products that are dry with aW <0.60 will not support microbial growth. The availability of water in a product, however, will dictate whether mold, yeast, or bacteria will predominate. Bacteria are often the cause of spoilage in high moisturecontaining products while yeasts tend to cause loss in quality for products that contain high sugars or carbohydrates. Growth of most microorganisms is optimal at pH values between 5 and 7. Acidophiles, however, can tolerate pH as low as 1.0 and thus a primary cause of spoilage in high acid foods. Use of preservation factors and techniques, therefore, will allow selection for the types of microorganisms that can tolerate the environmental conditions in a given food product. Some microorganisms exhibit unique degradation products that can direct an investigator to an organism or type of microorganisms that produce the characteristic spoilage pattern. For instance, a sour taste or aroma generally occurs as a result of growth of acidophiles such as lactic acid bacteria and evidence of gas in such a product may further direct the investigation to heterofermenters. An accurate description

3

Solving Microbial Spoilage Problems in Processed Foods

65

of the manifestations of spoilage of an intact product, therefore, can provide crucial information that could lead to the identification of the causative spoilage agent and the source of contamination. Key culprits involved in microbial gas production are heterofermentative lactic acid bacteria, coliforms, yeast, Clostridium spp., and Bacillus spp. However, an understanding of the types of expected interaction of the organism with the food will be critical to reduce the list of possible suspects (see Chapter 5).

3.2 Spoilage of Processed Meat, Poultry, and Fish It is generally recognized that Gram-negative, motile and non-motile aerobic rods and coccobacilli belonging to the genera Pseudomonas, Moraxella, Psychrobacter, and Acinetobacter can cause spoilage of raw meat stored aerobically under refrigerated conditions (Stanbridge and Davies, 1998). Low temperatures will not restrict growth of these psychrotrophic microorganisms, but application of a heat treatment adequate to destroy vegetative cells of pathogenic bacteria is often sufficient to kill a wide variety of these Gram-negative bacteria. Their isolation from a spoiled thermally processed meat or poultry product would therefore indicate that post-heat treatment contamination occurred as a result of poor hygienic practices during one or more of the subsequent processing steps such as slicing, portioning, skinning, and packaging. Spoilage is often characterized by the generation of sulfur-containing products such as H2 S, dimethylsulfide, and methylmercaptan. Slime formation, greening, and production of a characteristic spoilage metabolite, dimethylsulfide, can be attributed to the growth of Pseudomonas spp. The production of H2 S, on the other hand, generally results from growth of members of the Gram-negative Enterobacteriaceae family (Garcia-Lopez et al., 1998). Gram-negative aerobic bacteria are unable to compete with the Gram-positive lactic acid bacteria (LAB) in vacuum or modified atmosphere packaged meat and poultry products held under refrigerated storage conditions. Mesophilic and psychrotrophic strains of Lactobacillus and Leuconostoc (including Weissella) are the most prominent spoilage bacteria of such products (Stiles and Holzapfel, 1997). Typical organoleptic changes include souring and the formation of gas, slime, and/or white liquid. Table 3.1 lists the primary causative agents of spoilage in ready-to-eat (RTE) meat and poultry products. Some lactic acid bacteria that may be present in raw meat, e.g., Weissella viridescens (also called, Lactobacillus viridescens), can survive heat processing, while others may be introduced as a result of post-process recontamination during slicing and packaging from equipment that has been inadequately cleaned and sanitized. Their ability to grow under certain pH, temperature, and aw conditions will determine which strains will eventually predominate to cause spoilage of the product (Hamasaki et al., 2003). The observed sliminess/gassiness and souring in two spoiled, vacuum-packaged pork loins were due to growth of Leuconostoc mesenteriodes and Lactococcus lactis, respectively, while Leocunostoc citreum caused the

66

R. Clavero

Table 3.1 Causative spoilage agents and manifestations of microbial deterioration in ready-to-eat (RTE) meat and poultry products Type of product

Spoilage manifestations

Causative agents

Dry, shelf-stable

Whiskers and spots Off-odor, green discoloration

Fermented, shelf-stable

Yeast growth

Molds Alteromonas putrefaciens Enterobacter liquefaciens Shewanella putrefaciens Debaryomyces hansenii Cryptococcus laurentii Debaryomyces vanriji Rhodotorula mucilaginosa LAB Pseudomonas spp., Lactobacillus (Weissella) viridescens Leuconostoc mesenteroides H2 S producing LAB, clostridia LAB Leuconostoc mesenteroides Homo- and heterofermentative LAB Heterofermentative LAB Enterococci, homofermentative LAB, Leuconostoc spp. Weissella viridescens Carnobacterium divergens Carnobacterium piscicola Pseudomonas spp. Clostridium laramie Clostridium ctm Clostridium estertheticum Clostridium butyricum Clostridium pasteurianum Bacillus coagulans

Sliminess Greening Cured, shelf-stable

Chilled, vacuum-packaged

Sliminess Blackening/discoloration Souring Ropiness Souring Loss of vacuum, gassing Sliminess

H2 S odor/gas production

Canned, acidified

Souring/H2 swell Medicinal/phenolic

appearance of white liquid in wiener sausages (Metaxopoulos et al., 2002). Growth of LAB that produce hydrogen peroxide, e.g., W. viridescens, caused a green discoloration in frankfurters and other cured, vacuum-packaged, thermally processed meat products. Other strains that have caused greening include Lactobacillus jensenii, Lactobacillus fructivorans, L. mesenteroides, Enterococcus faecalis, Enterococcus faecium as well as some Pediococcus strains (Anifantaki et al., 2002). Lactic acid bacteria utilize fermentable carbohydrates for growth and depending on the strain will either produce primarily lactic acid (referred as homofermenters) or will release CO2 concomitant with the production of lactic acid (referred to as heterofermenters). Evidence of bloating, bulging, or blown-up packaged of refrigerated cooked meat and poultry products therefore suggests growth of a gas-producing heterofermentative lactic acid bacterium. Growth of homofermenters, on the other

3

Solving Microbial Spoilage Problems in Processed Foods

67

hand, can alter the flavor of the product due to production of lactic acid, which can be manifested by a sour taste or smell. The resulting low pH, however, can have a beneficial effect in the manufacture of fermented meat products such as pepperoni and salami. The reduction in pH below 5.5 has been found effective in inhibiting multiplication of spoilage bacteria. This low pH, however, favors the growth of acid-tolerant yeasts and also promotes growth of molds (Dillon, 1998). Yeasts and molds are considered opportunistic spoilage organisms in processed meat and poultry products. They tend to flourish when processing, preservation, and storage conditions cause suppression of major spoilage bacteria. Spoilage due to growth of yeasts is often manifested by release of CO2 and loss in vacuum. Candida, Cryptococcus, Rhodotorula, and Trichosporon spp. are the most frequently isolated yeasts in processed meat products (Wolter et al., 2000). Spoilage yeasts could originate from the raw meat itself and survive the applied thermal process or are introduced into the product as a result of post-process contamination from equipment surfaces and poor personal hygiene. Off-flavors and slime formation can also develop from proteolytic and lipolytic activities resulting from the release of enzymes during growth of certain yeasts and molds. Due to its low water activity, spoilage of dry, shelf-stable meat products such as jerky occurs as a result of growth of fungal contaminants such as Cladosporium sp. that produce black spots on meat and Sporotrichum carnis that causes white spots. Aspergillus and Penicillium spp. are the most common molds isolated from fermented sausages and cured meats with low aw . Molds are often considered environmental contaminants in the processed meat and poultry industry. The investigation on the source of mold contamination, therefore, would likely focus on poorly maintained ventilation and air handing systems, false ceilings, floors and walls that may have become wet, poor quality of air used in packaging systems, and/or the primary packaging materials itself. Loss of temperature control in the cold chain could render refrigerated vacuumpackaged meat products susceptible to spoilage due to growth of Clostridium spp. Raw meat can contain low levels of clostridial spores that can survive the thermal process. Temperature abuse in the cold chain could allow clostridial spores to germinate and grow, and spoilage is manifested by the release of H2 S and gassiness can be observed in the packaged product (Holzapfel, 1998; Kalinowski and Tompkin, 1999). Spoilage of fish is primarily fueled by a significant proportion of soluble components such as sugars, amino acids, trimethylamine oxide, creatine, taurine, anserine, uric acid, betaine, carnosine, and histamine and involves Pseudomonas, Shewanella, Acinetobacter, Moraxella, Photobacterium, and Vibrio. The characteristic odor of spoiled fish comes from the production of volatile amines, in particular trimethylamine and sulfur compounds (Leuchner and Hammes, 1999). In cold smoked salmon, the predominant spoilage flora is lactic acid bacteria and some Gramnegative bacteria. The off-odors and flavors that develop have been described as sour, acid, pungent, and occasionally fecal (Stohr et al., 2001).

68

R. Clavero

3.3 Spoilage of Dairy Products Milk is an excellent growth medium for bacteria. It provides the nutrients and moisture and has a near neutral pH. Microbial populations that survive the pasteurization process hence can potentially cause spoilage of milk and milk products. Table 3.2 lists the causative agents and the manifestation of spoilage in various milk products.

Table 3.2 Causative spoilage agents and manifestations of microbial defects in dairy products Type of product

Microbial defects

Causative spoilage agents

Milk, pasteurized

Fruity/fermented Bitter Acid/sour; tart Malty (“grape nuts”); sour Putrid; rancid Swelling, gas

Pseudomonas spp.

Milk, canned

Yogurt

Blown packages; off-flavor Yeasty

Cream Butter

Foamy Surface taint Malty (“grape nuts”); sour Gassy, butyric acid Gassy, floating, split curd

Cheese

Soft cheese

Cottage cheese

Moldy Browning defects Gassy; flavor defects

Slimy curd, putrid Discoloration Slimy gelatinous Fruity

Lactococcus lactis Lc. lactis var. maltigenes Clostridium sporogenes Clostridium thermosaccharolyticum Kluyveromyces marxianus Hansenula spp. Saccharomyces cerevisiae Pichia membranaefaciens Candida guillermondii Geotrichum candidum Candida, Torulopsis Pseudomonas spp., Alteromonas Lc. lactis Clostridium tyrobutyricum Leuconostoc, Lc. lactis subsp. Diacetylactis Penicillium, Mucor, other molds Yarrowia lipolytica Torulaspora delbrueckii Candida paralopsis; C. sake Cryptococcus spp. Debaryomyces hansenii Kluyveromyces marxianus Pichia fermentans P. guillermondii P. membranaefaciens P. norvegensis Rhodotorula spp. Yarrowia lipolytica Pseudomonas Flavobacterium, yeasts, molds Pseudomonas, Alcaligenes, Flavobacterium, coliforms Yeasts

3

Solving Microbial Spoilage Problems in Processed Foods

69

Gram-positive bacteria – primarily Micrococcus, Corynebacterium, Mycobacterium, Lactobacillus, Lactococcus, and Enterococcus spp. and Gram-negative bacteria – Pseudomonas, Achromobacter, Enterobacter, Escherichia, and Flavobacterium can be found in milk immediately after milking (Bordova et al., 2002). Complete inactivation of Gram-negative bacteria, however, can be achieved by exposure to pasteurization temperatures (72◦ C; 15 min). Spoilage of pasteurized milk due to the presence of viable Gram-negative bacteria, therefore, will only likely occur if there is a breach in the sanitary condition of processes downstream of the heat exchanger. Most problems arise in the filling lines as a result of poor sanitation and lack of preventative maintenance. Failure to establish a schedule for disassembly and inspection of heat exchangers can also lead to cross-contamination resulting from cracked plates or damaged gaskets. Isolation of Gram-negative bacteria in pasteurized milk is evidence of post-process contamination. Growth of enteric bacteria is manifested by off-odors, putrid smell, and slight rancidity. Gram-positive, thermotolerant Streptococcus, Corynebacterium, and Bacillus spp. and thermoduric micrococci and lactococci can survive the thermal process, and the shelf-life of the milk will largely depend on the levels present in the incoming raw milk. The coryneforms, micrococci, and lactococci are usually incapable of growth if pasteurized milk is held below 6◦ C. Certain Bacillus spp., however, possess the ability to grow under refrigerated conditions. While the vegetative cells of the bacilli are readily destroyed by pasteurization, the spores are more heat resistant. Given the optimum conditions for growth, Bacillus spp. have great potential to cause spoilage. Many bacilli can degrade milk protein and secrete the enzyme phospholipase that can cause destabilization of the fat emulsion in milk. Spoilage is often manifested by curdling and souring of milk, which signals the end of a product’s shelf-life. An overwhelming proportion of the spoilage psychrotrophic flora in milk produces heat-stable extracellular degradative enzymes that can survive pasteurization. Milk can contain as many as 60 indigenous enzymes which have the potential to affect cheese quality. Psychrotrophic strains can contribute to loss in product quality by producing heat-resistant lipases and proteases even though many of these strains may be destroyed by milk pasteurization. Consequently the microbiological quality of raw milk will affect the quality of the finished product. Acid, malty, fruity, bitter, stale, putrid, and unclean off-flavors in dairy products have been linked to bacterial contamination. Various chemicals associated with off-flavors are produced as metabolites during the bacterium’s exponential growth phase. Some characteristic off-flavor notes in milk are associated with contamination of milk with specific psychrotrophic bacteria. Contamination by Pseudomonas fragi often causes “fruity” off-notes in milk. P. fragi’s lipase and esterase enzymes hydrolyze short-chain fatty acids in milk fat, converting the acids to ethyl esters by reaction with ethanol (Rajmohan and Dodd, 2002). Some Bacillus strains have been observed to produce fruity off-notes in milk. Another common psychrotroph associated with a flavor defect is Lc. lactis var. maltigenes. Production of 3-methylbutanal causes the malty aroma of milk cultures. Other common metabolites include ethyl

70

R. Clavero

acetate, dimethyl sulfide, ethanol and other alcohols, methyl ketones, C4 –C10 fatty acids, lactic acid, pyruvic acid, and bitter peptides. Incidents of early gas production in cheese with added starter cultures can be ascribed to the activity of citrate-utilizing and/or heterofermentative strains in mixed-strain cultures (Muir and Banks, 2003). Commercial mixed-strain cultures may contain low levels of heterofermentative gas-producing, non-starter lactic acid bacteria (NSLAB), capable of growth at 6◦ C and at high salt in moisture (S/M) levels. Heterofermentative lactic acid bacteria have been isolated from raw milk and tanker samples of pasteurized skim. It is probable that the gas-producing NSLAB in cheese result from tanker skim-milk, mixed-strain cultures, and/or the processing facility environment previously seeded with raw milk microbiota. Small numbers of gas-producing non-starter lactic acid bacteria (NSLAB) can survive pasteurization. Hence, spoilage due to NSLAB resulting in early gas defects in cheese could be directly correlated to raw milk quality or a tanker skim-milk quality in some instances. Gas production in cheese of normal chemical composition and pH can also result from a number of interacting factors, including the starter used, lactose and citrate levels in the curd, temperatures of curd/cheese during pressing and curing, the salt sensitivity of the starter, the salt/moisture level in the cheese, the levels of gas-producing NSLAB bacteria in the cheese, and the level of phageinduced cell lysis in the curd at pressing and during early cheese maturation. Refrigerated cheeses can also be spoiled by fungal contaminants such as Penicillium that produces blue-green spores. The principal microbial spoilage defects in butter are “surface taint” and hydrolytic rancidity both of which can occur from contamination with Pseudomonas spp. (Kornacki et al., 2001). Such contamination can occur from growth of Pseudomonas spp. in raw milk or cream or from surface growth of the organism on butter resulting in production of heat-resistant lipases and proteases. Surface growth of mold species such as Rhizopus, Geotrichum, Penicillium, and Cladosporium can also produce hydrolytic rancidity when they grow on butter surfaces. Less commonly butter (a water-in-oil emulsion) can develop a malty flavor, skunk-like aroma, and black discoloration due to Lc. lactis var. maltigenes, P. mephitica, and P. nigrifacines, respectively (Kornacki et al., 2001). Poorly cooled cream may result in overproduction of acid from Lc. lactis cultures resulting in an overly acidic flavor in butter. Bacterial degradation also results from bacteria that get into such cream upon contact with improperly washed or sanitized equipment, and from external contamination, and is made worse by improper cooling. An extensive review of the microbiology associated with butter manufacture can be found in Kornacki et al. (2001) and a more detailed discussion of spoilage in food emulsions follows in this chapter. Sorbates are often added to inhibit growth of spoilage organisms. Nevertheless, deterioration of cottage cheese has occurred due to growth of psychrotrophic Gram-negative bacteria such as Pseudomonas, Alcaligenes, Proteus, Aerobacter, or Aeromonas resulting in undesirable flavors and slimy curd. Growth of yeasts and molds (e.g., Geotrichum, Penicillium, Mucor, and Alternaria) can cause flavor, textural, and visual spoilage.

3

Solving Microbial Spoilage Problems in Processed Foods

71

Non-starter Lactobacillus, Pediococcus, and Leuconostoc spp. have been isolated in cheddar cheese. Their proteolytic activity can contribute to development of desirable flavors but can also result in accumulation of bitter peptides. Heterofermentative varieties of NSLAB have also caused cracks and split defects due to gas formation. Furthermore, some NSLAB produce a racemase capable of converting L(+)-lactate produced by homofermentative starter cultures to the less soluble D(–)-lactate form resulting in development of white haze due to calcium lactate crystallization (Chou et al., 2003), which can be mistaken for mold growth. The source of NSLAB is primarily from post-pasteurization contamination through contact with equipment surfaces used for food handling and processing or from recalcitrant bacterial biofilms that developed due to inadequate cleaning and sanitation practices.

3.4 Spoilage of Beverage Products The high water activity of most ready-to-drink beverages typically allows microbial growth, and spoilage of beverage products has occurred (Table 3.3). The addition of fermentable sugars, acid, and chemical preservatives many of which are also acids, however, prevents growth of most organisms except those that can tolerate acidic conditions. Spoilage yeasts such as Saccharomyces cerevisiae, Candida lipolytica, and Zygosaccharomyces bailii are able to overcome these hurdles (Battey et al., 2002). Z. bailii, Zygosaccharomyces rouxii, and Zygosaccharomyces lentus have been reported to develop resistance to the common preservatives, sodium benzoate and sorbates (Steels et al., 1999). The fermentative nature of these yeasts allows the production of fruity flavors and aroma, organic acids, and gas causing containers to explode. The bacteria Acinetobacter calcoaceticus and Gluconobacter oxydans have also caused spoilage in beverages. Gluconobacter causes off-flavors while Acinetobacter has no perceivable effect on the odor, taste, or appearance of beverages. The latter, however, can alter the pH and preservative levels thereby allowing other microorganisms to grow and spoil the product. Contamination of pasteurized beverages during bottling with lactic acid bacteria, particularly Lactobacillus perolens, has been reported. Growth of lactic acid bacteria can cause release of a strong diacetyl aroma and flavor. Due to the low pH, soft drinks constitute a hostile environment and spoilage is thus caused by a limited number of yeasts, molds, and acid-tolerant bacteria. Spoilage defects include cloudiness, particulates, taints, and excessive gas. Improper sanitation practices particularly in filling machine-associated equipment including the filler itself, proportioning tanks, and pumps (especially those with worn or broken seals), have generally contributed to spoilage incidents. Contamination can also occur via raw materials, returned bottles, airborne vectors, and insects (Stratford et al., 2002). Condensation in compressed air used to blow out bottles prior to filling has been a source of yeast contamination to final containers (Kornacki, 2009, personal communication).

72

R. Clavero

Table 3.3 Causative spoilage agents and manifestations of microbial defects in processed beverages Type of product Alcoholic Wine Beer Non-alcoholic, carbonated Soda

Fruit drinks Non-carbonated, pasteurized Tea Water

Fruit juices

Microbial defects

Causative spoilage agents

Haze Rope Diacetyl; buttery flavor

Acetic acid bacteria, yeasts Pediococci spp. Lactobacillus spp.

Acid, off-flavors

Acinetobacter, Gluconobacter, fermentative yeasts, lactic acid bacteria Leuconostoc mesenteroides

Diacetyl, acetoin off-flavor

Haze, cloudy, slimy sediment Slimy particulate, haze, moldy Haze, sediment, gassiness, Diacetyl

Medicinal Dairy-based

Souring, gassiness

Canned, shelf-stable Fruit drinks

High acid

Pseudomonas, coliforms Pseudomonas, Enterobacteriaceae, molds, yeasts Oenococcus oenos, molds, Zygosacchromyces spp., Candida spp., Saccharomyces spp. Alicyclobacillus spp., L. perolens Thermophilic Bacillus spp., heterofermentative LAB Clostridium pasteurianum

3.5 Spoilage of Bakery Products The predominant spoilage problem in baked products is influenced by interrelated factors, specifically storage temperature, relative humidity, level of preservatives, pH, packaging material, and gaseous environment surrounding the product, and most importantly, by the moisture content and aw of the product. For low-moisture baked products (aw < 0.6), microbiological spoilage is not a problem. In intermediate moisture products (aw 0.65–0.85), osmophilic/osmotolerant yeasts and xerophilic molds are the predominant spoilage microorganism of concern. In highmoisture products (aw 0.94–0.99), many bacteria, yeasts, and molds are capable of growth (see Chapter 5 for aw growth limitations of selected pathogens). Many fillings can support the growth of pathogenic bacteria especially if they contain egg or dairy products. Yeast occur mainly in intermediate and high-moisture bakery products and spoilage is manifested by visible growth of yeasts on the surface of the products (white or pink patches) or fermentative spoilage manifested by alcoholic, fruity, or

3

Solving Microbial Spoilage Problems in Processed Foods

73

other odors and/or visible evidence of gas production, such as gas bubbles in jam and fondants (e.g., in pastry products) or expansion of flexible packaging. Visible yeast growth is generally associated with products of high aw and short shelf-life, while fermentative spoilage is usually associated with low aw and long shelf-life products. Yeasts which cause surface spoilage of bread are mainly Pichia burtonii (“chalk mold”) and to a lesser extent Candida guillermondii, Hansenula anomala, and Debaryomyces hansenii. The most common osmotolerant yeast that causes spoilage of high sugar coatings and fillings such as jam, marzipan, and mincemeat is Z. rouxii. Contamination of products by osmophilic yeasts results from unclean utensils and equipment and inadequate cleaning of syrup holding tanks. Many molds are capable of growing at aw values of <0.8 but a few xerophilic molds, although rare, have caused spoilage in products with aw as low as 0.65. The most important molds isolated from bakery products, however, are Eurotium, Cladosporium, and the highly xerotolerant species of Aspergillus and Penicillium (Table 3.4). Freshly baked products are free of viable vegetative molds and mold spores; products, however, become contaminated as a result of post-baking contamination by mold spores from the air, bakery surfaces and equipment, food handlers, and raw ingredients such as glazes, nuts, spices, and sugars. Molds are especially troublesome in summer months due to airborne contamination and the warmer, more humid storage conditions. Furthermore, products may be wrapped prior to being completely cooled. This results in formation of condensation inside the package and on the product’s surface – conditions that are conducive to mold growth. The most common mold spores found on cake were Wallemia sebi, Penicillium spp., Cladosporium spp., Eurotium glaucus group, and other aspergilli. Penicillium spp. particularly P. notatum, P. expansum, and P. viridicatum were the predominant spoilage molds in products with a high ERH >86%. E. glaucus group and Eurotium amstelodami predominate at values below this level. Table 3.4 Causative spoilage agents and manifestationsof microbial defects in ready-to-eat/heat and serve bakery products Type of product

Microbial defects

Causative spoilage agents

Bread

Moldy

Fruit-filled pastries

Gassing; souring

Cheesecake

Splits; gassiness

Aspergillus, Penicillium, Eurotium Zygosaccharomyces, Saccharomyces, Leuconostoc mesenteroides, Bacillus spp. Saccharomyces cerevisiae

Ropey bread is a bacterial spoilage condition of bread caused primarily by Bacillus subtilis and occasionally by B. licheniformis, B. pumilus, and B. cereus (Thompson et al., 1998). The spoilage is initially detected by a sweet fruity odor that resembles over-ripe melons or pineapples. This is followed by enzymatic degradation resulting in discoloration and the crumb eventually becomes soft and sticky due to production of extracellular polysaccharides. The major source of contamination

74

R. Clavero

is from raw ingredients. Bakers therefore need to identify those loaf formulations and processes with greater probability of producing rope-associated spoilage and establish raw material specifications that allow only use of ingredients with lowest levels of acceptable contamination.

3.6 Spoilage of Canned Foods Thermally processed canned foods that have a low acidity and are properly processed can still spoil if incubated at temperatures greater than 37◦ C. The thermophilic organism Bacillus stearothermophilus typically survives the canning thermal process and acidifies the canned food if incubated at higher temperatures. Improper processing of canned foods under anaerobic conditions favors the growth of gas producers such as Clostridium sporogenes. Improperly processed foods under aerobic conditions favor the growth of Bacillus coagulans that increases acidity without gas production resulting in “flat sour” spoilage defect (Palop et al., 1997). Improperly processed high acid foods such as tomatoes can be spoiled by the mold Byssochlamys fulva which causes softening of the fruit. A more extensive review of thermal processing of canned foods including and description of spoilage conditions associated with canned foods can be found in David et al. (1996).

3.7 Spoilage of Fruit and Confectionery Products Jams, jellies, and preserve products are characterized by low water activity and a low pH. Since they are also packed at high temperatures, thus they are not likely to harbor harmful foodborne illness organisms. However, mold spoilage may develop if the cap applied is not heated to a temperature adequate to destroy mold spores. This destruction of organisms on the cap is usually accomplished by inversion of the jar or by the use of a steam capper. Home canners often pour hot wax on the surface to exclude air from the surface of the product. Xerophilic fungi such as Bettsia alvei, Chrysosporium xerophilum, Neosartorya glabra, and Chrysosporium farinicola have been isolated from various spoiled manufactured chocolate products. Spoilage by Chrysosporium spp. may be mistaken for a “fat bloom” but the white fungal growth can be observed under an electron microscope (Kinderlerer, 1997).

3.8 Emulsions Emulsions are surfactant stabilized two-phase systems composed of two or more immiscible liquids and thus are inherently unstable systems. Growth of microorganisms can result in spoilage that is indicated by a rapid separation of continuous and the disperse phases. The physical breakdown of emulsions is a consequence of (1) pH alteration, usually during acid production which may affect surfactant solubility,

3

Solving Microbial Spoilage Problems in Processed Foods

75

(2) microbial utilization of an emulsion component essential for stability (e.g., surfactant or thickening agent), or (3) production of extracellular spoilage factors. Microorganisms produce a variety of extracellular molecules, which have the capacity to destabilize emulsions. Spoilage can result from direct recontamination of the finished product or contamination of the aqueous components prior to emulsification, both followed by growth of the microorganism. The main causes of spoilage of sauces and dressings are yeasts and lactobacilli; mold spoilage rarely occurs. Spoilage due to yeasts primarily Z. bailii, Pichia membranaefaciens, Z. rouxii, S. cerevisiae, and Candida magnolia is characterized by gas formation or growth of light brown surface colonies. Manifestation of spoilage due to Lactobacillus plantarum, Lactobacillus buchneri, and L. fructivorans include gas production and a change in product pH.

3.9 Isolation and Identification of Spoilage Organisms The recovery and identification of the causative agent of spoilage either in the food or on food contact surfaces form an integral part of the investigation process. The basis of methods used for the testing of microorganisms in foods is well established and typically relies on the collection of representative sample of the spoiled product (see Chapter 8) and incorporating into a growth medium on which the microorganisms can develop colonies thus resulting in a visual indication of growth. Recovery of spoilage organisms usually involves use of selective and non-selective depending on the type of organism. The Compendium of Methods for the Microbiological Examination of Foods provides methods for a wide variety of microbes, including spoilage organisms, in a wide variety of foods (Downes and Ito, 2001). If conventional test methods fail to recover the organism from an obviously spoiled product, a specialized media may have to be developed based on the nature of the product. Production of metabolic degradation by-products and chemical analysis such as pH could also provide clues to the identity of a group of organisms that exhibit similar properties. Specific compounds can also be used as spoilage markers such as dimethyl sulfide for chicken and eggs, diacetyl for orange juice, and trimethylamine for fish and milk. Mass spectroscopy and gas chromatography have been used to identify these substances. There has been considerable research into rapid and automated microbiological methods but the identification of isolates continues to rely on conventional pure culture techniques in the isolation of spoilage organisms (Betts, 2000). Traditionally, once an organism has been isolated from a food product, identification methods involving biochemical or immunological analyses of purified organisms are used. Major advances in molecular biology have made it possible to identify organisms by reference to their DNA structure. Comparison of the gene sequences using the PCR and DNA sequencing technology showed that the 16S rRNA gene is highly conserved and hence can be used as the new “gold standard.” Use of this technique will allow identification of the isolate to the species level that is not achievable using other techniques. (For a discussion of the value of molecular subtyping for methods for the food industry see Chapter 10.)

76

R. Clavero

3.10 Case Studies The manufacturer of vacuum-packaged frankfurters experienced a spoilage problem with the development of a sticky, tenacious slime. Some of the packages at retail outlets developed slime during storage but all packages had extremely high aerobic plate counts. Review of sanitation records did not reveal any unusual event that could result in recontamination of the finished product. The problem appeared to be unrelated to changes in hygienic measures, to the microbiological quality of meat trimmings or spices, or to the heat process applied. Review of production records, however, revealed that slime formation was associated with frankfurters produced for the Passover season. These frankfurters contained sucrose, whereas at other times of the year the formulation used dextrose. The cause of spoilage was later confirmed when both slimy and non-slimy frankfurters were plated on two separate media, one containing sucrose and the other containing dextrose. When both slimy and non-slimy frankfurters were plated into a laboratory medium containing sucrose as a carbohydrate source, all colonies that developed on the plate were mucoid and slimy. When both franks were plated onto the dextrose-containing medium, the colonies that developed were moist but not slimy. Storage of a freshly made sucrose-containing product confirmed the likely cause of the development of the sticky tenacious slime. Another case involved a perishable canned ham patty that developed soft swells during refrigerated storage. The aerobic plate counts on both normal and swollen cans were found negligible, however, counts on APT medium (a medium especially adapted for recovery of fastidious microbes such as lactic acid bacteria) were extremely high in swollen cans but not in unswollen cans. Preliminary findings suggest that the most likely cause of spoilage appeared to be microorganisms surviving the heat process of 64–65◦ C. Thermal death time studies of isolates from APT agar indicated that the organisms were destroyed at 67–68◦ C. The spoilage organism was determined to be Lactobacillus viridescens and the primary cause was related to the use of rework that allowed the organism to grow to high populations under refrigerated storage conditions. A manufacturer of a variety of salad dressings experienced a spoilage problem involving gassy fermentation. The defect was observed within weeks after manufacture of all varieties of salad dressings. Analyses of the spoiled product revealed the presence of large numbers of yeasts and lactobacilli. There was no evidence that any raw ingredient contributed to the high LAB and yeast count nor was there any change in formulation breakdown in clean-up practices. A review of production records, however, revealed that all spoiled products were manufactured in Line 1. These preliminary findings lead to the conclusion that some source of product contamination was associated with Line 1 and not Line 2. Surface swabs collected from pipe interiors, valves, faces of pumps, and filler heads taken before start-up and during operations did not reveal the source of contamination. Disassembly of the pumps leading to the filler heads, however, revealed that the gasket behind the face of the pump had deteriorated and a large accumulation of various types of salad dressing undergoing fermentation. High yeast and lactobacilli counts were

3

Solving Microbial Spoilage Problems in Processed Foods

77

obtained from the swabs collected. Further experiments revealed that the formulation is capable of preventing growth of small numbers of the spoilage organisms from the incoming ingredients. Of great significance, however, is the fact that the spoilage organisms became adapted to the acidic environment and the high numbers present significantly reduced the shelf-life of various salad dressings.

References Anifantaki K, Metaxopoulos J, Kammenou M, Drosinos EH, Vlassi M (2002) The effect of smoking, packaging and storage temperature on the bacterial greening of frankfurters caused by Leuconostoc mesenteroides subsp. Mesenteroides. Ital J Food Sci 14(2):135–144 Battey AS, Duffy S, Schaffner D (2002) Modeling yeast spoilage in cold-filled ready-to-drink beverages with Saccharomyces cerevisiaie, Zygosaccharomyces bailii, and Candida lipolytica. Appl Environ Microbiol 68(4):1901–1906 Betts RP (2000) Conventional and rapid analytical microbiology. In: Stringer M, Davis C (eds) Chilled Foods: A Comprehensive Guide, 2nd edn. CRC Press, Boca Raton, FL Bordova O, Baranova M, Laukova A, Rozanska A, Rola J (2002) Hygiene of pasteurized milk depending on psychrotrophic organisms. Bull Vet Inst Pulawy 46:325–329 Chou YE, Edwards CG, Luedecke LO, Bates MP, Clark S (2003) Nonstarter lactic acid bacteria and aging temperature affect calcium lactate crystallization in cheddar cheese. J Dairy Sci 86: 2516–2524 David JRD, Graves RH, Carlson VR (1996) Aseptic Processing and Packaging of Food: A Food Industry Perspective. CRC Press, New York Downes FP, Ito K (eds) (2001) Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC Dillon V (1998) Yeasts and moulds associated with meat and meat products. In: Davies A, Board R (eds) The Microbiology of Meat and Poultry, 1st edn. Blackie Academic and Professional, London Garcia-Lopez ML, Prieto M, Otero A (1998) The physiological attributes of Gram negative bacteria associated with spoilage of meat and meat products. In: Davies A, Board R (eds) The Microbiology of Meat and Poultry, 1st edn. Blackie Academic and Professional, London Hamasaki Y, Ayaki M, Fuchu H, Sugiyama M, Morita H (2003) Behavior of psychrotrophic lactic acid bacteria isolated from spoiling cooked meat products. Appl Environ Microbiol 69: 3668–3671 Holzapfel W (1998) The Gram-positive bacteria associated with meat and meat products. In: Davies A, Board R (eds) The Microbiology of Meat and Poultry, 1st edn. Blackie Academic and Professional, London Kalinowski RM, Tompkin B (1999) Psychrotrophic clostridia causing spoilage in cooked meat and poultry products. J Food Prot 62:766–772 Kinderlerer JL (1997) Chrysosporium species, potential spoilage organisms of chocolate. J Appl Microbiol 83:771–778 Kornacki JL, Bradley RL, Flowers RS (2001) Microbiology of butter and related products, Chapter 5. In: Marth EH, Steele JL (eds) Applied Dairy Microbiology, 2nd edn. Marcel Dekker, New York Leuchner R, Hammes WP (1999) Formation of biogenic amine in mayonnaise, herring and tuna fish salad by lactobacilli. Intl J Food Sci Nutr 50:159–164 Metaxopoulos J, Mataragas M, Drosinos EH (2002) Microbial interaction in cooked cured meat products under vacuum or modified atmosphere at 4◦ C. J Appl Microbiol 93:363–373 Muir DD, Banks JM (2003) Factors affecting the shelf-life of milk and milk products. In: Smit G (ed) Dairy Processing: Improving Quality, 1st edn. CRC Press, Boca Raton, FL

78

R. Clavero

Palop, A., Marco, A., Raso, J., Sala, F, and Condon, S. 1997. Survival of heated Bacillus coagulans spores in a medium acidified with lactic or citric acid. Int J Food Microbiol, 38:25–30. Rajmohan S, Dodd CER (2002) Enzymes from isolates of Pseudomonas fluorescens involved in food spoilage. J Appl Microbiol 93:205–213 Stanbridge LH, Davies AR (1998) The microbiology of chill-stored meat. In: Davies A, Board R (eds) The Microbiology of Meat and Poultry, 1st edn. Blackie Academic and Professional, London Steels H, James SA, Roberts IN, Stratford M (1999) Zygosaccharomyces lentus: A significant new osmophilic preservative resistant spoilage yeast, capable of growth at low temperature. J Appl Microbiol 87:520–527 Stiles ME, Holzapfel WH (1997) Lactic acid bacteria of foods and their current taxonomy. Int J Food Microbiol 36:1–29 Stratford M, Bond CJ, James SA, Roberts IN, Steels H (2002) Candida davenportii sp. Nov., a potential soft drink spoilage yeast silated from a wasp. Int J Sys Evol Microbiol 52:1369–1375 Thompson, J.M., Waites, W.M. and Dodd, C.E.R. 1998. Detection of rope spoilage in bread caused by Bacillus species. J Appl Microbiol 85:481–486. Wolter H, Laing E, Viljoen B (2000) Isolation and identification of yeasts associated with intermediate moisture meats. Food Technol Biotechnol 38:69–75

Chapter 4

Where These Contaminants Are Found Jeffrey L. Kornacki

Abstract The potential for in-factory environmental contamination exists for any food not biocidally treated in its end-use container. Microbes enter the factory environment from a variety of sources including worker’s skin, garments, air, and ingredients, among others. Air, water, tools, workers, traffic, and other means transfer microbes in the non-sterile factory environment into niches that are inaccessible for cleaning and sanitation. Within these niches many bacteria can attach themselves to underlying surfaces using cell wall-bound structures given enough time. Bacteria that attach and are allowed to form biofilms can be protected from cleaners and sanitizers. This chapter contains many examples of operating practices and structures that may create growth niches or transmit microbes in the factory environment. Over 30 photographs illustrate these practices.

4.1 The Significance of Environmental Contamination The potential for in-factory environmental contamination exists for any food not biocidally treated in its end-use container. It is the author’s contention that it is unreasonable, if not impossible to expect factory environments to be maintained in a sterile condition. Microbes enter the factory environment from a variety of sources including worker’s skin, garments, air, and ingredients. Logically the relative risk of food contamination from an in-factory growth niche containing 1 billion microbes is 10,000,000 times greater than that of the same area with 100 microbes. This illustrates the importance of controlling the growth of microbes in the factory environment. However, the relative risk of contamination from the environment is dependent on a number of key factors. Factors that affect the growth of microorganisms include moisture, nutrients, pH, oxidation–reduction potential, temperature, J.L. Kornacki (B) Kornacki Microbiology Solutions, Inc., McFarland, WI, USA e-mail: [email protected]

J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_4, 

79

80

J.L. Kornacki

presence or absence of inhibitors, interactions between microorganisms in a population, and time. Chapters 5 and 7 deal with these growth-affecting factors in more depth. However, moisture is the most critical of these as it is absolutely required for microbial growth (Gabis and Faust, 1988) and the most readily controlled. Air, water, tools, workers, traffic, and other means transfer microbes in the non-sterile factory environment into niches that are inaccessible for cleaning and sanitation. Within these niches many bacteria can attach themselves to underlying surfaces using cell wall-bound structures (e.g., proteins, polysaccharides, glycoproteins), given enough time. Bacteria that attach and are allowed to form biofilms can be protected from cleaners and sanitizers (DeBeer et al., 1994; Dhir and Dodd, 1995). Attached cells have also been correlated with increased heat resistance (Dhir and Dodd, 1995). Greater relative risk to product contamination occurs when entrapped environmentally resistant biofilms become moistened which may revive injured cells and allow microbial growth.

4.1.1 Causes of Microbial Growth Niche Development Microbial growth niches result from three things: (1) unsanitary operating practices, including some sanitation practices, (2) unsanitary maintenance and repair practices, and (3) unsanitary design of the factory or its equipment. Examples of these abound (see Figs. 4.1–4.35). Examples of operating practices that create or facilitate migration of microbial growth niches may include the following:

– Use of high-pressure hoses in drains or on wet floors causing aerosolization of (usually) high populations of bacteria into the air. – Wet cleaning in the same area with exposed product. – Failure to maintain sanitizer concentrations in germicidal footbaths. – Allowing for fluid and residues to build up underneath germicidal footbaths. – Failure to change gloves after they have become torn or have contacted the floor or other unsanitary areas. It is also important to note that failure to use hot enough water to liquefy fatty residues that entrap and protect microorganisms on food contact surfaces during cleaning will prevent aqueous sanitizer from contacting the cells. – Failure to clean and sanitize certain areas due to unproven assumptions regarding the perceived lack of the ability of an organism in a product residue to move past a certain point (e.g., gasket, valve – see Figs. 4.1, 4.8, 4.9, and 4.19). – Unrestricted forklift and wheeled vehicle traffic from raw to finished areas of the factory. – Failure to clean, sanitize, and thoroughly dry behind back-plates of positive displacement pumps.

4

Where These Contaminants Are Found

81

– Allowing accumulations of dust which may entrap moisture on or near equipment (see Fig. 4.6). – Storage of pallets in standing water (see Fig. 4.10). – Storage of packaging material on the floor (see Fig. 4.10). – Use of rotating pipe brushes which aerosolize wet residues. Maintenance/repair practices that create or facilitate migration of microbial growth niches include the following: – Failure to replace torn hoses allowing for creation of niches between the inner and outer hose skins (see Fig. 4.14). – Use of duct tape to repair leaking pipes allowing bacterial populations to build up and then drip into the factory. – Cleaning and sanitizing dirty equipment in the same area as cleaned sanitized equipment is stored. – Not hanging hoses to dry when not in use (see Figs. 4.10 and 4.14). – Failure to replace/repair torn insulation on storage tanks, freezers, refrigerant pipes, etc. (see Fig. 4.16). – Unsanitary welds that are not smooth and contiguous allowing for collection of residue. – Failure to effectively seal roof penetrations. – Failure to clean and sanitize tools when working on food processing equipment. – Failure of maintenance workers to use proper attire (hair nets, foot coverings, smocks, beard nets, etc.) in finished product areas of the plant. – Failure to repair or replace equipment that liberates excessive dust that may entrap moisture and create microbial growth niches. – Failure to correct peeling paint and rust (see Figs. 4.15 and 4.17). – Failure to repair holes in walls and ceilings (see Fig. 4.21). – Installation of false ceilings over wet processing areas (see Figs. 4.15 and 4.20). – Failure to install or maintain point of use filters on compressed air lines. – Improper use of impact wrenches to tighten fittings which deform or tear gaskets allowing product to leak and microbial growth to occur in the gasket/flange area (Fig. 4.19). – Failure to repair broken gasket seals on product pumps. – Failure to maintain condensate catch pan drains of air handling units in an unclogged state. Designs of the factory and/or equipment that create or facilitate migration of microbial growth niches include the following: – Allowing the creation of standing water in the factory through failure to tap catch pans, sinks, lubrication, etc., water to drains (see Figs. 4.24 and 4.25). – Design of food processing machinery that cannot be readily disassembled for cleaning and sanitation.

82

J.L. Kornacki

– Installation of ball–valves that allow product to collect between the inner ball and the valve housing (see Fig. 4.32). – Common loading dock for raw ingredients and finished product. – Common electrical charging rooms (e.g., for forklift or other motorized vehicles) for equipment used on both the raw and finished product areas of the factory. – Common elevators that move between raw and finished product floors (see Fig. 4.27). – Common receiving and load out docks for raw and finished liquid product tankers. – Restrooms located adjacent to processing areas. – Air handling units (AHUs) in which the final filter precedes the cooling coils.

Cooling coils and catch pans provide selective environments for the growth of psychrotrophic bacteria. Such bacteria may include pathogens like Listeria monocytogenes or Yersinia enterocolitica and/or spoilage bacteria such as Pseudomonas spp. and have been shown to be source of high populations of gram-negative bacteria even when well maintained (Hugenholtz and Fuerst, 1992). Designs in which final microbiological filters are placed after the coils (in the direction of air flow) should reduce the level of airborne contamination from AHU cooling coils and/or catch pans. Air flow ≥4 M/s will begin to aerosolize bacteria even in undisturbed standing water (e.g. in catch pans for HVAC units) aerosolized (Mitscherlich and Marth, 1984) from quiescent pools of water as may exist in AHU catch pans with clogged drains.

4.1.2 In-Factory Risk Assessments and Zones The author has found it useful to break down observations of operating, maintenance, and repair and equipment/factory design-related issues into high, medium, and indirect risk of product contamination. An example of high risk would be where moisture, which has been allowed to sit, is brought into contact with the product stream. Indirect risk maybe standing water on the floor of a factory where the consequences of its presence are unknown. This water, if disturbed, may or may not result in high, medium, or another indirect risk. Medium risk would be the same as high risk except for some subsequent step in the process or environment which may reduce the risk by some means. Medium risks nearly usually require further investigation, e.g., through challenge studies, process temperature verification, etc. Some have found it useful to divide the factory environment into zones of proximity to the product with zone 1 being the closest and zone 4 being the most removed from the product stream (ICMSF, 2002). This concept can be useful, but has been misapplied in some instances. One misconception is that if all the contaminationrelated issues in the earlier zones (e.g., zones 3 and 4) are cleared up, then this will resolve issues in zones 1 and 2 (ICMSF, 2002). However, once a biofilm or growth

4

Where These Contaminants Are Found

83

niche is established in zones 1 and 2, nothing short of decontaminating the site will correct it. That is not to say that controlling zones 3 and 4 will not help in the general control of zones 1 and 2.

4.2 The Significance of Ingredient Contamination The number of ways microbes may evade even well-developed HACCP plans is manifold. Introduction of contaminated ingredients into the factory environment can result in product contamination even when validated microbicidal CCPs are employed downstream in the process. This is because such ingredients may contaminate the factory environment and result in post-process contamination through the means described above.

4.3 The Significance of Process Failure Events Critical control points (CCPs) by definition provide process control, which if not within defined critical limits (CLs) could result in an unsafe product. Implicit in HACCP, is appropriate validation of CCPs. Oftentimes acceptable validation of CLs is based on laboratory inoculation studies that model the process; other times they are based on published research. Unacceptable so-called validations are based on tradition (e.g., stemming from a philosophy of “that’s the way we have always done it”) or extrapolation of appropriate CLs from one process/product combination to another very different process/product combination where they are inappropriate. Extrapolating the validated microbicidal time and temperature for high temperature short time (HTST) milk pasteurization (71.7◦ C for a minimum of 15 s) to milk chocolate would be an example of an unacceptable extrapolation between processes and food matrices. The heat resistance of microbes in milk chocolate will be markedly greater than that in fluid milk, in part, because of the difference in water activity between the two matrices (see Chapter 5). Some processes do not lend themselves to modeling in a laboratory due to their highly variable and dynamic conditions. In these situations validation of the process lethality should be done. Use of harmless surrogate microorganisms has been proposed (Ma et al., 2007; Eblen et al., 2005) as have time–temperature indicators. Recently the USDA has been advocating the use of surrogates for these purposes (Englejohn, 2004). The United States Food and Drug Administration’s Center for Food Safety and Applied Nutrition (USFDA CFSAN) has published guidelines on the selection of surrogates (Anonymous, 2000).

84

J.L. Kornacki

4.4 Examples of Operating Practices That May Create Growth Niches or Transmit Microbes in the Factory Environment

Fig. 4.1 Failure to disassemble pumps and gaskets adequately for cleaning (water marks/residue on finished product star valve gasket and flange (left) and on bottom of product tank (right), both previously not disassembled during cleaning)

Fig. 4.2 Inappropriate use of broom to move standing water

4

Where These Contaminants Are Found

85

Fig. 4.3 Uncleaned areas underside of floor-associated materials. These should be cleaned, sanitized, and thoroughly dried before use

Fig. 4.4 Floor scrubber (can transmit aerosols and wet residues)

Fig. 4.5 Failure to maintain segregation of work boots from street shoes

86

Fig. 4.6 Failure to control dust accumulation

Fig. 4.7 Failure to inspect and clean air line boxes

J.L. Kornacki

4

Where These Contaminants Are Found

Fig. 4.8 Failure to adequately disassemble, clean, and inspect behind diaphragm of pump

Fig. 4.9 Failure to adequately disassemble horizontal agitator shaft coupling for cleaning

87

88

J.L. Kornacki

Fig. 4.10 Storage of pallets, boxes, and packaging material on wet floor (movement of wheeled vehicles will track, splash, and aerosolize contaminated wet residues from floor)

Fig. 4.11 Failure to clean underside of product conveyor and thoroughly dry afterward

4

Where These Contaminants Are Found

Fig. 4.12 Storage of cleaned and dried equipment near wet floor

89

90

J.L. Kornacki

4.5 Examples of Maintenance/Repair Practices That May Create Growth Niches or Transmit Microbes in the Factory Environment

Fig. 4.13 Failure to replace torn or leaking gaskets

4

Where These Contaminants Are Found

Fig. 4.14 Failure to replace torn hoses

Fig. 4.15 Wet ceiling tiles

91

92

J.L. Kornacki

Fig. 4.16 Torn insulation (may collect moisture resulting in microbial growth niche development– note mold growth inside)

Fig. 4.17 Rusted opened back motor fan cover (failure to cover these during wet cleaning will also result in accumulation of wet residue inside fan which will be aerosolized when the motor turns on)

4

Where These Contaminants Are Found

93

Fig. 4.18 Poorly sealed roof (may allow for contaminated wet residues to drip into factory environment)

Fig. 4.19 Deformed gasket (from improper assembly with impact wrenches) with entrapped product residue

94

Fig. 4.20 Cealing leaks, bowed water-soaked ceiling tile

Fig. 4.21 Figure unsealed, pen-sized wall penetration

J.L. Kornacki

4

Where These Contaminants Are Found

Fig. 4.22 Penetrated hollow structures

Fig. 4.23 Failure to maintain seal on electrical box cover

95

96

J.L. Kornacki

4.6 Examples of Equipment or Factory Design That May Create Growth Niches or Transmit Microbes

Fig. 4.24 Untapped (e.g., to drain) pump coolant water (contributes water to the factory environment)

Fig. 4.25 Other examples of water lines not tapped to drains

4

Where These Contaminants Are Found

97

Fig. 4.26 Sandwiched areas in gears under product conveyor can collect product residues

Fig. 4.27 Tracked residues and standing water inside elevators (potential for cross-contamination throughout the plant via foot/wheeled vehicle traffic, e.g., when personnel and equipment are moved from raw to finished product floors)

98

J.L. Kornacki

Fig. 4.28 Poorly plumbed gurgling drain spraying droplets into factory environment (arrows depict movement of some of the droplets)

Fig. 4.29 Large immovable support bases for equipment (entraps moisture difficult to clean)

4

Where These Contaminants Are Found

99

Fig. 4.30 Sandwiched areas on cleaning equipment (can entrap residues and result in microbial growth niche development)

Fig. 4.31 Compressed air traps. These should be maintained. They are excellent places to sample as large volumes of air pass through them. Condensate development within compressed air lines may also grow microbes indicating the need for point of use filters where appropriate

100

J.L. Kornacki

Fig. 4.32 Use of ball–valves (entrapment of product between ball and valve housing)

Fig. 4.33 Equipment designed with numerous areas to entrap product residue (spiral freezer)

4

Where These Contaminants Are Found

101

Fig. 4.34 Steam exhaust over floor by trench drain (opportunity for splashing and aerosolization of contaminated residues from and near drain)

Fig. 4.35 Use of adjacent dissimilar flooring (may result in cracks that entrap wet residues)

References DeBeer D, Srinivasan R, Stewart PS (1994) Direct measurement of chlorine penetration into biofilms during disinfection. Appl Environ Microbiol 6:4339–4344 Dhir VK, Dodd CER (1995) Susceptibility of suspended and surface-attached Salmonella to biocides and elevated temperatures. Appl Environ Microbiol 61:1731–1738 Eblen DR, Annous BA, Sapers GM (2005) Studies to select appropriate nonpathogenic surrogate Escherichia coli strains for potential use in place of Escherichia coli O157:H7 and Salmonella in pilot plant studies. J Food Prot 68(2):282–291 Engeljohn D (2004) Regulatory Perspective of Validation and Verification Activities. Presented in Symposium S03, Validation and verification of pathogen interactions in meat and poultry processing, International Association for Food Protection Annual Meeting, Phoenix, 8–11 August 2004 Gabis DA, Faust RE (1988) Controlling microbial growth in the food-processing environment. Food Technol 42(12):81–82, 89

102

J.L. Kornacki

Hugenholtz P, Fuerst JA (1992) Heterotrophic bacteria in an air-handling system. Appl Environ Microbiol 58(12):3914–3920 International Commission for the Microbiological Specifications of Foods (ICMSF) (2002) Microorganisms in Foods 7: Microbiological Testing in Food Safety Management. Kluwer Academic Plenum Publishers, New York Ma L, Kornacki JL, Zhang G, Lin CM, Doyle MP (2007) Development of thermal surrogate microorganisms in ground beef for in-plant critical control point validation studies. J Food Prot 70(4):952–957 Mitscherlich E, Marth EH (1984) Units and Commentaries on Behavior of Bacteria. In: Microbial Survival in the Environment: Bacteria and Rickettsiae Important in Human and Animal Health. Springer-Verlag, NY, pp. 725–728 USFDA Center for Food Safety and Applied Nutrition (2000) Kinetics of Microbial Inactivation for Alternative Food Processing Technologies–Overarching Principles: Kinetics and Pathogens of Concern for All Technologies. June 2.http://www.cfsan.fda.gov/∼comm/ift-over.html. Accessed 17 July 2008

Chapter 5

What Factors Are Required for Microbes to Grow, Survive, and Die? Jeffrey L. Kornacki

Our lives are inextricably woven with the lives of these creatures who we ignore until they cause us trouble – Lynn Margulis and Dorion Sagin, Microcosmos, 1986

Abstract This chapter focuses on the impact of extrinsic and intrinsic factors that impact the growth of bacteria and fungi in foods. A bacterium with a generation time of 20 min can grow from 1 cell to over a million in 7 h. Intrinsic factors that impact microbial growth or survival are those properties within the food itself. Examples of such factors are the amount of available (not chemically bound) water (i.e., water activity), the oxidation/reduction potential (ORP) of the food, its pH, and the type of acid present. Extrinsic factors are those applied to the food such as thermal processes and refrigeration. Sometimes extrinsic factors such as heating result in creation of intrinsic factors such as a reduced ORP. The dynamic interaction between intrinsic and extrinsic factors will have a profound effect on the type of microbiota in the ingredient, food, and factory environment. The extrinsic and intrinsic factors that impact microbial survival and growth in food or in factory niches are manifold and can be quite dynamic. This highlights the need for research to better understand the relationship of microbes to their environments. Food processors should exercise appropriate caution (e.g., via challenge studies, appropriate testing, selection, and monitoring of valid CCPs) when formulating new products. Assumptions about microbial behavior in one product may not necessarily apply to another.

5.1 Introduction There are five significant groups of microorganisms that are of significance in foods; these include bacteria, fungi, Rickettsia, parasites, and viruses. However, the reader will note that the book is focused upon only bacteria and fungi (e.g., yeasts and J.L. Kornacki (B) Kornacki Microbiology Solutions, Inc., McFarland, WI, USA e-mail: [email protected] J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_5, 

103

104

J.L. Kornacki

molds). This is because these microbes have the greatest potential for growth in the food processing environment as they typically do not require a host for growth, and the methods for their detection are fairly well known and available to the food industry. Hence this chapter, like the others, will focus on the impact of extrinsic and intrinsic factors that impact the growth of bacteria and fungi in foods. It is important to note that bacteria, though very small (Fig. 5.1) relative to man (measured in millionths of a meter (microns or “μ”)) are still much larger than viruses yet smaller than yeasts or molds.

Fig. 5.1 Microbial size relationships

Bacteria are able to grow in a characteristic manner characterized by a “lag” phase, where they do not multiply but adapt to a new environment followed by transition to a “log” phase, where they grow exponentially (Fig. 5.2). A bacterium with a generation time of 20 min can grow from 1 cell to over a million in 7 h. This factor illustrates the importance of rigorous microbiological control efforts in the factory environment, especially since these environments are not sterile. Subsequent to the log phase they transition into a stationary phase where the population remains constant for a time at a maximum level. This is followed by a decline phase. A number of factors influence all these phases of growth or death. Populations in various phases of growth have altered resistance to stress as compared to those in other phases of growth (Fig. 5.3). Other factors may be employed by the industry to lengthen the lag phase, reduce the growth rate, or speed the destruction of microorganisms. This chapter is dedicated to providing a general overview of these factors. It is useful, when assessing the potential for microbial risk of contamination of the finished product from ingredients, the environment and the potential of finished product to support growth or survival, that an understanding of the basics that impact

5

What Factors Are Required for Microbes to Grow, Survive, and Die?

105

Fig. 5.2 Generalized microbial population growth (Banwart, 1979)

Fig. 5.3 Influence of Staphylococcus aureus culture age on its heat resistance (D-value) in autoclaved skim milk at 58◦ C (Adapted from Kornacki, 1986)

106

J.L. Kornacki

microbial growth and survival be understood. These factors can be broken down into intrinsic and extrinsic factors associated with the food or the environment. Intrinsic factors that impact microbial growth or survival are those properties within the food itself. Examples of such factors are the amount of biologically available (not chemically bound) water (i.e., water activity), the oxidation/reduction potential (ORP) of the food, its pH, and the type of acid present. Extrinsic factors are those applied to the food such as thermal processes and refrigeration. Sometimes extrinsic factors such as heating result in creation of intrinsic factors such as a reduced ORP. The dynamic interaction between intrinsic and extrinsic factors will have a profound effect on the type of microbiota in the ingredient, food, and factory environment.

5.2 Intrinsic Factors Affecting the Growth and Survival of Microbes 5.2.1 Water Activity (aw ) Bacteria and fungi have an absolute requirement for moisture, below which they cannot grow. The moisture available to the microbe is referred to as water activity and abbreviated “aw .” It is better to rely upon this parameter when considering the interaction of a microbe and moisture within a food product than simply that product’s percent moisture content. This is because some high-moisture foods may have a relatively low aw . Examples of these include syrups, jam, jellies, and pie fillings (see Table 5.1). Much of the moisture is chemically bound by the carbohydrates within the food in this example and thus is not available for cellular metabolism. Table 5.1 shows minimum reported aw values for selected microorganisms as well as some typical aw values for some foods. Microbial growth is retarded when the aw decreases, however, with regard to thermal processing, bacteria are more resistant to “dry” as opposed to “wet” heat. The specific solute(s) used to reduce the water activity also play a role in thermal protection (Kornacki and Marth, 1986, 1989, 1992, 1993). It is a common misconception that microbes will die in foods whose minimum water activity is below the minimum reported for their growth. Indeed some death may occur in these conditions, but it should be remembered that freeze drying is an import means of preserving a microbial population.

5.2.2 Acidity The pH of a food can also exert a profound effect on the ability of a microbe to survive or grow. Generally speaking, pH values near neutral (e.g., 6.5–7.5) are optimal for the growth of foodborne microbes of contemporary interest.

5

What Factors Are Required for Microbes to Grow, Survive, and Die?

107

Table 5.1 Limiting aw ’s of selected microbes compared to typical food aw ’s aw

Selected foods

0.98–1.00 0.97 0.96 0.95–1.00 0.93–0.96 0.92–0.95 0.90–0.98 0.92 0.90–0.94 0.9 0.88 0.84–0.92 0.83–0.87 0.82–0.94 0.80–0.90 0.79–0.84 0.8 0.75–0.91 0.69 0.65–0.75 0.61 0.60–0.75 0.54–0.75 0.2 0.10–0.20

Fresh poultry or fish Some ripened cheeses Fresh meats

Maple syrup

Microbe Clostridium botulinum type E Escherichia coli Salmonella Bacillus cereus C. botulinum Listeria Lactobacillus Most spoilage bacteria Most spoilage yeasts Staphylococcus aureus

Fermented sausages Jelly Aspergillus flavus Fruit juice concentrates Most spoilage molds Jams Chocolate candy Some cereals Xerophilic molds/osmophilic yeasts Sugars, syrups Honey Dried whole milk Some cereals

Adapted from Jay (2002), Banwart (1979), Ryser (1999)

Decreases in pH by addition of acids or fermentation have the impact of retarding and even killing certain microorganisms. Different microbes have different tolerances to pH. Yogurt which is produced by the fermentation of milk by selected lactic acid bacteria produces an environment very hostile to the growth and survival of many food pathogens (Frank and Marth, 1977). Table 5.2 depicts selected foods and the reported limiting pH for the growth of selected microorganisms. Consequently the pH of a food or a niche in a factory environment and the available moisture and oxidation–reduction potential will have a profound effect on the type of microbes that can survive and grow in such environments. 5.2.2.1 Inhibitory Substances Sometimes inhibitory substances (e.g., organic acid preservatives) are directly added to a food to retard microbial growth. Adjusting the pH of a food to a given level with one organic acid-based preservative will not provide the same germicidal power as another. It is important to recognize that the germicidal form of most organic acids is the undissociated acid. This makes sense when one realizes that bacterial

108

J.L. Kornacki Table 5.2 Limiting pHs of selected microbes compared to some typical food pHs pH

Selected food

7 6.6–6.8 6.5 6.3–6.5 6.1–6.4 6.2–6.4 5.9–6.1

Crabs Very fresh fish (most species) Cream Milk Butter Chicken Ham Vegetables Aspergillus Cabbage (green) Onion (red) Potatoes (tubers and sweet) Pumpkin

5.7–6.1 5.4–6.0 5.3–5.8 5.3–5.6 4.8–5.2 4.8–5.0 4.8 4.5 4.4–4.7 4.3–4.4 4.2–4.3 4.0–5.0 3.6–4.3 3.0–4.4 3 2.9–3.3 2.8–4.6 1.5–3.5

Microbe

Clostridium botulinum Vibrio parahemolyticus Eggplant Staphylococcus aureus Escherichia coli Tomatoes (whole) Salmonella Fruits Orange (juice) Lactobacillus spp. (most) Grapefruit (juice) Apples Plumbs Yeasts/molds

Adapted from Jay (2000), Banwart (1979)

cell membranes are charged and therefore the passage of charged ions into the cell would not occur apart from active transport. However, the undissociated acid portion should more easily pass through the cell membrane, dissolve in the cytosol of the bacterium, dissociate, and cause damage within the cell. This author well remembers a processor who changed the acid content of a product but kept the pH the same as the original product. Blown bottles due to excessive gas production by heterofermentative lactobacilli dramatically illustrated the danger of simply relying upon pH independent of the type of acid used.

5.2.3 Oxygen and Oxidation/Reduction Potential Just as every food has a minimal, maximal, and optimal pH and water activity, so they also have requirements in relation to oxygen and the oxidation/reduction potential (ORP) of their food or environment.

5

What Factors Are Required for Microbes to Grow, Survive, and Die?

109

ORP is the ability of a substance to donate or receive electrons. The ORP of a particular food is dependent on its natural ORP, the poising capacity of a food (resistance to ORP change), the oxygen tension in the atmosphere, and the access of the atmosphere to the food. Some microbes, like Pseudomonas spp. and related organisms, are obligate aerobes and thus have an absolute requirement for oxygen and high ORP in the food or environment for growth. In general, aerobic bacteria require an ORP of +200 to + 800 mV (Jay, 2000). Oxygen is toxic to other microbes, like Clostridium botulinum, considered to be a strict anaerobe, and thus they are intolerant of high oxidation–reduction potentials (ORP) also known as Eh . Reduced conditions for the growth of anaerobes have Eh values are around –200 mV. Obligate anaerobes may be lacking in certain enzymes (e.g., catalase) involved in aerobic respiration which detoxify otherwise toxic metabolites. Catalase converts toxic H2 O2 produced during oxidative metabolism to non-toxic O2 and H2 O. Hence the presence of oxygen will result in the strict anaerobic bacterial cell poisoning itself with hydrogen peroxide. Other microbes that we refer to as facultative anaerobes, such as Salmonella, will grow better with oxygen but can also grow under anaerobic conditions. Still other microbes like lactobacilli or streptococci are microaerobic and therefore grow in slightly reduced condition with some but limited oxygen. An illustration of the relationship of microbial growth to ORP in a medium is depicted in Fig. 5.4 where one can expect strict aerobes to grow at the top of a test tube of freshly prepared fluid thioglycollate medium (FTG). FTG medium has a small amount of agar and when freshly will have a dissolved oxygen and Eh continuum from low to high as one moves from the bottom of the tube to the top where air can interface with the top of the medium. Thus one can expect anaerobes to grow

Fig. 5.4 The relationship of microbial growth to oxygen

110

J.L. Kornacki

at the bottom, facultative anaerobes to grow throughout, and microaerobes to grow in a restricted zone intermediate between top and bottom. It is also important to note that, in general, the Eh of a growth medium is reduced when bacteria multiply (Banwart, 1979).

5.2.4 Competition Competition for nutrients and the production of toxic metabolites can result in retardation of one microbial type by another. Bacteria often produce a number of substances that inhibit other microbes. These may include bacteriocins, organic acids, toxic metabolites, antibiotics. They may also cause changes in the food which in turn inhibit other types of bacteria. For example, lipolytic bacteria break down fat into free fatty acids which may inhibit other microbes.

5.3 Extrinsic Factors Affecting the Growth and Survival of Microbes 5.3.1 Temperature Microbial growth may occur between –10 and 90◦ C. Temperatures within this range affect the lag phase, growth rate, maximal cell density, nutrition, and physiology of a microbial population (Kornacki and Gabis, 1990). Microorganisms have differences in the optimal, minimal, and maximal temperatures at which they may grow. Those microbes important in foods are either psychrotrophs (able to grow at 7◦ C or less but with optimal growth between 25 and 30◦ C) or mesophiles (have moderate optimal growth temperatures between 30 and 45◦ C), or thermophiles with optimal growth temperatures between 55 and 80◦ C. Given these definitions, it is possible to have mesophilic organisms like non-proteolytic C. botulinum, Yersinia enterocolitica, and Listeria monocytogenes, which are commonly thought to be mesophiles but are also considered psychrotrophs (Kornacki and Gabis, 1990). There are also thermoduric bacteria which can withstand elevated temperatures but have mesophilic optimal growth temperatures, such as the Enterococci which are able to grow at 50◦ C.

5.3.2 Impact of Temperature on Microbial Death It is well known that heating will destroy microorganisms. However, microbes have varying degrees of heat resistance which is dependent on the microbe and the matrix in which it is heated. In general, heat resistance is greatest in gram-positive spores > gram-positive vegetative cells > gram-negative cells.

5

What Factors Are Required for Microbes to Grow, Survive, and Die?

111

Fig. 5.5 Examples of bacterial survival at lethal temperatures (Adapted from Kornacki, 1986)

In theory bacteria die in a log-linear fashion when subjected to lethal temperatures; however, in practice this is not always the case (see Fig. 5.5). The time at a given temperature to reduce a bacterial population 10-fold or 90% is called a D-value, and it corresponds to the negative reciprocal of the slope of the survivor curve. Given that these plots are not always log-linear reductions, this creates some difficulties in proper calculation of the D-value. The time to destroy an entire population of bacteria is referred to as a thermal death time (TDT). By plotting the log10 TDT vs. temperature, one can see how the resistance of a bacterium changes at various temperatures. The parameter used to determine the heat resistance of a bacterium over such a range of temperatures is called a z-value and corresponds to the negative inverse of the slope of this plot. In theory, the same slope and therefore the same z-value would be obtained if the log10 D-value is also plotted against temperature. This plot is called a “phantom TDT curve” and is usually generated instead of the TDT curve. If one knows the heat penetration into the coldest portion of the product, the bacterium of concern, its z-value, and a D-value at any temperature, then in principle one can determine the lethality of a thermal process to destroy the target microbial population. One can see that this process has some flaws because of the microbiological “shoulder” (e.g., curve “A”) and “tail” (e.g., curve “C”) portion of the curves above. Frank and Chmielewski (2004) provided an alternate approach based on a probabilistic determination of total destruction of L. monocytogenes biofilms. This approach eliminates the confusion associated with calculating thermal death resulting from survivor curves that have “shoulders,” resistant “tails,” or that are concave or convex. Canning processes are based on a 12D concept to destroy maximal populations of C. botulinum. The time to total destruction of a microbial population at a given temperature is referred to as an F-value (in this case F = 12D for C. botulinum). The

112

J.L. Kornacki

industry generally uses many times “F ” to ensure the destruction of this organism in canned products (Lund, 1975).

5.3.3 Other Extrinsic Factors Other processes have been applied to foods to deactivate microorganisms including high pressure treatments, pulsed electric fields, pulsed light, X-ray, ultraviolet light, and drying.

5.4 Interactions Between Intrinsic and Extrinsic Factors The intrinsic properties of foods will also have an effect upon the heat resistances of a bacterium. Foods with low water activities (e.g., chocolate or milk powder) will thermally protect microbes to a greater degree than foods with higher water activity. Hence a pasteurization treatment of 71.7◦ C for 15 min could be relied upon to destroy many log10 cycles of Salmonella in fluid milk (with an aw approaching 100), but would be ineffective in destroying Salmonella in milk chocolate (D70◦ C = 12–17.5 h; D80◦ c = 1.6–2.4 h; Mitscherlich and Marth, 1984), which has an aw much closer to “0.” Lower molecular weight solutes will have a larger impact upon the aw of a product than larger macromolecules and, therefore, play a key role in microbial heat resistance of a product. It would be convenient if one could accurately predict microbial heat resistance based solely upon aw (all other extrinsic and intrinsic factors being equal); however, the individual solutes within a food matrix may also have differing effects on bacterial heat resistance at the same aw . Thus each food matrix and food processing environment provides a unique microbial environment which prohibits extrapolation to other quite different matrices, such as in our example of fluid milk and milk chocolate, above. Microbes themselves will change the intrinsic properties of food through increases (proteolysis and deamination reactions) or decreases in pH (e.g., through fermentation or spoilage), or Eh (through oxidation or reduction reactions). This will in turn create environments inhibitory to some microbes but conducive to the growth of others. One example depicts the changes in growth and survival of enteropathogenic Escherichia coli during the dynamic process of making Colby cheese (Fig. 5.6). Microbial-induced changes in a food can result in a succession of various microbial populations depending on the state of the process or ripening and the type of microbiota that is present. Extrinsic treatments can also influence the intrinsic properties of a food, such as when milk is heated. Reactions occur at different temperatures which can enhance or retard microbial growth. L. monocytogenes did not grow when inoculated into raw milk and incubated at 4◦ C, in one study (Ryser, 1999). However, its population increased about 10-fold in pasteurized milk during 7 days of incubation at the same temperature.

5

What Factors Are Required for Microbes to Grow, Survive, and Die?

113

Fig. 5.6 Behavior of enterotoxigenic E. coli H10407 and changes in pH during manufacture and storage of Colby-like cheese (Kornacki and Marth, 1982). Reprinted with permission from the Journal of Food Protection. Copyright held by the International Association for Food Protection, Des Moines, IA, USA. J. L. Kornacki is president and senior technical director of Kornacki Microbiology Solutions, Inc., Madison, WI. The late E. H. Marth was professor emeritus. Departments of Food Science and Bacteriology, 1204 Linden Dr., Madison, WI. 53704

5.4.1 Hurdles The ability of intrinsic and extrinsic factors to affect the ability of microorganisms to survive or grow has been used to promote food safety and retard spoilage. A “hurdles” model does not rely upon a single treatment to ensure microbiological quality but rather a combination of treatments (intrinsic modifications and/or

114

J.L. Kornacki

Fig. 5.7 Pasteurized processed cheese spreads inoculated with five strain cocktails each of C. botulinum types A and B and held 42 weeks at 30◦ C. Open circles represent toxin negative batches. Xs represent batches with at least one toxic sample (Adapted from Tanaka et al., 1986). Reprinted with permission from the Journal of Food Protection. Copyright held by the International Association for Food Protection, Des Moines, IA, USA. Author Tanaka was professor of Food Science University of Niigata University, Niigata, Niitsu Campus Research Building 3f 303, Japan

extrinsic treatments). In this approach none of the parameters (e.g., aw , pH, temperature, OPR) taken in isolation are necessarily adequate to achieve the desired affect. Tanaka et al. (1986) modeled the ability of C. botulinum to form toxin in processed cheese with a combination of such “hurdles” including moisture, pH, salt, and phosphates (Fig. 5.7). The author was involved with a food processing facility in which a mixture of solid and liquid products were being pumped through various lines and dispensed into a package over a period of 18 h. Concern was expressed that given the long operating time, the nutrients in the product, the sub-lethal pHs, and suboptimal temperature, microbial growth may occur in the lines and product resulting in a greater microbial risk. However, based on an extensive validation study the “hurdles” concept was shown to be effective and the company reassured regarding the safety of the process. In summary, the extrinsic and intrinsic factors that impact microbial survival and growth in food or in factory niches are manifold and can be quite dynamic. This highlights the need for research to better understand the relationship of microbes to their environments. It also means that food processors should exercise appropriate caution (e.g., via challenge studies, appropriate testing, selection, and monitoring of valid CCPs) when formulating new products. Assumptions about microbial behavior in one product may not necessarily apply to another.

5

What Factors Are Required for Microbes to Grow, Survive, and Die?

115

References Banwart GJ (1979) Basic Food Microbiology. AVI Publishing Company, Westport, CT Chmielewski R, Frank JF (2004) A predictive model for heat inactivation of Listeria monocytogenes biofilm on stainless steel. J Food Prot 67(12):2712–2718 Frank JF, Marth EH (1977) Inhibition of Enteropathogenic Escherichia coli by homofermentative lactic acid bacteria in skim milk. J Food Prot 40(11):754–759 Jay J (2000) Modern Food Microbiology, 6th edn. Springer, New York Kornacki JL, Marth EH (1982) Fate of non-pathogenic and enteropathogenic Escherichia coli during the manufacture of Colby-like cheese. J Food Prot 45:310–316 Kornacki JL (1986) Thermal inactivation of bacteria in ultrafiltered milk retentates. Dissertation, University of Wisconsin Kornacki JL, Gabis DA (1990) Microorganisms and refrigeration temperatures. Dairy, Food Environ Sanit 10(4):192–195 Kornacki JL, Marth EH (1986) Heat-inactivation of Streptococcus faecium var. casseliflavus in skim milk cultures with Pseudomonas fluorescens. J Food Prot 49:541–543 Kornacki JL, Marth EH (1989) Thermal inactivation of Staphylococcus aureus in retentates from ultrafiltered milk. J Food Prot 52:631–637 Kornacki JL, Marth EH (1992) Thermal inactivation of Enterococcus faecium in retentates from ultrafiltered milk. Milchwissenschaft 47(12):764–769 Kornacki JL, Marth EH (1993) Thermal inactivation of Salmonella senftenberg and Micrococcus freudenreichii in retentates from ultrafiltered milks. Lebensm Wiss U-Technol 26:21–27 Lund D (1975) Heat processing, Chapter 3. In: Fennema O (ed) Principles of Food Science Part II: Physical Principles of Food Preservation. Marcel Dekker, New York Mitscherlich E, Marth EH (1984) Microbial Survival in the Environment. Springer-Verlag, New York, Table 26, p. 584 Ryser ET (1999) Incidence and behavior of Listeria monocytogenes in unfermented dairy products, Chapter 11. In: Ryser ET, Marth EH (eds) Listeria, Listeriosis, and Food Safety, 2nd edn. CRC Press, Boca Raton, FL citing Northholt MD, Beckers HJ, Vecht U, Toepel L, Soentoro PSS, Wisselink HJ (1988) Listeria monocytogenes: Heat resistance and behavior during storage of milk and whey and making of Dutch types of cheese. Neth Milk Dairy J 42:207–219 Tanaka N, Traisman E, Plantong P, Finn L, Flom W, Meskey L, Guggisberg J (1986) Evaluation of factors involved in antibotulinal properties of pasteurized process cheese spreads. J Food Prot 49(7):526–531

Chapter 6

Where Do I Start (Beginning the Investigation)? Jeffrey L. Kornacki

Most impediments to scientific understanding are conceptual locks, not factual lacks . . . Stephen J. Gould (Bully for Brontosaurus)

Abstract No doubt some will open directly to this chapter, because your product is contaminated with an undesirable microbe, or perhaps you have been asked to do such an investigation for another company’s facility not previously observed by you and naturally you want tips on how to find where the contaminant is getting into the product stream. This chapter takes the reader through the process of beginning the investigation including understanding the process including the production schedule and critically reviewing previously generated laboratory data. Understanding the critical control points and validity of their critical limits is also important. Scoping the extent of the problem is next. It is always a good idea for the factory to have a rigorously validated cleaning and sanitation procedure that provides a documented “sanitation breakpoint,” which can be useful in the “scoping” process, although some contamination events may extend past these “break-points.” Touring the facility is next wherein preliminary pre-selection of areas for future sampling can be done. Operational samples and observations in non-food contact areas can be taken at this time. Then the operations personnel need to be consulted and plans made for an appropriate amount of time to observe equipment break down for “post-operational” sampling and “pre-operational” investigational sampling. Hence the chapter further discusses preparing operations personnel for the disruptions that go along with these investigations and assembling the sampling team. The chapter concludes with a discussion of post-startup observations after an investigation and sampling.

J.L. Kornacki (B) Kornacki Microbiology Solutions, Inc., McFarland, WI, USA e-mail: [email protected] J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_6, 

117

118

J.L. Kornacki

6.1 Understanding the Situation (Getting the Facts) No doubt some will open directly to this chapter, because your product is contaminated with an undesirable microbe, or perhaps you have been asked to do such an investigation for another company’s facility that you have never seen before and you want tips on how to find where the contaminant is getting into the product stream. The first place to start is the existing data. This may be data from environmental, operational, post-operational, and pre-operational testing, including the result of any trend analyses (Eiffert and Arritt, 2002) and results from in-line and finished product tests.

6.2 Know the Process 6.2.1 Review the HACCP Plan A copy of the factory’s Hazard Analysis and Critical Control Points Plan (HACCP) may be obtained ahead of the factory visit and should be reviewed to get a sense of the product flow and the critical control points (CCP’s). A thorough discussion of HACCP is beyond the scope of this book, although a more in depth discussion appears in Chapter 9. However, HACCP plans are preventative and are characterized by seven essential elements including (1) documentation of an appropriate Hazard Analysis that shows (2) established and appropriate CCPs (hence the acronym HACCP), (3) validated and appropriate critical limits (CLs) for those CCPs, (4) monitoring procedures, (5) predetermined corrective actions should CLs be exceeded, (6) verification procedures, and (7) record-keeping and documentation procedures.

6.2.2 Check Assumptions 6.2.2.1 Critical Limits There is value to challenging some assumptions. Critical limits (e.g., minimum thermal processing temperature, maximum refrigeration temperature, pH) should be based upon science. Often this is the case. However, on occasion a company will base a CL for a CCP upon tradition (“we’ve always done it this way”), or (worse) the absence of reported illness. Other times CL parameters which are effective for one product matrix (e.g., milk pasteurization at 71.7◦ C for 15 s) is misapplied to an inappropriate matrix (e.g., pie fillings with high sugar content – see Chapter 5). You will want to gain some confidence that the critical limits (CLs) for the CCPs are effective. This will help you to focus the investigation. For example, if the meat smokehouse or fluid dairy product pasteurizer times and temperatures are effective and the microbe in question is heat sensitive non-spore former, then the contamination event must be occurring after this point. However, the cooling cycle of a

6

Where Do I Start (Beginning the Investigation)?

119

smokehouse or other thermal treatment (e.g., cooling plates on a pasteurizer, cooling cycle of a retorted exposed product) could be considered in the investigation. Suspicious CLs for CPs In the event that you are not able to gain confidence in the ability of a CL to adequately control a pathogen then some type of validation should be considered. This may involve a laboratory-based challenge study or even in-factory trials with time temperature indicators or surrogate microorganisms (see Chapter 7). Often time constraints may require that this be done subsequent to the initial investigation. 6.2.2.2 Laboratory Assays and Quality Systems Sometimes non-pathogens are misidentified as “pathogens.” Find out how the testing was done and determine if the method was appropriate. Was a definitive result with an appropriate method used or is the result merely “presumptive” or “suspect”? A good deal can be a stake with merely accepting the information at face value. False positives are known to exist with essentially all methods and likely to occur (usually rare) at some frequency in all laboratories, so be sure to consider this possibility. The testing laboratory should be using recognized (e.g., AOAC, FDA Bacteriological Analytical Manual, USDA’s Microbiology Laboratory Guidebook or other recognized compendial methods) methods (Cuniff, 2003; USDA, 1998; Downes and Ito, 2001; Wehr and Frank, 2004). Other sources for microbiological methods include the International Organization for Standardization (ISO), Standard Methods for the Examination of Dairy Products (Wehr and Frank, 2007), USDA’s Microbiology Laboratory Guidebook. In the absence of recognized compendial or validated assays those published in peer review publications may be acceptable. Some companies may choose to validate their own methods and this validation can be reviewed by the investigator. It is important to remember that the data are only as good as the assay, the quality systems that back it up and the sampling technique. Any laboratory that is performing pathogen testing should have adequate controls in place to prevent and track laboratory contamination. This may involve testing employee hand samples before and after sample enrichments, transfers, and streaking activities and, also, routine laboratory environmental samples. Questions that can be asked of a laboratory reporting a positive pathogen result (among others) can be the following: (1) What evidence can you provide that the media used to run the specific assay in question was sterile? (2) Were negative controls used? If so, did any “negative” controls have a “positive” result? It is important to note that negative assays on truly negative samples (samples in which the target pathogen is not present) validate positive results on similar contaminated samples. (3) Were technician “hand samples” taken and did any test positive for the organism in question? (4) Did the testing facility perform routine environmental samples within the laboratory for the microbe of interest? If so, were any found in the past month? (5) What positive control isolate does the laboratory use? (positive controls validate negative findings on routine test samples). Is it the same as that found in my product? If so,

120

J.L. Kornacki

molecular subtyping (Chapter 10) techniques may be necessary to determine if the organisms are truly the same or different strains, and (6) Did any other samples test positive by the laboratory within several days on either side of the day the testing was begun on the implicated sample. The answers to these questions may cause the final result to be questioned. However, no amount of retesting of a previous positive lot will guarantee with absolute certainty that the first “positive” result was incorrect apart from very strong evidence of a laboratory cross-contamination (see Chapter 8). Other important questions that can be helpful with regard to a laboratory investigation related to a potential questionable result can be “Does the laboratory maintain tight records of the order in which samples are received, pre-enriched, transferred, streaked, etc.? (positive samples in close relation to each other in the same test tube rack suggests the potential for cross-contamination that should be investigated); What technician(s) did the assays on the samples in question?; Is there any evidence that the technician(s) may have handled positive samples several days on either side of the implicated sample receipt date. Is the isolate recovered rare or common subtype? If it is rare, was it found in another clients sample within the last several months? Such questions can assist with an investigation of a laboratory into an “equivocal” result.

6.3 Determine the Extent of Contamination (Scoping the Problem) It is always a good idea for the factory to have a rigorously validated cleaning and sanitation procedure that provides a documented “sanitation breakpoint.” However, sometimes contamination events transcend sanitation to sanitation production runs. In that event the full scope of product contamination needs to be determined. Application of rigorous statistically based, aseptic sampling of finished product may help to determine the extent of lot contamination (see Chapter 8). In the event of a Class I product recall the government will likely collect data related to specific lot codes associated with sickness. This information is also useful in scoping the extent of contamination.

6.3.1 Determine the Production Schedule If the factory is not already shut down, one needs to determine the production schedule. It is often best to arrive at the food processing plant when it is still in operation, especially if the product is subjected to handling by the employees. This allows time for microbes to grow and move through the environment often creating a “worse case” environment which is helpful to finding problems.

6

Where Do I Start (Beginning the Investigation)?

121

6.3.2 Touring the Plant and Preliminary Site Selection One has now looked over the original (pre-visit) data generated, met at and with the factory personnel, determined the production schedule, and understood the CCPs. It is now time to do the first factory walk-through. During the walk-through nonproduct contact environmental samples could be pre-selected and taken (see Chapter 7). This is a great time to understand the operation (ask lots of questions of your facility hosts) and make notes of where potential problems may exist in the context of a risk assessment walk-through (see Chapter 4), begin to pre-select future sampling sites to take when operations are down, and begin to formulate which assays one may do on samples taken (Chapter 4). One will often take more samples than initially anticipated, after the plant is shut down and you are able to have equipment taken apart for observation and sampling. It is very common to discover many more areas to sample than initially anticipated.

6.3.3 Prepare Operations for Disruption We have found it very useful to expend the vast majority of our in-factory investigation time during the period after the factory operations have been shut down. The time frame for effective sampling will impact the timing of subsequent sanitation and start-up. It is prudent, if not essential in many cases, to inform plant operational personnel of this likelihood. In our experience very often 8–12 h may be expended in this phase of a typical investigation. However, time frames will vary from situation to situation and could be a great deal longer depending upon the situation; after all it is an investigation into the unknown, not a cook book audit.

6.3.4 Assembling the Sampling Team It is wise to pre-number the sterile sample collection containers (e.g., pre-sterilized plastic bags, or jars) prior to entering into the processing area for the sampling phase of the investigation. The optimal team has four members: the principle investigator (PI), a scribe, a maintenance person, and an assistant sampling person. This will speed the investigation and free the PI for more detailed observations.

6.3.4.1 Maintenance Staff One important way that foods become contaminated is through residues entrapped within processing equipment that is not fully taken apart for effective cleaning and sanitation (see examples in Chapter 4). The assistance of maintenance personnel is essential in many instances. Their help with proper equipment breakdown and reassembly can be invaluable to the

122

J.L. Kornacki

investigational process. Those with greater longevity may also provide historical information about the various modifications of equipment and facility that have occurred over the years. This information can also be very useful in the investigational process.

6.3.4.2 Quality Assurance Personnel The cooperation, insights, and experience of these people within the factory are frequently invaluable. Sometimes their observations and opinions have been ignored and the input of the outside investigator can be very helpful to establish their credibility within the processing facility.

6.3.4.3 Scribes In our experience the facility quality assurance person is often the scribe or the assistant sampler. Scribes are ideally familiar with the factory and the terminology used within the facility to describe various rooms and pieces of equipment. These people are responsible for neatly recording sample descriptions that correlate to the numbered pre-sterilized containers. The author prefers that sample descriptions be recorded in order from general location to specific, noting if the sample was taken pre-, post-, or during operations (e.g., “Post-Operational-Packaging Room, East side, Slicer # 4, blade”). Many times various individuals in the factory will object to disassembly and sampling of certain equipment or areas in the factory. Phrases such as “nothing could ever get behind that gasket,” “we never take that apart,” “we’ve never broken down equipment this far before” all suggest areas that have not been effectively taken apart for cleaning and sanitation. These areas in or near product streams have often proven to be the source of the problem. Examples of this abound. In the past, contaminated or wet residues have been detected behind several layers of gaskets on product valve stems, gaskets for diaphragm, or other types of pumps, gable “crimpers” on “aseptic” fillers (see examples in Chapter 4).

6.3.5 Picture Taking Seeking permission to take close up pictures of various areas in the factory can be a very powerful tool when combined with sampling data (e.g., swabs) to provide further clarity to the sampling site and often has the advantage of effecting risk reducing changes with upper management. Such pictures can be very useful in report writing, singling out various areas for future sampling, and can be used in training plant personnel and equipment design.

6

Where Do I Start (Beginning the Investigation)?

123

6.3.6 Sanitation Observations and Validation Differences may exist between written sanitation standard operating practices (SOPs), what sanitation supervisors or management say occurs, and actual sanitation practiced in the factory. Observations of cleaning and sanitation by quality assurance personnel can be very revealing and are encouraged. Sometimes product contamination events result from improper application of “cleaning and sanitation.” For example, spraying Listeria-contaminated ready-to-eat product cooler floors with high-pressure hoses will likely result in contamination of product not previously removed from the cooler. However, knowledge of such occurrences may not be fully comprehended short of actual observation. Taking preoperational samples after sanitation is also an important way to assess the efficacy of sanitation (see Chapter 7).

6.3.7 Post-Startup Observations and Sampling Chapter 9 deals with start-up-related issues in more detail. However, if the factory is not permanently shut down, it is often prudent to spend time in the food processing facility after the sampling portion of the investigation and to make further observations of the factory process, traffic patterns, and airflow. This may result in critical observations and further sampling.

References Cuniff PA (ed) (2003) Official Methods of Analysis of AOAC International, 17th edn, 2nd revision. Association of Official and Analytical Chemists, Gaithersburg, MD Downes FP, Ito K (eds) (2001) Compendium of Methods for the Microbiological Examination of Foods. American Public Health Association, Washington, DC Eifert JD, Arritt FM (2002) Evaluating environmental sampling data and sampling plans. Dairy, Food Env Sanit 22(5):333–339 USDA (1998) Microbiology Laboratory Guidebook, 3rd edn. http://www.fsis.usda.gov/Science/ Microbiological_Lab_Guidebook/index.asp. Accessed 30 March 2010 US FDA Center for Food Safety and Applied Nutrition (2001) Bacteriological Analytical Manual (Online) http://www.cfsan.fda.gov/∼ebam/bam-toc.html. Accessed 25 July 2008 Wehr HM, Frank JH (2004) Standard Methods for the Examination of Dairy Products, 17th edn. American Public Health Association, Washington, DC

Chapter 7

How Do I Sample the Environment and Equipment? Jeffrey L. Kornacki

Abstract Food product contamination from the post-processing environment is likely the most frequent cause of contaminated processed food product recalls and a significant source of poisoning outbreaks, and shelf life problems in North America with processed Ready-To-Eat foods. Conditions exist for the growth of microorganisms in most food processing factories. Failure to clean and effectively sanitize a microbial growth niche can lead to biofilm formation. Biofilms may be orders of magnitude more resistant to destruction by sanitizers. Cells in some biofilms have been shown to be 1,000 times more resistant to destruction than those which are freely suspended. This has implications for cleaning, sanitizing, sampling, and training. Sampling the factory environment is one means of monitoring the efficacy of microbiological control as well as a powerful tool for in-factory contamination investigation. Many sampling techniques exist and are discussed. It is important to recognize the difference between cleaning (removal of soil) and sanitization (reduction of microbial populations). Knowing where, when, and how to sample, how many samples to take, and what to test for and how to interpret test information is critical in finding and preventing contamination.

7.1 Introduction Food product contamination from the post-processing environment is likely the most frequent cause of contaminated processed food product recalls and a significant source of poisoning outbreaks, and shelf life problems in North America (Kornacki, 1999/2000). It is impractical to create and maintain a sterile food processing environment. Therefore, manufacturers should strive to maintain strict control over the

J.L. Kornacki (B) Kornacki Microbiology Solutions, Inc., McFarland, WI, USA e-mail: [email protected] J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_7, 

125

126

J.L. Kornacki

microbial ecology and microbial growth within the factory environment. Factory conditions that promote growth of microbes increase the risk of post-processing product contamination. Sampling the factory environment is one means of monitoring the efficacy of microbiological control as well as a powerful tool for in-factory contamination investigation.

7.2 Factors That Result in Microbial Growth Niche Formation Many factors affect the growth of microorganisms, including moisture, nutrients, pH, oxidation–reduction potential, temperature, presence or absence of inhibitors, interactions between microorganisms in a population, and time (Faust and Gabis, 1988; and Chapter 5). Conditions exist for the growth of microorganisms in most food processing factories. Moisture is the most critical of these because it is absolutely required for microbial growth (Faust and Gabis, 1988). Areas where moisture, nutrition, and adequate time for microbial growth exist at a permissive temperature have proven to be excellent microbial growth niches and should be selected for sampling (see Fig. 7.1).

Fig. 7.1 Factors that result in microbial growth niches

7

How Do I Sample the Environment and Equipment?

127

7.3 Factors that Result in Biofilm Formation Given enough time, many bacteria can attach themselves to underlying surfaces within niches by cell wall-bound structures (e.g., proteins, polysaccharides, glycoproteins). Bacteria that attach and are allowed to form biofilms can be protected from cleaners, sanitizers, and heat (Debeer et al., 1994; Dhir and Dodd, 1995). Bacterial survival while attached to a surface is dependent on many factors. Temperature, time, and the availability of nutrients and water are also considered important for biofilm development. The attachment surface itself plays a role in how well an organism can survive. The physicochemical properties of a surface including nutrients influence cellular attachment. It has been previously reported that the chemical nature of a surface (surface charges, pH, hydrophobicity/ hydrophilicity) can either inhibit or promote attachment and survival of different organisms. Interestingly Fletcher (1976) showed that bovine serum albumin inhibited bacterial attachment to various surfaces. This effect, though, could not be a property of the conditioning layer only, but also the interaction of freely suspended albumin with cell surface structures involved in attachment that lead to a reduced ability of the cells to bind to the substratum-bound albumin.

7.3.1 Some Implications of Growth Niches and Biofilms to the Food Factory Environment Some lessons can be drawn from the information presented above regarding growth niches and biofilms. Some of these follow. Failure to clean and effectively sanitize a microbial growth niche site will likely lead to development of a biofilm at that site that may be orders of magnitude more resistant to destruction by sanitizers. Approved sanitizers will have passed a Chamber’s sanitation efficacy test (1956). Simply stated this assay measures the ability of a sanitizer to destroy 5 log10 CFU’s of freely suspended cells of a selected strain each of Staphylococcus aureus and Escherichia coli under defined conditions. However, cells in some biofilms have been shown to be up to 1,000 times more resistant to destruction than those which are freely suspended. Thus, sanitization may well eliminate freely suspended (and therefore unprotected) cells in a growth niche but the attached cells may remain. Given the right conditions of water, food, and time in a growth permissive temperature range, these cells begin another cycle of growth. If such conditions are created during food processing in equipment or areas in close proximity to product, then measurable contamination events may begin to occur. This is one reason why is it important to (1) effectively break down equipment for cleaning and sanitizing; (2) apply cleaning water at temperatures suitable to solubilize or liquefy proteinaceous or fatty residues, respectively, that may otherwise, entrap and protect bacteria from aqueous sanitizer; (3) monitor the effectiveness of cleaning and sanitation; (4) monitor and control the microbial ecology of the food factory environment;

128

J.L. Kornacki

(5) seek solutions to persistent sanitation failures; (6) effectively train sanitation personnel in proper techniques; (7) train all employees in, and enforce, good manufacturing practices; (8) prevent the excessive use of moisture; (9) discourage where possible the use of high-pressure hoses, etc.

7.4 Sampling Techniques Comparisons between some sampling techniques are shown in Table 7.1. Table 7.1 Comparison of environmental sampling approaches Sampling technique

Quantitative assays possible

Qualitative assays possible

Heavily soiled surfaces

Traditional swabs Pre-sterilized sponges Contact plates Pre-sterilized tongue blades

Yes Yes Yes Yes

Yes Yes Yesa Yes

No Yes No Yesb

a Possible b Biofilm

with non-traditional approaches removal possible

7.4.1 Sponge Samples Silliker and Gabis (1975) sampled a factory known to be contaminated with Salmonella using both sponge and swab techniques. The use of a pre-sterilized sponge technique successfully recovered Salmonella from the factory environment, whereas the swabs did not. Pre-sterilized sponges allow one to apply more pressure to the surface and cover more surface area than swabs, thus increasing the likelihood of effectively sampling biofilms and thereby finding contamination sites. Sponges have the disadvantage of not being able to effectively sample small crevices and penetrations. Furthermore, a uniform amount of pre-sterilized diluent can be added to the sponge sample, and dilutions prepared for quantitative analyses, e.g., coliforms, Enterobacteriaceae, coagulase-positive Staphylococci, Enterococci, yeast, and molds. Aliquots can be used for qualitative analysis as can the original sponge sample.

7.4.2 Swab Samples Swabs are the most traditional approach to sampling surfaces whereby moist presterilized swabs (e.g., cotton or alginate) are rubbed over a measured area (Silliker and Gabis, 1975). This approach offers similar advantages to sponges (e.g., able to be assayed by qualitative and quantitative methods) except that less pressure must

7

How Do I Sample the Environment and Equipment?

129

be applied but smaller surface areas can be effectively sampled. Small crevices and penetrations can be sampled.

7.4.3 Contact Plate Different direct contact approaches exist (e.g., prehydrated PetrifilmTM , RodacTM plates, agar slide method). These have the advantage of obtaining a quantitative result directly from the surface. It is often stated that qualitative analyses are not possible with such approaches. However, this author has found that removal of the agar subsequent to surface sampling followed by dilution, homogenization, plating, and enrichment may provide a means to do both quantitative and qualitative assays, respectively. They have the disadvantage of only being able to sample a flat surface. Moore et al. (2001) found that plate count agar (PCA) dip slides were able to recover a wider population range of E. coli cells from a wet surface than PCA spread plates, PCA pour plates, two ATP assay systems, and four protein detection kits.

7.4.4 Pre-sterilized Tongue Blades Personal experience has demonstrated the value of these in collection of dry scrapings or the ability to sample between sandwiched regions of certain equipment. The portion of the tongue blade in contact with the sampled region can also be aseptically broken-off into a pre-sterilized container, to which sterile diluent can be added. Both quantitative and qualitative assays can be done. Others have used pre-sterilized scoops and pre-sterilized metal scrapers, as appropriate.

7.4.5 Air Sampling A wide variety of devices for sampling the air exist. The basic designs include impingers, slit-type impactors, sieve-type impactors, filtration samplers, centrifugal samplers, and electrostatic precipitation impactors. Evancho et al. (2001) has an excellent review of the types of air samplers and the advantages and disadvantages of each.

7.5 ATP Bioluminescence and Protein Assays It is important to recognize the difference between cleaning (removal of soil) and sanitization (reduction of microbial populations). One can, in theory, have a soiled surface with no measurable microbial count; a microbiologically contaminated surface which is “clean” (e.g., no visible or measurable soil); a soiled and contaminated surface; and a clean and sanitized surface. All living cells contain ATP. Many types

130

J.L. Kornacki

of in-factory soil may be proteinaceous, including bacteria. However, these assays have been shown to be more effective at detection of soil or unclean surfaces rather than contaminated surfaces (Moore et al., 2001). Conversely, sampling surfaces and performing microbiological assays on those samples do not reveal the efficacy of cleaning, only the efficacy of sanitation. Given enough time under favorable conditions (e.g., water, food, time, growth-conducive temperature) unclean surfaces with no measurable microbial counts will become microbial growth niches. Consequently it is important for food production facilities to monitor cleaning and sanitization. Commercial ATP and protein-based assays provide a better indication of cleaning efficacy than mere visual inspection. In this author’s view monitoring cleaning either visually or (better) with ATP and/or protein-based assays (as appropriate) and sanitation via microbiological assays is a more prudent approach than merely relying upon one or the other.

7.6 In-Line Sampling Taking samples at selected locations throughout the process can provide useful information about the ability of the process to remain under microbiological control. In our experience more can be learned in an investigation from effective sampling of equipment and the environment by taking sponge or swab samples than from product sampling or in-line. However, this is not always possible. The statistics associated with sampling (see Chapter 8) mean that extensive numbers of samples must be taken to gain meaningful data. A microbial growth niche may contain millions or even billions of cells per sponged area. Hence if one takes the proper sample of this niche, it is relatively easy for an appropriate assay to detect the contaminant of concern. However, should one drop (0.05 ml) of fluid in this niche with bacterial count of 100 million CFUs per ml contaminate 10,000 pounds of product then only an average of 14 cells per gram would be present in the product. However, the distribution of these microbes is very often not homogeneous (see Fig. 7.2 and Chapter 8), and therefore the target organism may not necessarily be detectable in a single sample. This presents a need for rigorous sampling. Personal experience in numerous factory situations has shown that taking 20 × 375 g grab samples of wellmixed powdered product before and after an area of the process that is inaccessible for sampling has proven useful. In Fig. 7.2 raw ingredient(s) go through processing steps “A” through “F.” “A” represents the raw ingredient(s). In step “F” the product is packaged and stored.

Fig. 7.2 Hypothetical food process diagram

7

How Do I Sample the Environment and Equipment?

131

Step B is a CCP which destroys microbes native to the raw product. In this example, equipment between process B and C is very difficult to break down for routine cleaning, sanitization, and drying. The same is true for equipment between steps “D” and “E.” Hence aggressive sampling at the designated “in-line sample points” could prove useful in the context of a contamination investigation.

7.7 Where to Sample Air, water, tools, workers, traffic, and other means transfer microbes in the factory environment into areas that are inaccessible for cleaning and sanitation. Microbial growth niches result when conditions in the factory are conducive for microbial growth. Hence samples of tools, workers hands and foot coverings, forklift tires are some areas to consider. An extensive review of these areas can be found in Chapters 4 and 6.

7.8 Suggestions for Sampling Supplies Experience has shown the following supplies to be useful in conducting an infactory investigation. The items listed in Table 7.2 have been very useful in our investigations. These include the following: Sampling the factory environment is a crucial element in a processed food contamination investigation and a variety of techniques exist, each with their own advantages and disadvantages.

Table 7.2 Recommended sampling supplies Pre-sterilized

Disinfectable

Other

Sample collection bags (twist top) Sample collection jugs with screw cap hermetically sealable lid Sample collection scoops

Scissors

Duct tape

Knives

Alcohol wipes

Box cutters

Sanitizer concentration test strips pH test strips Flashlight Frozen coolant Insulated shipper

Tongue depressors Sponges (inhibitor free)a Swabs Neutralizing bufferb a Natural

Tools (screw driver) Electronic portable thermometer

sponges may contain microbicidal inhibitors buffers should contain chemicals to inactivate residual sanitizer that may be present in sample b Neutralizing

132

J.L. Kornacki

7.9 When to Sample 7.9.1 The Value of Pre-operational Swabs The factory environment and equipment should be effectively cleaned, sanitized, and thoroughly dried prior to food processing. Sampling and appropriate testing of the environment and product contact surfaces before operations (subsequent to sanitization and drying) will provide a good indication of the efficacy of cleaning and sanitation. APCs in excess of 100–1,000 per ft2 and/or the presence of any coliforms or Enterobacteriaceae on such a pre-operational surface indicate a failure to effectively clean and sanitize. Many times we have observed equipment surfaces that look visibly clean only to find aerobic plate counts (APC) in excess of 100,000 per sponged area (often about 1 ft2 ) or the presence of coliforms or Enterobacteriaceae. It is our view that such counts will likely result from microbial growth rather than from the air or transient contamination. Sometimes factories lack adequate hot water to effectively remove proteinaceous and fatty deposits that entrap and protect bacteria. Others have suggested guidelines with regard to counts per square inch (Evancho et al., 2001). However, I recommend against the use of templates in the context of a microbiological investigation, because the goal is to locate a problem area which can mean sampling irregular surfaces or crevices (e.g., rollers on conveyor belts, tines on slicing machines, gears on sausage peelers). In some instances (e.g., salad dressing manufacture) the presence of even one cell of Saccharomyces bailii or Lactobacillus fructivorans on a product surfaces can present a serious spoilage hazard to the product (Evancho et al., 2001).

7.9.2 The Value of Operational Swabs It is useful to perform a microbiological “risk assessment” of the processing environment during the course of operations (see Chapter 6). Sites for post-operational breakdown and sampling can be pre-selected at this time. In addition observations of personnel, product, vehicular (wheeled containers, forklifts, handcarts, etc.) and air movement can be made. These observations may result in selection of likely sites of cross-contamination in the factory environment.

7.9.3 The Value of Post-operational Swabs These samples often provide “worst case” data set and are very useful with infactory investigations. In many instances product residues will have accumulated on both product contact and non-contact surfaces for many hours, often under conditions conducive to microbial growth. Factory workers, and sometimes wheeled transport vehicles, conveyor belts, etc., will have been moving about the factory for sometime thus disseminating microbes.

7

How Do I Sample the Environment and Equipment?

133

7.10 What Do I Test for? Indicator Microorganisms 7.10.1 Index Organisms The concept of indicator organisms goes back about a century and stems from the practice of testing for E. coli as an indication of the possible presence of Salmonella. This made some sense then when the test methods for E. coli were better established and less expensive than those for Salmonella. The concept was that mammalian feces can contain enteric pathogens (e.g., Salmonella, Shigella) and they were also common sources of E. coli. The presence of E. coli in water supplies was taken to be an indication of direct fecal contamination. Thus, the concept of a non-pathogen indicating the possible presence of a pathogen was born (Kornacki and Johnson, 2001). The concept was useful a century ago and is still used today despite the fact that some water samples have tested negative for E. coli that were positive for the presence of Salmonella (Kornacki and Johnson, 2001). The concept of an indicator organism was eventually expanded to coliforms (not often related to fecal contamination) and then to dairy products and other foods. However, this is a classic example of extrapolating microbial behavior in one matrix (water) to very different matrices, such as milk in which microbes are likely to grow, rather than die. The presence of coliforms and E. coli in processed food may be useful as indicators of quality but their use as index organisms which indicate the possible presence of a pathogen in food has been discredited (Kornacki and Johnson, 2001). In recent times the use of a truncated Listeria assay has gained popularity in the processed meat industry. In these assays the test for Listeria monocytogenes is stopped prior to identification of the genus (e.g., Listeria-like organisms or “esculin-hydrolyzing bacteria” and “Listeria species,” respectively). Stopping short of confirming a Listeria-like organism or a Listeria (species unknown) as L. monocytogenes allows a factory to take corrective action without the regulatory implications of recovering this organism from a food contact surface. Organisms that may appear to be similar to Listeria spp. after enrichment and plating onto Modified Oxford Medium may include Lactobacillus spp., Enterococcus spp., and others (Kornacki et al., 1993). In principle, a negative Listeria-like organism or Listeria species assay should indicate a negative L. monocytogenes assay, assuming that the same assay steps are used with these “index tests” as are used up to the same point with approved L. monocytogenes assays. However a “positive” Listerialike organism assay on a surface may be an indicator of the need to improve the sanitary quality of that surface.

7.11 Indicators of Quality More appropriate uses of non-pathogen (e.g., coliforms, yeast and molds, APC) assays are as indicators of food quality. The author was involved with a manufacturer of an oat-based product which routinely had APCs in the 100–600/g range.

134

J.L. Kornacki

However, about once per month a sample was found to have counts numbering in the millions/g. In-factory investigation revealed a source of high populations of mesophilic aerobic spores that sporadically sloughed into the product stream. This indicator of quality (APC) was useful in stimulating an investigation and improving the microbiological quality of the product. In the context of the investigation the environment was extensively swabbed for mesophilic aerobic spores and a near product area was found which contributed to sporadic introduction of these organisms into the product stream.

7.12 Indicators of Sanitation Efficacy One of the best indicators of effective sanitation is the use of the APC and coliform count (see discussion of pre-operational samples above).

7.13 Surrogates These organisms are not used in the context of environmental testing but have been used to validate CCPs. For many years Clostridium sporogenes strain PA 3679 has been used as a thermal surrogate in place of Clostridium botulinum. This has allowed for thermal process development without the necessity of working with a dangerous pathogen (see Chapter 2). Enterococcus faecium NRRL B-2354, also called Pediococcus sp. NRRL B-2354 (formerly Micrococcus freudenreichii), was used to establish pasteurization time and temperature treatments used with milk, chocolate milk, and ice cream mix (Speck, 1947; Speck and Lucas, 1951). Later it was used in validation of a variety of juice pasteurization treatments (Annous and Kozempel, 1998). A review of the data generated by Kornacki and Marth (1993) indicated that heat resistance (D and z values) of this organism in milk and ultrafiltered concentrated milk was very similar to that of Salmonella senftenberg 775 W believed to be one of the most heat-resistant Salmonella strains in high water activity foods (Ng et al., 1969). Ma et al. (2007) indicated that this organism was more heat resistant than Salmonella senftenberg 775 W and L. monocytogenes in lean ground beef. Hence destruction of this microbe would ensure destruction of Listeria and Salmonella in lean ground beef. Sometimes natural contaminants can be used in place of added surrogates in a process to validate its efficacy to control microbes. Several examples follow: Naturally contaminated ground beef was shown to contain 200 enteric bacteria/g prior to oven treatment after which, 3 of 10 enrichment samples (each of which were 375 g) tested positive despite average temperatures of 160◦ F in an unpublished study by this author. Other assays revealed undercooked portions within some of the finished products.

7

How Do I Sample the Environment and Equipment?

135

Kornacki (2006) observed destruction >3 log10 CFU of naturally contaminated cocoa beans with mesophilic aerobic spores during roasting. These data were used to validate the ability of the in-factory roasting process to destroy Salmonella species, widely known to be a much less heat-resistant microbe. Samples of raw sausage naturally contaminated with Enterococci, Enterobacteriaceae, and lactic acid bacteria were cooled instantly upon emergence from an in-factory continuous convection oven. Product temperatures were also taken. Assays were done for Enterococcus spp., Lactobacillus spp., and Enterobacteriaceae. Product temperatures in excess of 180F, absence of Enterobacteriaceae and reductions in heat-resistant Enterococcus and Lactobacillus product isolates (as shown to be more heat resistant than Salmonella and Listeria monocytogenes in a subsequent laboratory study) indicated an effective process (Kornacki, 2006).

7.14 Summary Environmental sampling is a powerful investigational technique. A variety of approaches can be used depending on the type of surface or product one wishes to sample; these include but are not necessarily limited to sponge, contact plate, tongue blade, traditional swab, ATP and protein assays, and air sampling. Selection of the proper test or indicator organism is important as “rules of thumb” exist for interpretation of recovered populations of indicator organisms associated with pre-operational and post-operational surfaces. The usefulness of indicator organisms, truncated Listeria testing, surrogate microbes, and in-line testing were also described.

References Annous BA, Kozempel MF (1998) Influence of growth medium on thermal resistance of Pediococcus spp. NRRL B-2354 (formerly Micrococcus freudenreichii) in liquid food. J Food Prot 61:578–581 Chambers CW (1956) A procedure for evaluating the efficiency of bactericidal agents. J Milk Food Technol 19:183–187 DeBeer D, Srinivasan R, Stewart PS (1994) Direct measurement of chlorine penetration into biofilms during disinfection. Appl Environ Microbiol 6:4339–4344 Dhir VK, Dodd CER (1995) Susceptibility of suspended and surface-attached Salmonella to biocides and elevated temperatures. Appl Environ Microbiol 61:1731–1738 Evancho GM, Sveum WH, Moberg LJ, Frank JF (2001) Microbiological monitoring of the food processing environment, Chapter 3. In: Downes FP, Ito K (eds) Compendium of Methods for the Microbiological Examination of Foods. American Public Health Association, Washington, DC Faust RE, Gabis DA (1988) Controlling microbial growth in food processing environments. Food Tech 42:81–82, 89 Fletcher M (1976) The effects of proteins on bacterial attachment to polystyrene. J Gen Microbiol 94:400–404

136

J.L. Kornacki

Kornacki JL, Evanson DJ, Reid W, Rowe K, Flowers RS (1993) Evaluation of the USDA protocol of detection of Listeria monocytogenes. J Food Prot 56(5): 441–443 Kornacki JL (1999/2000) The nuts and bolts of food safety. Food Testing Anal 40:18–22 Ma L, Kornacki JL, Zhang G, Lin CM, and Doyle MP (2007) Development of thermal surrogate microorganisms in ground beef for in-plant critical control point validation studies. J Food Prot 70(4):952–957 Kornacki JL (2006) Industry Case Studies, Applied Use of Surrogate Microorganisms, Surrogate Microorganisms: Selection, Use, and Validation, Session S07, IAFP Annual Meeting, Calgary, 14 August 2006 Kornacki JL, Marth EH (1993) Thermal inactivation of Salmonella senftenberg and Micrococcus freudenreichii in retentates from ultrafiltered milks. Lebensm–Wiss U-Technol 26:21–27 Kornacki JL, Johnson JL (2001) Enterobacteriaceae, coliforms and Escherichia coli as quality and safety indicators, Chapter 8. In: Downes FP, Ito K (eds) Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC Moore G, Griffith C, Fielding L (2001) A comparison of traditional and recently developed methods for monitoring surface hygiene within the food industry: A laboratory study. Dairy, Food Environ Sanit 21(6):478–488 Ng H, Bayne HG, Garibaldi JA (1969) Heat resistance of Salmonella: The uniqueness of Salmonella senftenberg 775 W. Appl Microbiol 17:78–82 Silliker JH, Gabis DA (1975) A cellulose sponge sampling technique for surfaces. J Milk Food Technol 38(9):504 Speck ML (1947) The resistance of Micrococcus freudenreichii in laboratory-high-temperatureshort-time pasteurization of milk and ice cream mix. J Dairy Sci 30:975–981 Speck ML, Lucas HL (1951) Some observations on the high-temperature short-time pasteurization of chocolate milk. J Dairy Sci 34:333–341

Chapter 8

How Many Samples Do I Take? Jeffrey L. Kornacki

Abstract It is sometimes falsely assumed that microbes are evenly distributed throughout the entire food mass. This is rarely true. Sporadic contamination of food production lots is a very common phenomenon requiring increased sampling to gain some measure of confidence in results. Sample composites are often used to reduce the testing burden. However, before using a compositing scheme, one should verify that it is valid for the food matrix, the microbe of interest, and the assay used. Focused sampling and testing of the portion of a product in which the microbe is most likely to be is one approach which can be considered. Often “positive” results are questioned. Verification of “positive” samples through retesting is not an effective or appropriate means to determine whether or not the testing laboratory made an error. Nevertheless, there are diagnostic questions that can be addressed to the testing laboratory before accepting a “positive” result. Food processors should establish appropriate specifications for their incoming ingredients. Certain key elements should be included in any microbiological specification. Sampling plan inadequacies usually have to do with taking an insufficient number of samples without regard to accepted statistical sampling plans. Sampling plans to consider include attributes plans and variables plans as described in ICMSF plans and others.

8.1 The Statistics of Sampling and Resampling “How many samples should I take?” is a common question that arises whenever one seeks to find an organism in food. It is sometimes falsely assumed that microbes are evenly distributed throughout the entire food mass. This is rarely true. Environmental contamination of foods is frequently sporadic. This can occur when microbes are sloughed into a food from a processing stream or contaminants drip J.L. Kornacki (B) Kornacki Microbiology Solutions, Inc., McFarland, WI, USA e-mail: [email protected] J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_8, 

137

138

J.L. Kornacki

Fig. 8.1 Nonrandom distribution of contaminants (as represented by white dots) in production sublots 1–20

or splash into the product (see Chapter 7). Hence, one can expect that an inoculum may enter the food stream at some point during processing and gradually diminish to non-detectable levels as depicted in Fig. 8.1. This may happen multiple times (see Fig. 8.1) and from multiple sources. Even when microbes are homogeneously distributed, there is no guarantee that the microbe will be present in a particular sample as depicted in the even-numbered boxes in Fig. 8.2. Given the complex nature of environmental contamination, there is a great need to understand the statistics behind sampling plans and the role of testing to detect environmental contamination of foods. A brief glance at the statistical probabilities associated with finding a contaminated lot reveals the unsuitability of exclusive reliance on finished product testing to detect a contaminated lot (Fig. 8.3). A glance at Fig. 8.3 reveals a 95% chance of finding a 10% contaminated lot if one takes and tests 30 random samples. In this instance there is a 90% chance that at least 1 of the 30 samples will test positive. However, one would need to take 299 random samples to detect a lot that is contaminated at the 1% level with 95% confidence. The better a

Fig. 8.2 Randomly distributed contaminants in food sublots (contaminants represented by white dots)

8

How Many Samples Do I Take?

139

Fig. 8.3 Finished product sampling: a statistical perspective. Adapted from: Compendium of Methods for the Microbiological Examination of Foods, 4th ed

food processor controls product contamination, the less likely they are to find a contaminant given the same level of testing. In practical terms, companies producing lots contaminated at the 10% or even 1% level with a dangerous pathogen would soon cease to exist due to foodborne illness, litigation, recalls, and negative public exposure. Consequently, if one is dealing with an infectious pathogen, it is both impractical and dangerous to exclusively rely upon finished product testing to detect a contaminated lot. This is why it is so important that companies have effectively implemented good manufacturing practices (GMPs), other prerequisite programs, Hazard Analysis and Critical Control Points (HACCPs), and appropriate ingredient specifications.

8.2 Investigational Sampling In the course of an investigation, it may be necessary to take samples before and after a portion of the process where environmental sampling is not possible. Personal experience in hundreds of factories (Kornacki, 1999/2000) indicates that twenty 375 g sample composites comprised of 300 randomly selected 25 g samples have been effective in locating contaminated regions within factories, when the organism of concern is Salmonella. In some instances, twenty 375 g grab samples from a well-mixed lot have been effective. The author has noted the same with 6 × 125 g samples tested for Listeria in a cheese product wherein each of the 125 g samples

140

J.L. Kornacki

was composed of 5 × 25 g subsamples (this approach is unlikely to be acceptable for all cheeses, as shown by McMahon, 2005, see also Evanson et al., (1992). However, before using a compositing scheme one should verify that it is valid for the food matrix, the microbe of interest, and the assay used. For example, 125 g sample composites of ready-to-eat meat products have been shown to be acceptable with some Listeria assays but not others (Curiale, 2000). It is important to note that the only way of determining whether or not a lot is contaminated with absolute confidence is by testing the entire lot with a perfect assay, neither of which is likely to occur.

8.3 Other Approaches to Enhanced Detection of a Pathogen It is to one’s advantage to use the most sensitive sampling techniques and assays in the context of an investigation. In this regard a few techniques can be considered. Focused sampling on the portion of a product where the microbe is most likely to be is one approach which can be considered. For example, ready-to-eat meat products that have passed through a validated smokehouse treatment are likely to be contaminated only on or near the surface. Hence, sampling proportionately huge amounts of cooked material in relation to the surface of the product is not likely to be an efficient means of finding the target microbe. In this event it may be wise to rinse the exterior of the product or the package and test the rinse fluid. This concept has also been recognized by the USDA (Wallace et al., 2003; Luchansky et al., 2002). Another approach that worked quite well in one of the author’s investigations with Cold Pack cheese involved taking 2 × 500 composite samples, stomaching these 1:1 with enrichment broth, then removing 15 × 50 g aliquots of these and enriching each of these in 450 ml of appropriate enrichment broth, and performing a PCR assay. In this instance one production line was considered to have a contamination problem which was ostensibly corrected prior to testing according to this scheme. The other production line had no previously detected contamination. When this composite scheme was applied to finished product from both lines, 28 of 30 enrichments tested positive in the previously implicated line. However, 17 of 30 enrichments tested positive by PCR in the production line previously not considered to be contaminated. It seems likely that had immunocapture techniques been applied, the sensitivity of the assay may have been even greater.

8.4 Resolution of Questionable Data 8.4.1 Retesting Verification of previous “positive” samples through retesting is not an effective or appropriate means to determine whether or not the testing laboratory made an error (e.g., cross contamination from within the analytical laboratory). The basis for this statement follows:

8

How Many Samples Do I Take?

141

1. If the lot is contaminated at the 1% level, then there is a 1/100 × 1/100 or 1/10,000 chance that two consecutive positive results will occur. Hence, a lot which is contaminated at the 1% level which tests positive will seldom test positive by a retest. The “negative” retest is therefore misleading. 2. Another reason for the inadequacy of retesting is because microbial populations may be reduced to undetectable levels between the time testing has begun on the first sample and the time testing is done on a sample from the same lot. Consequently one should not question a “positive” result on the basis of a negative retest, but rather on the basis of good laboratory practices (GLPs). For a discussion of GLPs see Smittle and Okrend (2001) and Miller (1987).

8.4.2 Resolution of Questionable Results Through GLPs – Questions to Ask Questions to ask the performing laboratory before accepting a positive result can include, but are not necessarily limited to, the following (as described in more detail in 6.2.2.2): 1. Was the same organism found in the laboratory environment? Good laboratories will routinely test their environment and seek to control the microbial ecology therein. 2. Was the organism found on any of the laboratory worker’s hand samples? Good laboratories will require that their technicians test their hands for the presence of the invasive pathogen before and after pre-enrichments, transfers, and streaking. 3. Is the recovered microorganism the same as the laboratory’s positive control? If so, and if the positive control strain is rarely, if ever, found in foods, then there is evidence that cross contamination may have occurred. 4. Was the recovered microbe found in more disparate samples or samples from other suppliers? This may suggest a cross-contamination event in the laboratory. 5. Did the laboratory assay’s negative control give a positive reaction? Negative control microorganisms which test properly (i.e., negative) on the medium or assay of choice can be used to validate the medium’s or assay’s positive results obtained from other samples, respectively. However, should a negative control test “positive,” all positive results from other samples tested at the same time are questionable. It is likely that ISO 17025 laboratories will be able to answer these questions (American Society for Quality Assurance, 2002). In the event that evidence is found of potential false-“positive” data in the laboratory, a review of the specifics of each assay with the appropriate technician(s)

142

J.L. Kornacki

should be considered. It may become necessary in such events to retest the sample in a statistically relevant (e.g., extremely rigorous) manner.

8.5 Microbiological Criteria and Specifications There are three types of microbiological criteria. These are (1) standards, (2) guidelines, and (3) specifications. Standards are legal and mandatory requirements established by regulatory bodies and usually pathogen oriented. Guidelines are advisory. Regulators and/or industry use guidelines to determine if hygienic provisions have been met (GMPs or CCPs). These may include organisms of no public health significance. However, specifications are purchase requirements set by industry.

8.5.1 Establishing Specifications Food processors should establish appropriate specifications for their incoming ingredients. This should occur whether or not the final product is subjected to a microbicidal treatment since post-process contamination may occur from microbes which are introduced into the factory environment via ingredients. Microbiological specifications should always include five key elements. They should (1) identify the food or ingredient to be tested, (2) specify the contaminant of concern, (3) indicate the analytical method (e.g., FDA BAM Salmonella) to be used, (4) define the sampling plan (e.g., n = 15 samples per lot – or 375 g composite of 15 samples – per FDA BAM category III, if appropriate), and (5) note the microbiological limit appropriate to the food and consistent with the sampling plan (e.g., “negative in 375 g”). 8.5.1.1 Common Errors in the Establishment of Microbiological Specifications These errors are often associated with key elements (4) and (5), the sampling plan and the microbiological limit, respectively. Sampling plan inadequacies usually have to do with taking an insufficient number of samples without regard to accepted statistical sampling plans. An example of an error related to limits occurs when a specification of “negative” occurs without any indication of the gram quantity tested. Hence “negative” could mean negative in a 1 g sample or some other quantity. The vagueness of such a specification can lead to a misleading sense of confidence in the safety of a product. However, “negative in 1,500 g per FDA BAM category I Salmonella sampling plan” is not only a defined approach but also associated with a specific probability of finding a microbe if it is randomly distributed (and randomly sampled) in a product. The phrase “negative in 1,500 g” means that a product meeting that specification has 95% chance that it has <1 cell in 500 g under the conditions described above. It should be pointed out, however, that a negative Salmonella assay

8

How Many Samples Do I Take?

143

in 1,500 g also could represent a sample in which 1 Salmonella per 750 g of sample may exit. A product which has one positive in 2,300 g also has a 95% change that <1 cell per 500 g is present (National Research Council, 1985). Thus, it is important to recognize that the presence of a “negative” does not prove the absence of the microbe.

8.6 Types of Sampling Plans There are essentially two types of sampling plans. These are “attributes” and “variables” sampling plans. Attributes sampling plans do not require knowledge of the distribution of microbes in a product, unlike variables plans which generally assume these are distributed in a log normal fashion.

8.6.1 Attributes Plans There are two types of attributes plans. The first is a one-class attribute sampling plan and the second is a two-class attribute sampling plan. One-class plans are typically used for pathogenic bacteria wherein the detection of any is unacceptable. Two-class plans are typically used with indicator organisms or those pathogens which require some quantity to initiate illness (e.g., Bacillus cereus, Clostridium perfringens).

8.6.2 ICMSF Plans The International Commission for the Microbiological Specifications for Foods (ICMSF) has developed plans based upon three broad product-related attributes, which either decrease, not affect, or increase risk, and five levels of pathogen or toxin severity. Thus, there are 15 cases arrayed as a grid with three column and five rows (see Fig. 8.4). Risk associated due to extrinsic (e.g., handling, environment, temperature, see Chapter 4 for examples) and intrinsic factors (e.g., pH, water activity, see Chapter 5) associated with the product increases from left to right across the product attributes, whereas virulence of the microbe (see Chapter 2)/toxin increases as one moves from top to bottom. The ICMSF specifies how many samples one should take per lot as represented by “n” and the number of these “c” allowed between “m” and “M,” where M is the absolute limit of tolerance above which rejection of the lot must always occur and “m” is an intermediate value. Thus, “c” specifies how many marginally acceptable samples one can have per lot. The ICMSF does not specify values for “m” or “M.” In the case of pathogens which require a certain level to create historical risk of illness, “M” can be set to a number equal to or somewhat below the risk level (e.g., < 100,000/g for B. cereus in many products). Naturally if one is targeting a high-risk group, such as neonates or immunocompromised individuals, a lower number may be chosen for “M.”

144

J.L. Kornacki

Fig. 8.4 ICMSF attributes sampling plans. n = number of samples per lot; c = maximum number of samples allowed between m and M

A rule of thumb for the value “m” is that this number can be set equal to the mean of the log10 CFU/g as determined from product testing. In instances where nonpathogens (e.g., indicator or spoilage organisms) are to be tested, the value of M can be set equal to 2 or more standard deviations above the mean log10 CFU/g or ml of product.

8.6.3 FDA Plans The FDA plans for Salmonella, in effect, cover three levels. FDA category II foods are those which are not subjected to a bactericidal process between sampling and consumption. Category I foods share the same definition with the addition that they are intended for consumption by high-risk populations. Category III foods are those which are to receive a bactericidal treatment between sampling and consumption (e.g., dry soup mixed to be hydrated and boiled before consumption). In these instances, the FDA has recommended that 60 × 25, 30 × 25, and 15 × 25 g samples are to be tested for foods in categories I, II, and III, respectively. However, the FDA recognized the legitimacy of compositing the randomly selected samples of a given production lot into 375 g quantities. Thus, FDA category I, II, and III foods could be tested at multiples of 375 g equivalent to 4, 2, and 1, respectively (Andrews and Hammack, 2001). It should be pointed out that these plans were designed to be used for “investigational purposes” but have become standard operating procedure for many food companies. This author is aware of situations

8

How Many Samples Do I Take?

145

where multiples of category I testing have even been used by government and industry alike. The FDA has a compositing approach for testing food for Listeria monocytogenes (Hitchens, 2003). In this approach one takes 10 × 50 g (or ml) subsamples from the lot which is divided into two sets each of 5 × 50 g samples. These are combined into two 250 g samples and stomached in 250 ml of buffered Listeria enrichment broth. Then 50 g (or ml) quantities from each of these two preparations are added to two separate 200 ml BLEB enrichments and the assay continued. The assay, when preformed in this manner, is effectively on two 25 g (ml) portions of the sample.

8.6.4 Variables Plans Variables plans require that one know the frequency distribution of microbes in a given product. This is usually assumed to be “log10 normal.” This is a disadvantage of these plans as compared to the attributes plans which do not have a requirement of knowing the distribution of microbes in a product. Variable plans have been used by the military (MIL standards) are considered to be applicable to products produced from a common source, under uniform conditions. More information about variables plans can be found in Midura and Bryant (2001) and in Smoot and Pierson (1997).

References American Society for Quality Assurance (2002) ANSI/ISO 17025. In: American National Standard: General Requirements for the Competence of Testing and Calibration Laboratories. Quality Press, Milwaukee Andrews WH, Hammack TS (2001) Salmonella. In: FDA/CFSAN (eds) Bacteriological Analytical Manual (Online) Updated April 2003, September 2005 and December 2005. http://www.911emg.com/pages/library/FDA%20Bacteriological%20Analysis.pdf. Accessed 29 July 2008 Curiale M (2000) Validation and the use of composite sampling for Listeria monocytogenes in ready-to-eat meat and poultry products. Research report prepared for the American Meat Institute Foundation. http://www.fsis.usda.gov/PDF/Seminar_Listeria_Sample_Composite_ Validation.pdf. Accessed 25 July 2008 Evanson D, McGiver JDE, Richter E, Decker SJ (1992) Comparison of 25 gram samples and 375 gram samples (composites of 15–25 gram subunits) for the detection of Listeria in dairy products. Paper presented at IFT Annual Meeting, New Orleans, 20–26 June 1992 Hitchens AD (2003) Detection and enumeration of Listeria monocytogenes in foods, Chapter 10. In: FDA/CFSAN (eds) Bacteriological Analytical Manual Online. http://vm.cfsan. fda.gov/∼ebam/bam-10.html . Accessed 25 July 2008 Kornacki JL (December 1999/January 2000) Environmental control programs: The nuts and bolts of food safety. Food Testing Anal 5(6):18–22 Luchansky JB, Porto ACS, Wallace FM, Call JE (2002) Recovery of Listeria monocytogenes from vacuum-seal packages of frankfurters: Comparison of the U.S. Department of Agriculture

146

J.L. Kornacki

(USDA) Food Safety and Inspection Service (FSIS) product composite enrichment method, the USDA Agricultural Research Service (ARS) product composite rinse method, and the USDA/ARS package rinse method. J Food Prot 65:567–570 McMahon WA, McNamara AM, Schultz AM (2005) Validation and use of composite sampling for the detection of Listeria monocytogenes in a variety of food products. Poster presented at the International Association for Food Protection, Baltimore, MD, August 17 Midura TF, Bryant RG (2001) Sampling plans, sample collection, shipment, and preparation for analysis, Chapter 2. In: Downes FP, Ito K (eds) Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC, pp. 13–23 Miller JM (1987) Quality control in microbiology. In: CDC Laboratory Manual. US Department of Health and Human Services, Public Health Service, Centers for Disease Control, Atlanta National Research Council (1985) An evaluation of the role of microbiological criteria for foods and food ingredients. National Academic Press, Washington, DC Smoot LM, Pierson MD (1997) Indicator microorganisms and microbiological criteria. In: Doyle MP, Beuchat LR, Montville TJ (eds), Food microbiology fundamentals and frontiers. ASM Press, Washington, DC, pp. 66–80 Smittle RB, Okrend AJ (2001) Laboratory quality assurance, Chapter 1. In: Downes FP, Ito K (eds) Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC Wallace FM, Call JE, Luchansky JB (2003) Validation of the USDA/ARS package rinse method for recovery of Listeria monocytogenes from naturally contaminated, commercially prepared frankfurters. J Food Prot 66:1920–1973

Chapter 9

When Can I Start Up My Factory or Processing Line Again? Jeffrey L. Kornacki

Abstract This chapter stems from personal experiences assisting companies after plant or line shuts down in response to contamination events and FDA- or USDAassociated recalls. Questions addressed include “How does one know that the problem has been eliminated?” “When am I allowed to start up?” “Should I hold the product and test and, if so, for how long?” “What unique activities are helpful on the first (and subsequent) day(s) of start-up to ensure a contamination-free product?” Gaining confidence that one can start up a previously shutdown line will come from thorough review and validation or revalidation of a variety of programs including the Hazard Analysis Critical Control Points (HACCPs) plan, ingredient specifications, sanitation standard operating procedures (SSOPs), and good manufacturing practices (GMPs). In addition, finding and eliminating microbial growth niches for the pathogen of concern, especially those associated with product contact or near product contact surfaces, will provide some assurance of safety.

9.1 Introduction This chapter stems from personal experiences assisting companies after plant or line shuts down in response to contamination events and FDA- or USDA-associated recalls. Others who have been through this process may have additional thoughts to contribute in future editions of this book and we invite their wise counsel to the benefit of all. If one’s factory or a processing line in the factory is shut down due to microbiological concerns a number of questions will spring to mind that need to be asked and answered after the investigation is complete. These include “How does one know that the problem has been eliminated?”; “When am I allowed to start up?”; “Should J.L. Kornacki (B) Kornacki Microbiology Solutions, Inc., McFarland, WI, USA e-mail: [email protected]

J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_9, 

147

148

J.L. Kornacki

I hold the product and test and, if so, for how long?”; “What unique activities are helpful on the first day of start-up to ensure a contamination-free product?”

9.2 Principles and Suggestions 9.2.1 How Can One Be Sure the Problem Is Eliminated? The short, though frustrating answer is “One cannot be completely sure.” It is simply impossible to prove a negative. The fact that one cannot observe (or sample) what they failed to see (or sample), along with the fact that no amount of testing (except 100% of the product with a perfect test method) will be adequate to guarantee an absolutely safe product (see Chapter 8) means that companies need to be (1) diligent in their investigation, (2) effective in their corrective actions, and (3) cautious at start-up. Much of this chapter is devoted to providing practical suggestions in this regard. Gaining confidence that one can start-up a previously shutdown line will come from thorough review and validation or revalidation of a variety of programs including the Hazard Analysis Critical Control Points (HACCPs) plan, ingredient specifications, sanitation standard operating procedures (SSOPs), and good manufacturing practices (GMPs), and others. In addition, finding and eliminating microbial growth niches for the pathogen of concern, especially those associated with product contact surfaces, will provide some assurance of safety.

9.2.2 Revalidate HACCP and CCPs A detailed discussion of HACCP is beyond the scope of this book and many excellent resources exist (e.g., Mortimore and Wallace, 1998; ICMSF, 1988; NACMCF, 1997). However, there are seven key principles in any HACCP plan. These are (1) hazard analysis, (2) critical control points (CCPs), (3) critical limits (CLs), (4) monitoring procedures, (5) corrective actions, (6) verification procedures, and (7) record keeping and documentation procedures. A few comments about these will follow. 9.2.2.1 Hazard Analysis The purpose of the hazard analysis is to determine potential hazards which present a reasonable risk to the consumer if not controlled. The concept of reasonable risk is reminiscent of the words of my former graduate advisor, the late Emeritus Professor Marth (1979) who wrote, “Because of the many ways by which a given food could become unsafe for a given consumer, it is impossible for the food industry or for regulatory agencies that control the food industry to guarantee that all foods will be completely safe for all consumers at all times. Consequently those persons working

9

When Can I Start Up My Factory or Processing Line Again?

149

in this field should invest their time and resources in those areas where the hazards are likely to be greatest and hence where such efforts are likely to benefit the health and welfare of the greatest number of people.” There are two phases of hazard analysis. The first phase involves identification of the following: ingredients, activities involved in each step of the process, equipment used, final product and how it is stored and distributed, its’ intended use, and consumers. The second phase is where the severity and likelihood of occurrence of the hazard(s) are evaluated. The National Advisory Committee HACCP document (1997) contains a series of questions regarding ingredients, intrinsic factors of the food during and after processing, processing procedures, microbial content of the food, factory and equipment design and use, and packaging when conducting a hazard analysis. Determination of potential microbiological hazards will require a thorough review of the literature. Experts in food safety microbiology may also be very helpful at this point. A properly done hazard analysis should lead readily to identification of CCPs. 9.2.2.2 Establish CCPs A particular process or procedure that is necessary to prevent, eliminate, or control a hazard to an acceptable level is called a CCP (NACMCF, 1997). 9.2.2.3 Establish CLs A critical limit (CL) is a maximum and/or minimum value to which a biological, chemical, or physical parameter must be controlled at a CCP to be effective. Implicit in the design of a HACCP plan is the validation of effective CLs. These should be established based on sound science not tradition, guess work, or extrapolation from inappropriate matrices. Revalidation of the entire HACCP plan and CCPs should be done before start up of operations (see Chapters 6 and 7). Most HACCP plans will not require more than three CCPs. The hazard analysis identified biological, chemical, or physical agents reasonably likely to cause illness or injury if not controlled as hazards (NACMCF, 1997). Therefore, deviations outside of a CL indicate that a product has been produced under unsafe operating conditions at that particular CCP. 9.2.2.4 Establish Monitoring Procedures Monitoring of CCPs is ideally continuous and performed in real time using appropriately validated measures. An example includes thermometers which should be calibrated to appropriate standards at the proper frequency. Monitoring does several things including the following: (1) It allows one to track the operating parameters related to the CCPs. If there is an observed trend away from control, then corrective action can be taken before the CLs are exceeded and the CCP is out of control; (2) Monitoring provides information as when a CCP was out of control; and (3) It provides a record which is essential to review in the verification step. It

150

J.L. Kornacki

is also critical to assign responsibility for monitoring at each CCP by appropriately training individuals. These individuals have knowledge of what to do when a process is trending toward an out-of-control conditions and report deviations from CLs promptly. The individuals responsible for monitoring CCPs must date and sign (or initial) all records and documents associated with CCP monitoring. 9.2.2.5 Establish Corrective Actions (CAs) It is important to recognize that exceeding a CL at a CCP is evidence that a product is a potential health risk. This fact demonstrates the importance of appropriate corrective actions in such an event. Consequently, it is essential to establish clearly well-defined corrective actions (CAs). CAs defined by phrases such as “see manager” are inappropriate. CAs should be well thought through in non-crisis times such that when deviations from CLs occur effective, prompt, documented corrective action can be taken. Several elements should be included in each CA. These include (1) determine and correct the cause of the deviation, (2) describe the disposition of the product, and (3) record the CAs taken. Specificity in what is to be done is very important and includes (1) defining what is done when a CL is exceeded, (2) defining who is responsible to implement the CA(s), and (3) establishing that a record will be developed and maintained of the CAs taken. Describing the location of such records and the responsibility for maintaining such records is also recommended. 9.2.2.6 Establish Verification Procedures Verification includes (1) all non-monitoring activities that determine the validity of the HACCP plan and conformance of the HACCP system to that plan. Hence, verification that the plan is being followed and a review of CCP monitoring and CA records is one aspect of verification. Thus, the plan should be audited routinely and after a change in a product formula or a new product. Presently, monitoring CCPs using bacterial test results cannot be done in real time. Hence, such approaches are typically discouraged in that context. It may be possible that one day real-time biosensors will make this possible. However, statistically based finished product testing can be used and is often recommended as verification of the efficacy of the HACCP plan to control the pathogen of interest. An unbiased independent party should comprehensively verify the HACCP plan on a periodic basis. This audit should include a critical review of the hazard analysis, all the elements of the HACCP plan, an on-site verification that all process flow diagrams correspond to the operation, and all appropriate records. The HACCP plan will need to be modified in the event that deficiencies are found. Validation. Another aspect of verification includes validation that the HACCP plan is (1) technically sound from a scientific standpoint and (2) all hazards reasonably likely to occur have been accounted for and will be effectively controlled by such a plan. The information needed for such a validation may include expert advice, scientific studies and/or in-plant observations, measurements, and evaluations.

9

When Can I Start Up My Factory or Processing Line Again?

151

Validation of the HACCP plan and CLs is so important that some have proposed it as an eighth HACCP principle (Sperber, 1999). Validation of critical limits (CLs) for CCPs of HACCP plans can be done with surrogate microorganisms as described in Chapter 7 or with time temperature indicators and other approaches. The USDA and FDA have acknowledged the use of surrogate microorganisms as a valid approach, where appropriate (Engeljohn, 2004; Eblen et al., 2005; Annous and Kozempel, 1998; Annous et al., 1999; FDA, 2000). Surrogate microorganisms are harmless microbes with correlated survival and growth parameters to specific pathogens. An advantage of using non-pathogenic surrogate microbes is that they can be used under actual factory or pilot plant conditions to validate the efficacy of a process without the need to simulate it in a laboratory. Other approaches may include controlled laboratory challenge studies or thorough literature searches. Be exceedingly cautious not to accept validation of CLs based on notions that “no body ever got sick from our product before,” extrapolations from different food matrices (see Chapter 5), or tradition (Ma et al., 2007). It may be appropriate to consult with an expert on a case-by-case basis to ensure the use of a surrogate is appropriate and properly used in the validation trials. Revalidation of the HACCP plan should occur under certain circumstances, e.g., when there are unexplained system failures, significant product, process, or packaging changes, or new hazards are identified. 9.2.2.7 Establish Record Keeping and Documentation Procedures Documentation of the HACCP plan, the HACCP team members, hazard analysis, description of the food, its distribution, intended use and consumer, CCPs, CLs, monitoring procedures, CAs, process flow diagram, test records, and verification is important and will be reviewed in any proper HACCP verification audit.

9.2.3 Ingredients Microbiological criteria for ingredients should be reviewed and revised or updated as necessary. Proper microbiological criteria include a listing of the identity of the food or ingredient (e.g., potato flakes), the contaminant of concern (e.g., Salmonella spp.), the analytical method (e.g., current FDA Bacteriological Analytical Method), the sampling plan (e.g., ICMSF plan case 13 or FDA category I), and the microbiological limits appropriate to the food and consistent with the sampling plan (National Research Council, 1985). It may be useful to risk rank the ingredients with regard to those intrinsic or extrinsic factors that result in ingredients of high, medium, or low risk. Ingredients with high aw and high pH which are not sterilized in their container and which are blended into a product without further microbicidal treatment or refrigeration would clearly represent a high risk. One could plot the ingredients across a relative risk ranking (e.g., 1–10 on the y scale with 10 being the highest risk) vs. risk associated with supplier’s externally performed good manufacturing practice (GMP) audit

152

J.L. Kornacki

Fig. 9.1 Risk associated with product vs. risk associated with external GMP audits

scores (see Fig. 9.1). Suppliers and their products can be ranked in this way. Those falling into zone 3 would be the highest risk (Fig. 9.1).

9.2.3.1 Risk Associated with GMP Score Suppliers falling into zone 3 risk should be visited at an appropriate frequency and a microbiological in-factory risk assessment performed (see Chapter 4).

9.2.3.2 Sanitation Standard Operating Practices (SSOPs) SSOPs should be reviewed critically as should their correspondence with what actually occurs. Cleaners and sanitizers serve different purposes. For example, chlorinated alkaline cleaners are sometimes confused with chlorine-based sanitizers. The purpose of a cleaner is to remove soil, whereas the purpose of a sanitizer is to reduce bacterial populations. The principle germicidal form of aqueous chlorine is hypochlorous acid (HOCl). This is because the molecule is uncharged and thus can pass more readily through the charged cell membrane (see Chapter 5). The majority of the chlorine is not in the germicidal form but rather as the hypochlorous acid anion (OCl–) in chlorinated alkaline cleaners due to their high pH. This is one reason why it is important to use both cleaners and sanitizers for their intended function. It follows that chlorine-based sanitizer should be used at the manufacturers recommended concentration rather than trying to increase the concentration. Increasing the concentration will increase the alkalinity of the solution resulting in less chlorine in the germicidal form.

9

When Can I Start Up My Factory or Processing Line Again?

153

Rotation of sanitizers is a prudent idea because otherwise resistant microbial populations may build up. The idea is that one would use one type (e.g., chlorinebased) sanitizer for several days/weeks and then switch to another type of sanitizer. However, some have misinterpreted the concept of sanitizer rotation to mean the one immediately follows one sanitizer with another in the same treatment. This is potentially dangerous and likely to be ineffective. For example, mixing chlorine-based sanitizer with acidic sanitizer (e.g., acidic iodine sanitizer) may result in liberation of toxic gas.

9.3 When Am I Allowed to Start Up? In the author’s experience permission from outside parties is rarely gained but rather determined and announced by the plant once they are confident that the operation can be started with a minimum of risk to the consumer. However, great caution should still be taken. The following may seem technically counter to the principles of HACCP with its reduced emphasis on finished product testing, but these have been helpful in this author’s experience.

9.4 Hold and Test Considerations and Unique Line Start-Up Considerations Let’s say you have been diligent in your investigation and the plant has made effective corrective actions that you believe that you are likely to prevent recurrence of the contamination event. However, as stated earlier, one cannot be completely sure they have eliminated all the potential sources, thus it remains to be very cautious at start-up. The last thing anyone wants is to release contaminated product to the consumer, if for some reason the problem was not completely resolved. Consequently a hold and test program for finished processed product is generally recommended.

9.4.1 Sampling the First Product Off the Line In the event that some of the sources of the contaminant have been missed in the investigation and still reside in the system, it is often useful to extensively sample the first product produced on the processing line in question. One should do this in a manner likely to find a pathogen. This is no time for routine measures. Hence, taking a single 25 g sample is likely not to be effective (see Chapter 8). However, one could consider taking multiple large (e.g., 375 g in the case of Salmonella) grab samples at start-up using the theory that the flow of product may dislodge microbes resident in the processing lines and closely associated areas that may have not been adequately sanitized. These samples can then be appropriately handled, taken to

154

J.L. Kornacki

a suitable laboratory, aseptically subdivided into appropriate analytical units, and enriched for the pathogen of concern.

9.4.2 Sampling the Lot In addition the facility should seriously consider testing randomly selected 25 g samples throughout the lot for appropriate compositing. If a composite sampling scheme has been approved for the microbe of concern (e.g., Salmonella) then sampling at a multiple of the FDA BAM recommended levels (e.g., I, II, or III) may be appropriate until one gains confidence that the product is free of the pathogen. Several food processors that I have worked with have utilized 4–5 times the FDA BAM sampling level across several successive clean-up to clean-up product start-ups. Please note: all products produced from such processing lines sent out for pathogen testing should be held by the plant under appropriate conditions until the results of all pathogen tests are complete. Clearly only appropriately defined lots (e.g., clean-up to clean-up) should be released assuming all appropriate product and sanitation tests have been conducted and found to be within acceptable levels. Any set of product(s)/lots produced from the same line from “clean-up to clean-up” which test positive should be held, and in this case the entire line should be shut down for further investigation and corrective actions. Extensive finished product testing for pathogens can be backed away from over time (e.g., multiple lots) as the plant gains increasing confidence that the system is under control. 9.4.2.1 The Use of Indicator Organisms and Statistical Process Control It seems prudent and relatively inexpensive that product could be sampled and monitored at selected times throughout the day and from lot to lot for non-pathogenic indicator bacteria (e.g., APC, Enterobacteriaceae, etc.) as appropriate. These data could be tracked on a graph using the principles of statistical process control (e.g. Evans and Lindsay, 1996). This is not a real-time measurement. However, this provides data that can be responded to when it trends toward being out of control. Responses could include increasing the number and frequency of environmental samples (and appropriate documented corrective actions) in the context of initiating an investigation as to the cause of the trend.

References Annous BA, Kozempel MF (1998) Influence of growth medium on thermal resistance of Pediococcus spp. NRRL B-2354 (formerly Micrococcus freudenreichii) in liquid food. J Food Prot 61:578–581 Annous BA, Kozempel MF, Kurantz MJ (1999) Changes in membrane fatty acid composition of Pediococcus sp. strain NRRL B2354 in response to growth conditions and its effect on thermal resistance. Appl Environ Microbiol 65:2857–2862

9

When Can I Start Up My Factory or Processing Line Again?

155

Eblen DR, Annous BA, Sapers GM (2005) Studies to select appropriate nonpathogenic surrogate Escherichia coli strains for potential use in place of Escherichia coli O157:H7 and Salmonella in pilot plant studies. J Food Prot 68:282–291 Engeljohn D (2004) Regulatory Perspective of Validation and Verification Activities. Presented in symposium S03, Validation and verification of pathogen interactions in meat and poultry processing, International Association for Food Protection Annual Meeting, Phoenix, 8–11 August 2004 Evans JR, Lindsay WM (1996) Fundamentals of statistical process control, Chapter 15. In: Evans JR, Lindsay WM (eds) The Management and Control of Quality, 3rd edn. West Publishing Company, New York, pp. 639–698 FDA (2000) Kinetics of microbial inactivation for alternative food processing technologies. http://vm.cfsan.fda.gov/∼comm/ift-toc.html. Accessed 25 July 2008 ICMSF (1988) HACCP in microbiological safety and quality. In: Silliker JH, Baird-Parker AC, Bryan FL, Christian JHB, Roberts TA, Tompkin RB (eds) Microorganisms in Foods, Volume 4. Blackwell Scientific Publications, Boston Ma L, Kornacki JL, Zhang G, Lin CM, Doyle MP (2007) Development of thermal surrogate microorganisms in ground beef for in-plant critical control point validation studies. J Food Prot 70(4):952–957 Marth EH (1979) Food safety: Can it be achieved. Presented at the Spring Meeting of the Central States Association of Food and Drug Officials, Madison, WI, April 25, 1979. Reprinted from the Quarterly Bulletin of the Association of Food and Drug Officials, Volume 44(1) January, 1980, ISSN:0195-4865, pp. 12–21. Mortimore S, Wallace C (1998) HACCP: A Practical Approach, 2nd edn. Aspen Publishers, Gaithersburg, MD NACMCF (1997) Hazard analysis and critical control point principles and application guidelines: Adopted August 14th, 1997. http://vm.cfsan.fda.gov/∼comm/nacmcfp.html. Accessed 25 July 2008 National Research Council (1985) An Evaluation of the Role of Microbiological Criteria for Foods and Ingredients. National Academic Press, Washington, DC Sperber WH (1999) Thoughts on today’s food safety. . .The role of validation in HACCP plans. Dairy, Food Environ Sanit 19(12):920

Chapter 10

Value and Methods for Molecular Subtyping of Bacteria Mark Moorman, Payton Pruett, and Martin Weidman

Abstract Tracking sources of microbial contaminants has been a concern since the early days of commercial food processing; however, recent advances in the development of molecular subtyping methods have provided tools that allow more rapid and highly accurate determinations of these sources. Only individuals with an understanding of the molecular subtyping methods, and the epidemiological techniques used, can evaluate the reliability of a link between a food-manufacturing plant, a food, and a foodborne disease outbreak. In principle, the goal of molecular subtyping methods is to compare the genetic material of two or more bacterial isolates to determine whether they have shared a recent common ancestor. The chapter addresses some of more commonly applied subtyping methods including pulsed field gel electrophoresis (PFGE), ribotyping, PCR methods applied to fragment length polymorphisms (RAPD and REP-PCR), DNA sequencing-based subtyping, and other characterization methods. This chapter also includes case studies. In preparing for potential emergencies, food companies may consider adding an outside expert in molecular subtyping to their emergency response team.

10.1 The Value of Molecular Subtyping Methods for the Food Industry Tracking sources of microbial contaminants has been a concern since the early days of commercial food processing. However, recent advances in the development of molecular subtyping methods have provided tools that allow more rapid and highly accurate determinations of microbial contamination sources. These techniques have improved the food industry’s ability to track the spread of foodborne pathogens and spoilage microorganisms along the food chain with much greater effectiveness than M. Moorman (B) The Kellogg Company, Battle Creek, MI, USA e-mail: [email protected] J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0_10, 

157

158

M. Moorman et al.

ever before. In criminal forensics, similar genetic methods have provided a newer DNA-based tool for individuals falsely incarcerated for crimes often after many years in prison. Others may be indicted based on findings (such as the presence of a blood stain) that would have been insufficient evidence years ago. Similarly, the evidence that can be used to implicate a food processor as the source of human disease cases is different now than in years past. For example, it is not uncommon for foodborne disease surveillance systems, such as PulseNet in the United States (Swaminathan et al., 2001), to ascertain that three to four patients in different states have been infected by the same, rare, subtype of Listeria monocytogenes, thus indicating a likely link of these cases to a common food source. Application of food consumption questionnaires and use of traditional epidemiological techniques (i.e., case–control studies) can subsequently define a suspect food source, such as sliced deli meat or soft cheese. From there, it may only be a short period until a molecular subtype of L. monocytogenes is found in the product or even within the environment of a plant producing the food item linked to these initial cases. A call from a federal or state agency announcing that products manufactured by a given plant have been linked by “molecular subtyping” to human listeriosis cases then becomes a company’s worst nightmare. Only individuals with an understanding of the molecular subtyping methods, and the epidemiological techniques used, can evaluate the reliability of a link between a food-manufacturing plant, a food, and a foodborne disease outbreak. This chapter has been written to provide an overview of the application of molecular subtyping methods in areas relevant to the food industry. While the casual reader will surely not become an expert in molecular subtyping after reading this he or she will be able to better understand some of the methods used and their value and limitations. In addition, this chapter will provide some guidance of how plant personnel with food safety and food microbiology responsibilities can prepare themselves and their companies for D-day . . . Are you ready?

10.1.1 Limitations of Traditional Cultural Methods for Tracking Food-Associated Microorganisms – Why Do We Need Molecular Subtyping Methods? While traditional biochemical and serological methods have been used for many years to characterize bacterial isolates beyond the species level (e.g., to serotypes), these methods often do not permit determination of definitive relationships between isolates obtained from clinical specimens, foods, or manufacturing environments. For example, detection of an L. monocytogenes serotype 1/2a strain in a food implicated as a possible source for a human listeriosis outbreak (20 individuals) and from clinical specimens does not provide strong evidence for a causal relationship as more than 40% of food isolates are serotype 1/2a. The strength of this evidence would be similar to murder case where a blood spot found on the victim reveals blood group A, the same blood group carried by one of many potential suspects,

10

Value and Methods for Molecular Subtyping of Bacteria

159

including perhaps, a suspect who also has a poor motive and no other evidence linking him and her to the murder victim. While some highly discriminatory phenotypic subtyping methods (e.g., phage typing) can be used to provide stronger evidence, molecular subtyping methods, which may differentiate hundreds or more subtypes, have truly revolutionized our ability to conclusively track the sources and spread of food-associated microorganisms. This chapter will focus on the application of molecular subtyping methods, excluding phenotypic methods such as serotyping, phage typing, antimicrobial susceptibility typing, or fatty acid analysis. Readers interested in more information on phenotypic subtyping methods are referred to one of the numerous reviews and book chapters on this topic.

10.1.2 Evolution of Molecular Subtyping Methods Phenotypic as well as molecular subtyping methods used in food microbiology can trace their origins to different sources and many of these have evolved from tools used in the bacterial population genetics (e.g., multilocus enzyme electrophoresis, multilocus sequence typing), clinical diagnostics, basic genetic and molecular biology and/or from subtyping methods developed for various other applications, ranging from characterization of environmental or clinically important pathogens to genetic characterization of mammalian and eukaryotic individuals. Evolution of methods progressed from methods with poor standardization and reproducibility, often plagued by limited discrimination between strains, to highly standardized and reproducible methods with high levels of discriminatory power. In food microbiology, molecular subtyping was initially primarily used to characterize foodborne pathogens, but these methods are also increasingly applied by the food industry to track sources of spoilage organisms with the goal to implement controls to increase product shelf life and decrease flavor defects (see Chapter 3 for a discussion of bacterial spoilage organisms). Federal agencies have played an important role in developing and implementing standardized molecular subtyping methods for surveillance of infectious disease, with a major focus on foodborne infectious diseases. The most well-known example of these types of efforts is “PulseNet” (Swaminathan et al., 2001), which started as a network of selected state health department and CDC laboratories performing standardized PFGE typing for Escherichia coli O157:H7. Some in the food industry use either in-house or contract laboratory expertise to perform subtyping using similar or the same PFGE methods applied by PulseNet laboratories. However, a wider variety of different subtyping methods (e.g., REP-PCR, ribotyping) can and are used by the industry to track sources and transmission of foodborne pathogens and spoilage microorganisms. Most of these methods have gone through the same evolution as PFGE though and have been developed into more standardized, reproducible, and discriminatory methods than when first implemented. Almost all currently used molecular subtyping methods in food microbiology still require initial isolation of a pure bacterial culture for subsequent subtype characterization. The use of 16s rDNA sequencing might be considered by some as an exception, since some recent publications (Randazzo et al., 2002) have used direct

160

M. Moorman et al.

amplification of 16S rDNA sequences from food samples to identify culturable and unculturable bacterial species and subtypes present. However, sequencing of this gene provides very limited subtype discrimination, due to the conserved nature of the 16S rDNA. It is highly likely that in the not-so-distant-future an increasing number of combined PCR-based detection and DNA-sequencing-based subtyping methods will be developed, which will allow for integrated detection and subtyping of specific target pathogens. The future is likely to bring further improvements in molecular subtyping approaches. This will enable the food industry to track the presence of subtype-specific gene sequences that will define pathogen or spoilage microorganism sources. These tools will have the added benefit of detection and subtyping within 6–12 h rather than the multiple days currently required.

10.2 Principles of Molecular Subtyping Methods In principle, the goal of molecular subtyping methods is to compare the genetic material of two or more bacterial isolates to determine whether they have shared a recent common ancestor. Even though subtyping of bacteria is often referred to as “DNA fingerprinting,” conceptually, bacterial subtyping is different from fingerprinting humans. In human fingerprinting, the goal is to track a single individual (“has suspect A touched this knife?”), while bacterial fingerprinting tries to establish recent common ancestry (“Are the bacteria found in the blood stream of a human closely related to bacteria isolated from a food sample implicated by being produced in the same plant and time period as the source of a human case?”). Correct interpretation of bacterial subtyping results thus requires a an understanding of evolutionary biology and population genetics. While the following section will provide a short overview of the different molecular methods, interpretation of subtyping results requires specific expertise often not found with many food manufacturers. In preparing for potential emergencies, food companies may consider adding an outside expert in molecular subtyping to their emergency response team. For this purpose, companies may elect to have a list of potential experts on file as part of their emergency preparedness plan. Since a variety of reviews and book chapters on the general topic of molecular subtyping (e.g., Wiedmann, 2002a; van Belkum et al. 2001; Olive and Bean, 1999; Woodford and Johnson, 2004) as well as on subtyping of specific pathogens (e.g., Wiedmann, 2002b; Wassenaar and Newell, 2000; Graves et al., 1999; Strockbine et al., 1998; Threlfall and Frost, 1990) have been published, the following summary will be brief and the reader is referred to the appropriate references for more detailed information.

10.2.1 Overview of Molecular Subtyping Genetic Techniques The emergence of more accessible, cheaper, and reliable methods for manipulating and sequencing DNA has been critical for the boom in molecular subtyping

10

Value and Methods for Molecular Subtyping of Bacteria

161

methods. Most importantly, the advent of PCR in 1985 and the subsequent refinement of this technique for DNA amplification have been critical for the evolution of molecular subtyping methods. More recently, the continued development of more rapid and inexpensive methods for DNA sequencing, culminating in the completion of the human genome as well as the genomes of more than 100 bacterial species, has had tremendous impact on the development and implementation of improved molecular subtyping methods. The reader interested in a more in-depth understanding of molecular methods may refer textbooks or WWW-based resources (e.g., http://www.microbeworld.org/htm/aboutmicro/tools/genetic.htm).

10.2.2 Overview of Molecular Subtyping Methods In an ideal world, the best way to determine whether two bacteria are closely related (i.e., closely enough to share a very recent, i.e., weeks or months, ancestor) would be to determine the full DNA sequences of their entire genomes. This is technically possible, but cost-prohibitive with current sequencing technologies and sizes of bacterial genomes ranging from 1 to 10×106 base pairs (bp). This approach can be, and has been, applied in situations where cost is not a driving factor. For example, full genome sequencing data have been used to determine that the different Bacillus anthracis isolates recovered during the 2002 “anthrax attacks” in the United States were nearly identical, indicating a common source (Read et al., 2002). For large-scale applications, such as detection of foodborne disease outbreaks and determination of outbreak sources, these approaches are not currently feasible. Rather, most molecular subtyping methods for microorganisms characterize either defined or random fragments of the genome for their relative similarity. All currently used molecular subtyping methods only characterize small parts of the overall genome for similarities or differences. Detection of genome similarities between organisms can be achieved by banding pattern-based methods, which characterize genome fragments through restriction digestion (e.g., PFGE, ribotyping), generation of PCR product length polymorphisms (e.g., RAPD, REP-PCR), or by methods which determine the actual DNA sequence present at one or more sites in the genome (i.e., DNA sequencing-based methods). Some of more commonly applied subtyping methods falling into these categories are discussed in more detail below. 10.2.2.1 Pulsed Field Gel Electrophoresis (PFGE) PFGE characterizes bacteria into subtypes (sometimes referred to as “pulsotypes”) by generating DNA banding patterns after restriction digestion of the bacterial DNA. Specifically, complete bacterial DNA is purified and subsequently cut into diagnostic DNA fragments using restriction enzymes, which cut DNA at specific short DNA sequences. For example the restriction enzyme AscI will cut the bacterial DNA whenever a sequence of GGCGCGCC is present in the bacterial DNA. Restriction enzymes are chosen such that they cut DNA only rarely to yield between approximately 8 and 25 large DNA bands ranging from 40 to >600 kb (Wiedmann, 2002a).

162

M. Moorman et al.

Since DNA fragments this large cannot be separated by standard gel electrophoretic techniques, a specific electrophoresis technique using alternating electric fields is used to separate DNA fragments based upon base pair length (Lai et al., 1989) which is visualized as a DNA banding pattern. The resultant DNA banding patterns for different bacterial isolates are compared to differentiate distinct bacterial subtypes from those that share identical (or very similar) DNA fragment patterns. Often, different restriction enzymes are used for PFGE typing of various foodborne pathogens. For example, while the restriction enzymes AscI and ApaI are often used for L. monocytogenes (Brosch et al., 1994), the restriction enzymes XbaI and BlnI have been used by the US Centers for Disease Control and Prevention (CDC) for discriminatory PFGE typing of E. coli O157:H7 (Breuer et al., 2001). PFGE of a given isolate is often performed using different restriction enzymes in separate reactions to achieve improved discrimination. For example, two L. monocytogenes isolates with identical PFGE types following exposure of their DNA to AscI restriction enzyme may have two distinct ApaI PFGE patterns. Selection of enzymes to be used for PFGE typing of a specific bacterial species is generally based on initial preliminary experiments to determine the most discriminatory enzymes producing easy to interpret and reproducible patterns. The cost of the enzymes may also play a role in the selection. The CDC and state health departments have developed a national network (PulseNet) to rapidly exchange standardized PFGE subtype data for isolates of foodborne pathogens (Swaminathan et al., 2001). PFGE subtyping shows a high level of discrimination for many foodborne bacterial pathogens and, thus, is often considered the current “gold standard” for strain discrimination. It is important to realize, however, that PFGE (as well as other subtyping methods) may detect small genetic differences (e.g., two to three different bands) that may not be epidemiologically significant (Tenover et al., 1995). On the other hand, the detection of an identical PFGE type (or a subtype determined by another method) in two samples (e.g., a food sample and a sample from a clinically affected human) does not necessarily imply a causal relationship between two isolates. In outbreak investigations, molecular subtyping information needs to be analyzed in conjunction with epidemiological data to determine causal relationships between two or more isolates. 10.2.2.2 Ribotyping Ribotyping is another DNA-based subtyping method in which bacterial DNA is initially cut into fragments using restriction enzymes. While PFGE uses restriction enzymes to cut bacterial DNA in very few large pieces (40–600 kb), the initial DNA digestion for ribotyping cuts DNA into many (>300–500) smaller pieces (approximately 1–30 kb). These DNA fragments are separated by size using agarose gel electrophoresis followed by Southern blot using DNA probes to specifically label and detect DNA fragments containing the ribosomal RNA (rRNA) bacterial genes. The resultant DNA banding patterns are thus based on those DNA fragments that contain the rRNA genes (Grimont and Grimont, 1986). A completely automated, standardized system for ribotyping (the RiboPrinter Microbial

10

Value and Methods for Molecular Subtyping of Bacteria

163

Characterization system) has been developed by Qualicon-DuPont (Wilmington, DE) and is commercially available (Wiedmann, 2002a; Bruce, 1996). 10.2.2.3 Polymerase Chain Reaction (PCR) Methods Generating Fragment Length Polymorphisms The most common PCR-based approaches generating fragment length polymorphisms can be used to group bacterial isolates into identical subtypes. These methods include random amplified polymorphic DNA (RAPD), amplified polymorphic (AP)-PCR, REP-PCR, and Enterobacterial Repetitive Intergenic Consensus (ERIC) PCR. While all of these methods can provide substantial discriminatory power, they often suffer from poor reproducibility between different days, operators, and laboratories. Generally, these procedures are not used to develop large databases (such as the PulseNet database); yet may be suitable for rapid characterization of a small number of isolates. RAPD and AP-PCR were both described at about the same time, albeit by two different research groups (Williams et al., 1990; Welsh and McClelland, 1990). Both methods use small PCR primers or larger primers at low annealing temperatures to facilitate random priming at multiple chromosomal sites. This generates multiple (random) PCR fragments which can be visualized by gel electrophoresis. Conversely, REP and ERIC PCR represent PCR-based methods that use PCR primers to amplify defined conserved regions, found in multiple copies within the bacterial chromosome, generating banding patterns after gel electrophoresis. While REP-PCR targets Repetitive Extragenic Palindromic sequences, which were initially described in E. coli and Salmonella, ERIC PCR targets amplification of ERIC sequences. The presence of multiple copies of these two repetitive sequence elements in bacterial genomes provides for generation of multiple PCR products of different sizes. Both methods originally were used to subtype Gram-negative organisms. However, suitable primers have been developed to allow application of these methods to Gram-positive pathogens. Kits and equipment for REP PCR are also commercially available through Bacterial Barcodes Inc. (Houston, TX). A good overview of both of these methods is available in a book chapter by Farber (2001); in addition, a comprehensive review on REP-PCR has been published by Kerr (1994). 10.2.2.4 DNA Sequencing-Based Subtyping and Characterization Methods DNA sequencing-based subtyping methods can be used to differentiate bacterial isolates by sequencing a single gene or multiple genes. Sequencing single genes often provides limited discrimination, but may provide a suitable approach for characterizing bacterial isolates into species. For example, while sequencing of the gene encoding 16S rRNA provides a highly suitable tool for characterizing bacterial isolates to the species level, the conserved nature of this gene and its limited sequence diversity limit the use of this tool for sensitive subtype discrimination. Multilocus sequence typing (MLST) refers to a highly discriminatory molecular subtyping approach that uses DNA sequencing of multiple genes or gene fragments

164

M. Moorman et al.

to differentiate bacterial subtypes, determine the genetic relatedness of isolates, and make inferences about a genetic parameter of interest in a population. While MLST often is used to sequence multiple housekeeping genes (Spratt, 1999) it is also a highly discriminatory tool for sequencing multiple virulence genes and as a subtyping method (Wiedmann, 2002a). A major advantage of DNA sequencing-based subtyping, as compared to other molecular subtyping methods, is that sequence data are considerably less ambiguous (Spratt, 1999) and easier to interpret than banding pattern-based subtype data obtained by many other DNA-based subtyping approaches (e.g., PFGE and ribotyping). DNA sequencing data also provide an opportunity to reconstruct ancestral relationships among bacterial isolates, allowing further insight into the evolution, epidemiology, and ecology of foodborne pathogens. The development of Internet accessible databases for MLST information (such as the MLST database at http://www.mlst.net/) will also facilitate global, large-scale surveillance, and tracking of bacterial foodborne pathogens as well as large-scale evolutionary studies (Spratt, 1999). Sequencing of housekeeping genes provides an opportunity to probe the evolutionary relatedness of bacteria without the confounding effects of adaptive selection often occurring among virulence genes. Sequencing of virulence genes, however, may provide critical information on the evolution of virulence characteristics among clonal groups of bacterial pathogens. In addition, studying the effects of adaptive selection in specific lineages or strains of foodborne pathogens may help explain variations in the ability of these clonal groups to cause disease. As DNA sequencing continues to become less expensive and more broadly accessible, and as DNA chip-based DNA sequencing methods are developed, the use of DNA sequencing data for bacterial subtyping and evolutionary studies will continue to expand.

10.2.3 Others Methods Recent advances in genomics and genomic technologies have led to the development of new subtyping approaches likely to be applied in the food industry. Technologies that most likely will translate into subtyping applications in the food industry include microarrays, full genome sequencing methods, and high-throughput single nucleotide polymorphism (SNP) detection methods. These techniques permit rapid and reproducible subtype characterization by detecting high numbers of genetic polymorphisms (including “DNA sequencing on a chip”), as well as detection of gene presence/absence polymorphisms (e.g., through use of genomic microarrays). In addition, full genome sequencing data will facilitate the development of other highly discriminatory molecular subtyping methods such as multilocus variablenumber tandem repeat analysis (MLVA) (Top et al., 2004; Keys et al., 2005), a technique recently used in the peanut butter outbreak investigation by the CDC (2009). While many of these methods may ultimately be more rapid and discriminatory than current methods (e.g., PFGE, ribotyping), their application and use will likely be similar.

10

Value and Methods for Molecular Subtyping of Bacteria

165

10.2.3.1 Importance of Combined Use of Molecular Subtyping and Epidemiological Data An understanding of molecular subtyping methods is important when investigating outbreaks or identifying sources of spoilage organisms in a food processing environment. It is also critical to understand the application of such techniques as part of epidemiological approaches to detect and define foodborne disease outbreaks (e.g., surveillance methods) and sources (e.g., case–control studies) during outbreak investigations. Epidemiological information typically first defines a suspect food source for an outbreak as subtype data alone can generally not be used to link an outbreak to a source. For example, strains of E. coli O157:H7 or L. monocytogenes with respective identical subtypes may be found in multiple processing plants and/or other environments. Thus, identification of a pathogen subtype in a plant, which matches a subtype implicated in a human outbreak, alone does not provide sufficient evidence that a plant is an outbreak source. Rather, an epidemiological link in addition to a subtype link is required to implicate a processing plant as an outbreak source. Conversely, strong epidemiological evidence in conjunction with less than perfect subtype evidence (i.e., isolation of a closely related PFGE type from patients in an outbreak and foods products with a strong epidemiological link) can be sufficient to conclusively implicate a specific food product as an outbreak source. As mentioned above, when PFGE typing is used as a subtyping method, two bacterial isolates (e.g., one from a food item and one from a patient) may be able to accumulate sufficient genetic changes during multiplication in a food and a human that they show slightly different PFGE types (i.e., as many as three bands difference or more according to Tenover et al., 1995, JLK, Editor) even if they share a recent common ancestor (e.g., they both originated from the one L. monocytogenes cell that was initially present on a conveyer belt).

10.3 Users of Molecular Subtyping Methods Application of molecular subtyping techniques has increased in today’s food testing laboratory. Innovative work by manufacturers and universities has aided the transition of molecular assays into government, commercial, and industry laboratories where they can now be employed more cost effectively and with greater ease. As many assays have become highly automated, laboratories are no longer required to maintain highly skilled personnel on staff to perform these procedures. However, just as with any methodology, appropriate good laboratory practices (GLP) are still required to ensure accuracy and reliability of results. In the last decade, many outbreaks, recalls, and in-plant investigations have utilized molecular subtyping techniques to pinpoint the ultimate source of microbial contamination. The remainder of this section will present two case studies (foodborne illness investigation and an in-plant investigation of a spoilage problem), illustrating the value and public health benefit of these molecular subtyping tools. It is important to emphasize that molecular subtyping methods have found other

166

M. Moorman et al.

applications relevant to the food industry, such as determination of laboratory cross contamination (e.g., Braden et al., 1997). The case studies included below are fictional, however, some may be partially based on true outbreak investigations and case scenarios, which were modified to provide maximum didactic benefits.

10.3.1 Foodborne Illness Outbreak Investigation – Salmonellosis from Raw Pecans During a 3-month period (June–August, 2002) 110 laboratory-confirmed cases of salmonellosis caused by Salmonella Enteritidis (SE) were reported in California, Oregon, and Washington. PFGE analysis of clinical SE cultures from these cases revealed that 95 human cases were caused by the same PFGE type (PFGE-type XY), as determined by PFGE with the enzyme XbaI. In the preceding 5 months (January through May), only nine cases of salmonellosis caused by SE were identified in the same three states; none of these cases were PFGE type XY. State health departments and CDC thus concluded that they had detected an on-going outbreak of salmonellosis due to a rare SE PFGE type. In order to determine an outbreak source, the three involved state health departments initiated a case–control study in October 2002. A total of 12 cases and 12 matched controls (i.e., persons that matched cases by age, gender, ethnicity, and place of residence) were enrolled in this case–control study. Analyses of the food questionnaire administered in this case–control study showed a strong association between infection with SE PFGE type XY and consumption of raw pecans (odd ratio = 18:1; 95% confidence interval = 2.8–17.3). In many responses, patients indicated specifically that they had eaten raw, prepackaged pecans purchased from either grocery stores or convenience markets, with brand F mentioned by 8 of the 12 cases and brand G mentioned by 5 of the patients (some patients mentioned consumption of more than one brand). Based on the information from the case–control study, FDA initiated investigations of two pecan processors, including plant F, which processes and packages brand F and plant G, which produces and packages brand G. Both plants were located in southern California. While product processed in plant F was distributed predominantly in California, Oregon, and Washington, products produced in plant G were distributed more widely, including 15 states in the western and mid-western United States. Numerous environmental swabs were obtained from sites within both processing plants, including the hulling, shelling, and packaging equipment, along with non-product contact areas such as surrounding walls and floors. In addition, several intact packages of product from both brand F and G were collected from stores and warehouses over the course of the investigation. Neither brand G product samples nor environmental samples taken from plant G were positive for Salmonella. While all brand F product samples were also negative for Salmonella, 5 of the 30 swabs taken from plant F were positive for Salmonella including two from the hulling equipment, one from the shelling equipment, and two from floor areas. Serotyping of four Salmonella isolates from each sample revealed

10

Value and Methods for Molecular Subtyping of Bacteria

167

that all isolates were SE. Subsequent PFGE typing of one isolate from each sample revealed that all isolates represented PFGE type XY. In addition, in-plant investigations by the FDA investigative team revealed that a greater than average rainfall during the spring of 2002 had caused several roof leaks throughout the plant F facility. Plant personnel confessed numerous challenges in controlling moisture within the raw pecan-processing areas through the spring and early summer. Plant F had no microbiological verification control programs (e.g., weekly Salmonella environmental samples) in place to determine the potential negative impact of increased moisture within the facility. Based on (i) strong epidemiological support for an SE PFGE type XY outbreak caused by consumption of raw pecans, including implication of brand F by a considerable number of case patients, (ii) product F distribution patterns that matched with the outbreak distribution, (iii) in plant microbiological investigations indicating the presence of type XY strain, and (iv) strong subtype evidence, FDA requested plant F to recall of all pecans produced within the last 6 months. Plant F contested this request, arguing that the fact that none of the more than 50 brand F product samples tested were positive for Salmonella indicated that plant F was not the source of this outbreak. Plant F microbiologist argued that it is likely the SE PFGE type XY represents a pecan-associated Salmonella subtype, which may be present in pecans from different manufacturers. Thus, the presence of this subtype in environmental samples of plant F is irrelevant since this subtype could likely be found in a number of pecan-processing plants. The fact that none of brand F finished product samples were positive indicated that plant F had appropriate environmental control strategies in place that prevented contamination of finished product. The plant F expert witness specifically suggests that another, yet unidentified, pecan-processing plant that also received raw material contaminated with SE PFGE type XY may have had a temporary breakdown in environmental control leading to production of one or more highly contaminated batches of pecans, which represent the true cause of this outbreak. In response to the arguments raised by plant F, FDA presents data from a recent university-based research study on Salmonella contamination in 15 pecanprocessing plants and 30 pecan farms in California, which led to isolation of Salmonella from environmental samples collected in seven processing plants as well as from 16 pecan farms. Among the 50 Salmonella isolates obtained as part of this study, only three were identified as SE and none of the three SE isolates matched PFGE type XY. In addition, as part of the specific outbreak investigation, FDA had collected drag swabs from orchards located on seven pecan farms in California that supplied raw materials to plants F and G. While samples from five of the farms were found positive for Salmonella, only the three Salmonella isolates from one farm, which was the main supplier to plant F, were found to be SE, which were confirmed the be PFGE type XY. The Salmonella isolates from the other four farms represented Salmonella serotypes other than SE. Presentation of these results convinced the management of plant F that a recall of all raw pecan products produced in their plant was appropriate and necessary. Conclusions and lessons learned from this outbreak investigation. The above case study is typical for the type of foodborne disease outbreak investigation that is

168

M. Moorman et al.

now commonly conducted around the world. With minor adjustments, the same type of outbreak investigation may occur in response to clusters of human infections with L. monocytogenes (see Olsen et al., 2005; MacDonald et al., 2005), E. coli O157:H7 (see Jay et al., 2004), Shigella (see Kimura et al., 2004), Yersinia enterocolitica (see Ackers et al., 2000), or any number of foodborne pathogens. In the above case study, the lack of isolation from the finished packaged product of a Salmonella strain that matched the PFGE type associated with the human outbreak represents a missing piece of evidence. While many outbreak investigations similar to the one described above may include isolation of a subtype matching the outbreak subtype from an intact sample of a food product, isolation of an isolate with a matching subtype from a food product is not always necessary to conclusively link an outbreak to a specific manufacturer. In the absence of a food isolate that matches the subtype associated with an outbreak, very strong epidemiological evidence and/or other additional evidence is required. In the above case study, this additional evidence is provided by (i) a strong epidemiological link of the outbreak with plant F and (ii) isolation of a subtype matching the outbreak-associated subtype from the plant environment and from the environment of a farm supplying pecans to plant F. Nevertheless, isolation of a subtype matching an outbreak-associated subtype from a plant environment alone should not be considered sufficient evidence for a causal link between a plant and an outbreak. It is easily feasible that an isolate matching a subtype associated with an outbreak can be isolated from a plant environment, even though a plant is not responsible for an outbreak. Plant F’s request for more information on the frequency of PFGE type XY was thus highly appropriate and justified. In the absence of the information that PFGE type XY was rare among pecan and pecan-processing associated sources (or if PFGE type XY would have been found in a number of other plants and/or farms, indicating that this subtype may have been a pecan-associated subtype), the subtype evidence may have not been sufficient to request a recall by plant F. Good information on the frequency of different pathogens subtypes among a variety of sources is critical in the application of molecular subtyping methods in foodborne disease outbreak investigations.

10.3.2 In-Plant Investigation – Microbiological Spoilage of Salad Dressing Company C produces a variety of refrigerated salad dressings that are nationally distributed across the United States. In late fall of 2003, the operation began receiving consumer complaints regarding spoilage of two of its creamy dressing brands. Product defects reported included off-flavor, separation, and/or gas bubbles. Previously, Company C had only fielded sporadic spoilage complaints regarding these two dressing types. For the most part, the environmental sampling history of the operation verified good sanitary controls, with only a low incidence of yeast or bacteria being observed sporadically over a number of routine evaluation sites. The plant was able to obtain several bottles of spoiled product from retailers and warehouses. The on-site laboratory immediately began microbiological analysis of

10

Value and Methods for Molecular Subtyping of Bacteria

169

returned samples, largely focusing their efforts on detection and identification of yeast and lactic acid bacteria. In addition, the microbiology support and quality assurance teams initiated plant investigations (environmental sampling, equipment evaluations, etc.) of the two dressing processing lines (lines 1 and 2) where the creamy formulations were typically produced. The equipment used was mostly a closed system conveying product through pipes, and ultimately to filler heads and bottles. Routine sanitation was performed by a CIP process. Periodically, and where feasible, equipment was disassembled for visual evaluations of condition, detection of product build-up, and environmental testing. After a week, the laboratory had recovered a number of yeast and lactic acid bacterial isolates from spoiled product samples. Quantitative analysis by plating had shown that most spoiled products contained high levels of lactic acid bacteria, with moderate to no detectable levels of yeast. Environmental swabbing had recovered both low levels of yeast and lactic acid bacteria from only a few product and nonproduct contact sites. The majority of sampling sites had no detectable levels of microorganisms. Reviewing the data in hand, the microbiology and quality assurance teams had determined that either processing line could have contributed microbial contamination causing the spoilage problem. There was no compelling evidence that would specifically implicate one line or sampling site as a source of this spoilage problem. While the plant laboratory had limited biochemical identification capabilities, the microbiologists could quickly characterize isolates from the environmental samples using a commercial Rep-PCR instrument. While REP-PCR analysis of lactic acid bacteria isolated from environmental samples revealed a number of DNA pattern types, 80 of the 100 isolates obtained from finished products showed the same pattern (type AB). In addition, REP-PCR analysis demonstrated that only one environmental isolate (obtained from a worn in-line gasket within a pipe on Line 1) matched REP PCR type AB. Based on these subtyping results, the gasket was immediately replaced. Plant sanitation and equipment maintenance procedures were revised to include more frequent disassembly and inspection of equipment components, including gaskets. Once these corrective actions were made, increased testing for lactic acid bacteria from environmental samples and from finished product samples (immediately after production) was continued. While low levels of lactic acid bacteria were occasionally recovered from environmental and from product samples, REP-PCR typing revealed that none of these isolates were pattern type AB, verifying the effectiveness of the corrective actions. In addition, the number of spoilage complaints received for the creamy dressings was reduced to baseline levels of one to two complaints per month. Conclusions and lessons learned from this in-plant investigation. The above investigation demonstrated the power of DNA analysis for quickly pinpointing the source of microbial contamination. Traditional food microbiological identification techniques will typically involve biochemical, or even serological techniques, to determine the ultimate source of microorganism(s) responsible for the product spoilage. These procedures can be quite time-consuming.

170

M. Moorman et al.

Although pursuing speciation of lactic acid bacterial isolates would be the normal response or instinct in such an investigation, the availability of REP-PCR to the laboratory obviated the need for traditional identification procedures. In the above scenario, expediency in determining the source(s) of contamination was more critical than fully biochemically characterizing isolates. The discriminating power and speed of REP-PCR allowed quicker and more effective resolution of the spoilage problem than traditional microbiological approaches. Other lactic acid bacterial strains had contaminated the product. However, pattern type AB obviously grew better and to higher levels in the creamy dressing, resulting in the recent spoilage event. Examples of the application of molecular subtyping methods to investigate in-plant sources of spoilage organisms and pathogens have also been described in the peer-reviewed literature (e.g., Ralyea et al., 1998; Lappi et al., 2004).

10.3.3 Company Response to Molecular or Subtyping Data Depending on its technical resources, a company either can adequately interpret subtyping data (generated internally or by an outside laboratory) or may require assistance from outside experts. Determination of this need obviously relates to the situation. If a company is utilizing molecular techniques internally for investigations of microbial contamination sources, relying on training and support provided by technology vendors, or consultation with university experts, may be sufficient to build competency. After gaining experience and confidence with a subtyping methodology, laboratory scientists should become skilled in interpreting results and understanding assay nuances. Nevertheless, in urgent situations involving the implication of a manufacturer’s product due to claims of microbial adulteration or foodborne illness, even the most highly skilled and technically staffed companies may need advice from external resources. Outside contacts can provide guidance to help a company better protect its interests or make crucial decisions regarding public health protection. Obviously, no company looks forward to receiving this type of news regarding its products. Therefore, it is in a company’s best interest to develop a policy to anticipate and manage such events. The policy should clearly outline employee responsibilities and communication plans in dealing with suppliers, regulators, media, company customers, and consumers. Most companies already have crisis management procedures in place for a variety of emergencies. The response to unfavorable, regulatory data should be included in such a document. An emergency response team should be established to move quickly in the event of a potential recall. Table 10.1 summarizes what the composition of this team could look like based on common positions or departments in larger businesses. Although smaller operations may not be staffed to the same degree or with equivalent positions, practical alternatives can be developed. In working with regulatory agencies, companies must be mindful that the primary role of both parties is to assure that the appropriate consumer protection decisions are made.

Will be advised and updated by company’s technical experts. Must clearly understand impact to business and why certain decisions and actions will be necessary Works with regulatory agency to understand level and magnitude of issue. To the extent possible, will need to evaluate microbiological data and make decisions regarding its reliability. Will serve as internal coordinator regarding company communications and direction. Tracks and leads retrieval of affected products and determines disposition Provides technical support for any data interpretations. Quickly accesses databases or laboratory records to demonstrate verification of microbiological control, which could prevent a recall or limit its scope With guidance from other response team members, will craft public communications, including company statements and notices. Reviews regulatory recall notices and leads efforts to ensure their accuracy

Senior management

Communications

Laboratory

Quality assurance/food safety

Role/responsibility

Position/department

Effective communicator at management level or a human resources employee could be trained to serve similar role

Commercial laboratories performing testing on behalf of a smaller company may serve similar role

May need to consult with outside third party laboratory scientist or subject matter experts (e.g., university professors, consultants) May need outside technical experts to deal with more rudimentary interpretations of data in addition to more complex results

Smaller company alternatives

Table 10.1 Some approaches to recall emergency response teams

10 Value and Methods for Molecular Subtyping of Bacteria 171

Will work with other members to protect business interests, focusing on minimizing financial losses and damages to the company’s reputation. Subsequent activities may include dealing with at-fault vendors that may have contributed to or actually caused recall Will communicate and work with suppliers that may have impacted event. May need to find alternative ingredient sources until vendor corrective actions are confirmed May be needed to advise and ensure that company asks appropriate questions regarding molecular subtyping data. Can provide supporting data that indicate prevalence of certain strains (e.g., commonality of a Listeria PFGE pattern in ready-to-eat meats). Information could minimize recall damages or ultimately prevent regulatory actions Are staffed to support companies in crisis situations. Can draw from internal or external experts to provide supporting data or help refute dubious interpretation of results or regulation

Legal

Industry organizations

External scientific experts

Purchasing/procurement

Role/responsibility

Position/department

Table 10.1 (continued)

Strongly recommend that smaller companies establish membership with at least one reputable industry organization. Fee structures may be proportionally adjusted to company size or revenue

Smaller companies should have established relationships with outside experts

Should have position responsible for this activity

These services can be outsourced. Legal firm should have experience with food regulation and has dealt with similar cases around food manufacturing

Smaller company alternatives

172 M. Moorman et al.

10

Value and Methods for Molecular Subtyping of Bacteria

173

In the aftermath of a recall, companies will need to quickly resolve any microbial contamination issues that caused product adulteration. Additional external resources may be required to accelerate the investigation (e.g., determine the ultimate microbial contamination source) and effect remedial actions, so that the processor may return to normal operations as quickly as possible.

10.4 Conclusions Molecular subtyping methods provide powerful tools that can help to detect foodborne disease outbreak sources and define outbreak sources. In addition, these methods also provide food processors with an opportunity to rapidly and accurately define sources and transmission of spoilage microorganisms. Many of the subtyping methods currently used have matured into robust technologies that are commercially available (either through kits or through service laboratories), however, considerable knowledge and skill is still required for appropriate interpretation of subtyping results. Technological advancements will likely provide further improvements in subtyping methodologies, including the potential for combined PCR-based detection and subtyping, nevertheless a larger need may lie in the development of comprehensive subtype databases to facilitate correct interpretation of subtype results.

References Ackers ML, Schoenfeld S, Markman J, et al. (2000) An outbreak of Yersinia enterocolitica O:8 infections associated with pasteurized milk. J Infect Dis 181:1834–1837 Braden CR, Templeton GL, Stead WW, Bates JH, Cave MD, Valway SE (1997) Retrospective detection of laboratory cross-contamination of Mycobacterium tuberculosis cultures with use of DNA fingerprint analysis. Clin Infect Dis 24:35–40 Brosch R, Chen J, Luchansky JB (1994) Pulsed-field fingerprinting of listeria: Identification of genomic divisions for Listeria monocytogenes and their correlation with serovar. Appl Environ Microbiol 60:2584–2592 Breuer T, Benkel D, Shapiro R, et al. (2001) A multistate outbreak of Escherichia coli O157:H7 infections linked to alfalfa sprouts grown from contaminated seeds. Emerg Infect Dis 7: 977–982 Bruce J (1996) Automated system rapidly identifies and characterizes microorganisms in food. Food Technol 50:77–81 CDC 2009. Salmonella Strains Tables for Outbreak Related to Peanut Butter and Peanut Containing Products. January, 29. http://www.cdc.gov/salmonella/typhimurium/strains_ table.html Farber JM, Gendel SM, Tyler KD, Boerlin P, Landry WL, Fritschel SC, Barrett TJ (2001) Molecular typing and differentiation, Chapter 11. In: Compendium of Methods for the Microbiological Examination of Foods, 4th edn. American Public Health Association, Washington, DC Graves LM, Swaminathan B, Hunter SB (1999) Subtyping Listeria monocytogenes. In: Ryser E, Marth E (eds) Listeria, listeriosis, and Food Safety. Marcel Dekker, New York, pp. 279–297 Grimont F, Grimont PAD (1986) Ribosomal ribonucleic acid gene restriction patterns as potential taxonomic tools. Ann Inst Pasteur/Microbiol (Paris) 137B:165–175

174

M. Moorman et al.

Jay MT, Garrett V, Mohle-Boetani JC, et al. (2004) A multistate outbreak of Escherichia coli O157:H7 infection linked to consumption of beef tacos at a fast-food restaurant chain. Clin Infect Dis 39(1):1–7 Kerr KG (1994) The rap on REP-PCR-based typing systems. Rev Med Microbiol 5(4):233–244 Keys C, Kemper S, Keim P (2005) Highly diverse variable number tandem repeat loci in the E. coli O157:H7 and O55:H7 genomes for high-resolution molecular typing. J Appl Microbiol 98(4):928–940 Kimura AC, Johnson K, Palumbo MS, et al. (2004) Multistate shigellosis outbreak and commercially prepared food, United States. Emerg Infect Dis 10(6):1147–1149 Lappi VR, Thimothe J, Nightingale KK, Gall K, Scott VN, Wiedmann M (2004) Longitudinal studies on Listeria in smoked fish plants: Impact of intervention strategies on contamination patterns. J Food Prot 67:2500–2514 Lai E, Birren B, Clark S, et al. (1989) Pulsed field gel electrophoresis. BioTechniques 7:34–42 MacDonald PD, Whitman RE, Boggs JD, et al. (2005) Outbreak of listeriosis among Mexican immigrants as a result of illicitly produced Mexican-style cheese. Clin Infect Dis 40(5): 677–682 Olsen SJ, Patrick M, Hunter SB, et al. (2005) Multistate outbreak of Listeria monocytogenes infection linked to delicatessen turkey meat. Clin Infect Dis 40(7):962–967 Olive DM, Bean P (1999) Principles and applications of methods for DNA-based typing of microbial organisms. J Clin Microbiol 37:1661–1669 Randazzo CL, Torriani S, Akkermans AD, de Vos WM, Vaughan EE (2002) Diversity, dynamics, and activity of bacterial communities during production of an artisanal Sicilian cheese as evaluated by 16 S rRNA analysis. Appl Environ Microbiol 68:1882–1892 Ralyea RD, Wiedmann M, Boor KJ (1998) Bacterial tracking in a dairy production system using phenotypic and ribotyping methods. J Food Prot 61:1336–1340 Read TD, Salzberg SL, Pop M, et al. (2002) Comparative genome sequencing for discovery of novel polymorphisms in Bacillus anthracis. Science 296(5575):2028–2033 Spratt BG (1999) Multilocus sequence typing: Molecular typing of bacterial pathogens in an era of rapid DNA sequencing and the Internet. Curr Opin Microbiol 2:312–316 Strockbine NA, Wells JG, Bopp CA, et al. (1998) Overview of detection and subtyping methods. In: Kaper JB, O’Brien AD (eds) Escherichia coli O157:H7 and other Shiga-toxin producing E. coli strains. ASM Press, Washington, DC, pp. 331–356. Swaminathan B, Barrett T, Hunter S, et al. (2001) PulseNet: The molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg Infect Dis 7:382–389 Tenover FC, Arbeit RD, Goering RV, et al. (1995) Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: Criteria for bacterial strain typing. J Clin Microbiol 33:2233–2239 Threlfall E, Frost J (1990) The identification, typing and fingerprinting of Salmonella: Laboratory aspects and epidemiological applications. J Appl Bacteriol 68:5–16 Top J, Schouls LM, Bonten MJ, Willems RJ (2004) Multiple-locus variable-number tandem repeat analysis, a novel typing scheme to study the genetic relatedness and epidemiology of Enterococcus faecium isolates. J Clin Microbiol 42:4503–4511 Van Belkum A, Struelens M, de Visser A, et al. (2001) Role of genomic typing in taxonomy, evolutionary genetics, and microbial epidemiology. Clin Microbiol Rev 14:547–560 Wassenaar TM, Newell DG (2000) Genotyping of Campylobacter spp. Appl Environ Microbiol 66:1–9 Welsh J, McClelland M (1990) Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Res 18:7213–7218 Wiedmann M (2002a) Subtyping of bacterial foodborne pathogens. Nutr Rev 60:201–208 Wiedmann M (2002b) Molecular subtyping methods for Listeria monocytogenes. J AOAC Int 85:524–531 Williams JG, Kubelik AR, Livak KJ, Rafalski JA, Tingey SV (1990) DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531–6535 Woodford N, Johnson AP (2004) Genomics, Proteomics, and Clinical Bacteriology: Methods and Reviews. Humana Press, Totowa, NJ

Index

Note: Page numbers featuring figures and tables are given with ‘f’ and ‘t’ followed with the corresponding figure/table number. A Abdominal cramping, 21 Abdominal pain, 9, 15, 23, 32, 37–38 Acetoin, 72t3.3 Acid, 34, 69, 72t3.3 Acidophiles, 64 Acid-tolerant, 28, 67, 71 bacteria, 71 Acid tolerant yeasts, 67 Acinetobacter, 65, 67, 71 Acinetobacter calcoaceticus, 71 Ackers, M. L., 168 Active transport, 108 Adams, 36 Aerobacter, 70 Aerobic respiration, 19, 109 Aeromonas, 70 Agar slide method, 129 Air, 22, 67, 80 Air flow (unsanitary design), 82 Air handling units (AHUs) (unsanitary design), 82 Air line boxes, 86 Air sampler sieve-type impactor, 129 slit-type impactor, 129 Alcaligenes, 69t3.2, 70 Alcohols, 70, 72t3.3 Alfalfa sprout, 10, 34, 40 Alicyclobacillus spp., 72t3.3 Allan, J. T., 6 Altekruse, S. F., 14–16 Alter, T., 18–19 Alternaria, 70 Alteromonas, 66t3.2 Alteromonas putrefaciens, 66t3.1 Amino acid, 67 Anaerobes, 19, 35, 109–110 Ancestor, 160–161, 165

Andrews, W. H., 144 Anemia, 32 Anifantaki, K., 66 Animal feed, 10 Annous, B. A., 134, 151 Anserine, 67 Antibiotics, 110 Antibiotic-resistant, 16, 21 multi-antibiotic-resistant, 20 Antimicrobials, 28, 159 APC (aerobic plate counts), 132–134, 154 AP-PCR (amplified polymorphic polymerase chain reaction), 163 Apples, 108t5.2 Apple cider, 34 Arcobacter, 38–39 characteristics of, 38 food and environmental sources, 39 nature of disease, 38 Arcobacter butzleri, 38–39 Arcobacter cryaerophilus, 38–39 Arcobacter skirrowii, 38–39 Arritt, F. M., 118 Arseni, A., 41 Ascorbates, 28 Aseptic sampling, 120 Aspergillus, 67, 73t3.4, 73, 107, 108t5.2 Aspergillus flavus, 107t5.1 Aspic glaze, 10 Assadi, M. M., 41 Assays, 27, 47, 119–120, 129–130, 133, 140, 165 Assistant sampler, 122 Atabay, H. I., 39 ATP assay, 129–130 Attached cells, 80 Attachment, 31–32, 127 Attributes plans, 143 Audit, 150

J.L. Kornacki (ed.), Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment, Food Microbiology and Food Safety, C Springer Science+Business Media, LLC 2010 DOI 10.1007/978-1-4419-5518-0, 

175

176 Audit scores, 151–152 Autoclave, 23, 105f5.3 Autoclaved skim milk, 104f5.3 aw (water activity), 27–28, 36, 106, 112, 143 limiting, 107 B Bacillus spp., 65, 69, 108 Bacillus cereus, 6, 22–25 factors related to cost, 23 food processing environments, 24–25 foods associated, 24 infectious dose, 24 organism, 22–23 reservoirs, 24 symptoms, 23 Bacillus coagulans, 66t3.1, 74 Bacillus licheniformis, 73 Bacillus pumilus, 73 Bacillus stearothermophilus, 74 Bacillus subtilis, 73 Bacteremia, 9 Bacteria, 2, 37, 64–69, 104, 106, 109–111, 169 Bacterial cell membranes, 107–108 Bacterial survival at lethal temperatures, 111f5.5 Bacteriocins, 18, 28, 110 Baetz, A. L., 38 Bakery products, 22, 72–74, 73t3.4 spoilage of, 72–74 Baksh, F. K., 44 Ball, W. S., 41 Ball–valves (unsanitary design), 82, 100f4.32 Banding pattern, 161–162 Banks, J. M., 70 Banwart, G. J., 107–108, 110 Barletta, R. G., 44–51 Bar-Oz, B., 41 Battey, A. S., 71 Bean, P., 160 Beans, 26 Beef, 17, 33–34, 40, 46, 134 Beer, 72t3.3 Beginning investigation, food contamination determine extent of contamination, 120 assembling sampling team, 121–122 determine production schedule, 120 post-startup observations and sampling, 123 prepare operations for disruption, 121 sanitation observations and validation, 123 touring plant and preliminary site selection, 121

Index facts collection, 118 know process check assumptions, 118–120 review HACCP plan, 118 Belay, N., 23 Bennett, R., 23 Besser, R. E., 29–30, 32, 34 Betts, R. P., 75 Bettsia alvei, 74 Beuchat, L. R., 41 Beverage products, 40, 71–72 spoilage of, 71–72 Beverages (processed), Causative spoilage agents and manifestations of microbial defects in, 72t3.3 Biering, G., 41 Biofilms, 18, 19, 71, 80, 127–128 formation, 127 Biosecurity, 17 Bioterrorism, 21 Birds, 11, 16–17, 39 Bitter, 69 Bitter peptides, 70, 71 Bloating, 66 Block, C., 41 Bloodborne, 13 Blood pressure, 21 Blown-up, 66 Body temperature, 21 Boils, 21 Boor, K., 45 Bordova, O., 69 Bowen, A. B., 42 Braconnier, A., 28 Braden, C., 41 Braden, C. R., 42, 166 Bradshaw, J. G., 36 Bradshaw, M., 27 Brain, 13 Bread, 40, 73t3.4, 73 Breathing (difficulty), 26 Breeuwer, P., 41 Breuer, T., 162 Broccoli, 28 Broken gasket seals on product pumps (unsanitary maintenance and repair practices), 81 Brosch, R., 162 Browning defects, 68t3.2 Bruce, J., 163 Brunner, K. G., 20 Bryan, F. L., 45 Bryant, R. G., 145

Index Bulging, 66 Bull, T. J., 44 Burdette, J. H., 41 Butter, 7t2.1, 10, 14, 68t3.2, 70, 72t3.3, 108t5.2 Butzler, J. P., 16 Buzby, J. C., 2, 8, 12, 15, 30 Byssochlamys fulva, 74 C C. botulinum, 25–28 control treatments, 27–28 detection of neurotoxins, 27 disease incidence and syndrome, 25–26 organism, 25 physiological characteristics, 26–27 reservoirs and prevalence in foods, 26 C. esterthicum, 66t3.1 C. piscicola, 66t3.1 Cabbage, 108t5.2 Calcium lactate, 71 Callicott, K. A., 17 Campylobacter, 6, 7t2.1, 14–19, 29, 38 costs, 15 disease syndromes, 15–16 foods associated, 16–17 infectious dose, 16 organism, 15 and poultry, 17–19 reservoirs, 16 Campylobacteriosis, 15, 17–18 Candida, 67, 72t3.3 Candida guillermondii, 68t3.2, 73 Candida lipolytica, 71 Candida magnolia, 75 Candida paralopsis, 68t3.2 Canned foods, 27, 74 spoilage of, 74 Canning, 19, 25, 28, 36, 74, 111 Cantaloupe, 10 Carbohydrates, 64, 66, 76, 106 Carcasses, 17, 19, 35, 39, 46 Carnobacterium divergens, 66t3.1 Carnosine, 67 Carrots, 26 Catalase, 19, 109 Catch pans (unsanitary design), 81–82 Cattle, 16, 33, 44–46 Caubilla-Barron, J., 41 Causal, 158, 162 C4 –C10 fatty acids, 70 Cealing leaks, 94f4.20 Ceiling tiles, 91f4.15 Cell lysis, 70

177 Cellular metabolism, 106 Centrifugal samplers, 129 Cereal, 24, 40 Cereulide, 23 Chalk mold, 73 Challenge studies, 82, 103 Chamber’s sanitation efficacy test, 127 Cheddar cheese, 10, 71 Cheese, 7t2.1, 10, 13, 14, 34, 40, 46, 68t3.2, 69, 70–71, 140 early gas defects in, 70 Cheese curds, 34 Chemical disinfection, 63 Chemically bound, 106 Chicken, 17–19, 22, 39, 75 Children, 13, 15, 21 Chiodini, R. J., 44–45 Chlorine, 18–19, 152–153 Chmielewski, R., 111 Chocolate, 7t2.1, 10, 22, 40, 74, 83, 112, 134 Chocolate candy, 107t5.1 Chou, Y. E., 71 Chrysosporium farinicola, 74 Citrate-utilizing strains, 70 Cladosporium spp., 67, 70, 73 Clark, D. L., 46 Clark, N. C., 41 Clavero, R., 63–77 Cleaner, 40, 80, 127, 152 Cleaning and sanitation, effectiveness of, 127 Cleaning efficacy, 130 Clean-up to clean-up, 154 Clonal, 164 Clostridium spp., 65, 67 Clostridium botulinum, 6, 7t2.1, 25–28, 35–36, 107t5.1, 110–111 Clostridium botulinum type E, 107t5.1 Clostridium butyricum, 66t3.1 Clostridium ctm, 66t3.1 Clostridium laramie, 66t3.1 Clostridium pasteurianum, 66t3.1, 72t3.3 Clostridium perfringens, 2, 5, 6, 7t2.1, 35–37, 143 cost, 36 disease syndromes, 37 enterotoxin, 36 infectious dose, 37 organism, 35–36 reservoirs, 37 Clostridium sporogenes, 68t3.2, 74, 134 Clostridium thermosaccharolyticum, 68t3.2 Clostridium types A and B, 25–26, 35–37 Clostridium tyrobutyricum, 68t3.2

178 Cloudiness, 71 CO2 , 46, 66, 67 Coagulase positive Staphylococci, 128 Coccobacilli, 65 Cocoa beans, 135 Coignard, B., 41 Colby cheese, 112 Colby-like cheese, 113f5.6 Cold chain, 67 Cole, M. B., 28 Coliforms, 30, 42, 65, 72. 128, 132 Colitis, 30, 32 Collins, C. I., 39 Collins, M. T., 43, 45–46 Colon, 29, 31–33 Colonization, 16, 31–32 Coma, 32 “Commercially processed” foods, 6 Compendium of Methods for Microbiological Examination of Foods, 24, 75, 139 Competition, 110 Composite sampling, 154 Composites/compositing (sample), 140, 144, 154 Compressed air, 71, 81, 99 Condensate catch pan drains (unsanitary maintenance and repair practices), 81 Confectionery products, 74 Confectionery products, spoilage of, 74 Conserved regions, 163 Constipation, 26 Contaminant sources, 79–101 equipment or factory design creating growth niches/transmit microbes, 96–101 maintenance/repair practices creating growth niches/transmit microbes, 90–95 operating practices creating growth niches/transmit microbes, 84–89 significance of environmental contamination, 79–83 significance of ingredient contamination, 83 significance of process failure events, 83 Contamination, 1–3, 22, 35, 42–43, 65, 67, 71, 73, 79–83, 118, 120–123, 125–126, 130, 138f8.1, f8.2 Contamination events, 120, 123, 127, 147 Conveyor, 88f4.11, 97f4.26 Conveyor belt, 132

Index Cooling, 25, 82, 118–119 Cooling coils (unsanitary design), 82 Corn, 26, 43 Cornblath, D. R., 16 Corn syrup, 43 Corrective actions (CAs), 118, 148, 150, 169 Corry, J. E. L., 39 Corynebacterium, 69 Coryneforms, 69 Cost of food contamination costs associated with selected foodborne pathogens, 2 costs associated with spoilage and foods with microbial indicators of “unacceptable quality,” 2–3 microbial foodborne illness in America, 1–2 Cottage cheese, 14, 68t3.2, 70 Cottyn, B., 40 Cox, N. A., 17 Cracked plates, 69 Cracks, 71, 101 Cramps, 9, 15, 23 Cream, 14, 68t3.2, 70, 108t5.2 Creatine, 67 Cress, 24 Crevices, 128, 129, 132 Critical control points (CCP), 83, 117, 118, 139, 148–150 surrogate organisms, 47 validation, 83 Critical limits (CL), 83, 117, 118 Crohn’s disease, 44, 47 Cronobacter (Enterobacter sakazakii), 5, 6, 39–40, 43 food industry concerns, 42–43 introduction, background, and bacterial characteristics, 39–40 pathogenicity and infectious dose, 41–42 regulation, 42 reservoirs and presence in food and environment, 40–41 Cross contamination, 17, 69, 97, 132, 166 Cruz, A. C., 40 Cryptococcus, 67, 68t3.2 Cryptococcus laurentii, 66t3.1 Cultured milk products, 13 Cuniff, P. A., 119 Cunningham, A. F., 46 Curd, 68t3.2, 70 Curiale, M., 140 Cuttlefish, 10 Cytosol, 108

Index D Dairy products, 10, 13–14, 22, 24, 45, 68–71, 133 causative spoilage agents and manifestations of microbial defects in, 68t3.2 spoilage of, 68–71 Damaged gasket/s, 69 D’Aoust, J. Y., 9–10 Datta, A. R., 12 David, J. R. D., 20, 74 Davies, A. R., 65 12D concept, 111 Deamination, 112 Death, 1–2, 29, 32, 104, 106, 110–111 Debaryomyces hansenii, 66t3.1, 68t3.2, 73 Debaryomyces vanriji, 66 DeBeer, D., 80, 127 Decline phase, 104 Dehydration, 9, 18, 21, 63 Deli meats, 13, 158 Den Aantrekker, E. D., 6 Desiccation resistance, 41 Detection, 11, 24, 27–28, 140, 164 Dextrose, 12, 76 Dhir, V. K., 80, 127 Diacetyl aroma, 71 Diarrhea, 9, 12, 15, 23, 31, 38 Dillon, V., 67 Dimethyl sulfide, 70, 75 Disassembly (unsanitary design), 76, 122 Discriminatory/discrimination, 159, 160, 162, 163–164, 170 Disease syndromes/infectious process, 5, 8–9, 12–13, 15–16, 21, 30, 37 Disruptions, 121 Dizziness, 26 D(–)-lactate, 71 DNA, 46, 75, 160–162 fingerprinting, 160 probe, 162 16s rDNA, 159–160 DNA sequencing technology, 75, 159, 161, 163–163 Dodd, C. E. R., 69, 80, 127 Dogs, 16, 34 Double vision, 26 Downes, F. P., 75 Doyle, L. P., 16 Doyle, M. P., 5, 18, 29–30, 32–34 Doyle, T. M., 44–45 Drinking water, 39 Drudy, D., 41

179 Dry cereal, 10 Dry foods, 10 Drying, 106, 112 Dryness of skin, mouth and throat, 26 Dry soup mixes, 10 Duct tape (unsanitary maintenance and repair practices), 81, 131 Dust (unsanitary operating practices relating to), 22, 37, 40, 81, 86f4.6 D-value, 105f5.3, 111 E E. sakazakii, see Cronobacter (Enterobacter sakazakii) Eblen, B. S., 28 Eblen, D. R., 151 Edelson-Mammel, S. G., 41 EDTA, 28 Egg drink, 10 Eggplant, 108t5.2 Eggs, 7f2.1, 10, 22, 24 Egg salad, 10 E h , see Oxidation/reduction potential (ORP/Redox, E h ) Eifert, J. D., 39 Electrical box, 95f4.23 Electrical charging rooms for electrically powered vehicles (unsanitary design), 82 Electrophoresis, 159, 161–162 Electrostatic precipitation impactors, 129 Elevators (unsanitary design), 82, 97f4.27 Ellingson, J. L., 45 Ellingson, J. L. E., 44 Emergency response team, 160, 170, 171t10.1 Emerging pathogens, 6 concern to food processors (industrial) Arcobacter, 38–39 Cronobacter (Enterobacter sakazakii), 39–43 M. avium subsp. paratuberculosis, 43–47 Emetic, 20, 23 Emilani, F., 41 Emulsions, 74–75 Endocarditis, 12 Endophthalmitis (eye infection), 23 Endospores, 26, 37 Endotoxin, 9, 41, 44 Engeljohn, D., 151 Enrichments, 40, 119, 133–134, 140–141 Enteric/paratyphoid fever, 9 Enteric pathogens, 14, 133

180 Enterobacteriaceae, 8, 40, 43, 65, 72t3.3, 128, 132, 154 Enterobacter liquefaciens, 66t3.1 Enterobacter sakazakii, 6, 39–43 Enterococci, 66t3.1, 69, 110, 128, 133, 135 Enterococcus faecalis, 66 Enterococcus faecium, 66, 134 Enterocolitis, 9, 40 Enterohemorrhagic E. coli, 6, 29–35 cost, 30 disease syndromes, 30 attachment and colonization, 31–32 hemorrhagic colitis, 32 HUS, 32 onset time, 30–31 TPP, 32 food processing environments, 34–35 foods associated with, 34 infectious dose, 33 organism, 29–30 other sources of infection, 34 pathogenic mechanisms, 33 reservoirs, 33–34 unique acid tolerance in foods, 34 Enteropathogenic E. coli, 15, 112 Enterotoxigenic E. coli, behavior of, 113f5.6 Enterotoxigenic E. coli H10407, 113t5.6 Enterotoxins, 6, 20 Entrap, 80–81, 127, 132 Environment, 11, 14, 24–25, 34–35, 39–41 Environmental contamination, 6, 7t2.1, 79–83 examples of outbreaks attributed to, 7t2.1 significance of, 79–80 causes of microbial growth niche development, 80–82 in-factory risk assessments and zones, 82–83 Environmental monitoring, 43 Environment and equipment, sampling, 125–126 ATP bioluminescence and protein assays, 129–130 biofilm formation, factors resulting in, 127 some implications of growth niches and biofilms to food factory environment, 127–128 comparison of approaches, 128t7.1 factors resulting in microbial growth niche formation, 126 indicator microorganisms index organisms, 133 indicators of quality, 133–134 indicators of sanitation efficacy, 134

Index in-line sampling, 130–131 location, 131 recommended sampling supplies, 131T7.2 sampling techniques, 128 air sampling, 129 contact plate, 129 pre-sterilize tongue blades, 129 sponge samples, 128 swab samples, 128–129 suggestions for sampling supplies, 131 surrogates, 134–135 time value of operational swabs, 132 value of post-operational swabs, 132 value of pre-operational swabs, 132 Enzymes, 19, 162 Equipment, 22, 96–101, 125–135 Equipment or factory design creating growth niches/transmit microbes, 96–101 compressed air trap, 99f4.31 equipment designed with numerous areas to entrap product residue, 100f4.33 large immovable support bases for equipment, 98f4.29 poorly plumbed gurgling drain spraying droplets into factory environment, 98f4.28 sandwiched areas in gears under product conveyor can collect product residues, 97f4.26 sandwiched areas on cleaning equipment, 99f4.30 steam exhaust over floor by trench drain, 101f4.34 tracked residues and standing water inside elevators, 97f4.27 untapped (to drain) pump coolant water, 96f4.24 use of adjacent dissimilar flooring, 101f4.35 use of ball–valves, 100f4.32 water lines not tapped to drains, 96f4.25 Erickson, M., 22 ERIC PCR (Enterobacterial Repetitive Intergenic Consensus PCR), 163 Escherichia coli, 2, 5–6, 7t2.1, 29–35, 113f5.6, 133 Esculin-hydrolyzing bacteria, 133 Eslava, C., 33 Espeland, E. M., 41 Esterase, 69 Ethanol, 69–70 Ethyl esters, 69

Index Eurotium, 73t3.4, 73 Eurotium glaucus, 73 Evancho, G. M., 129, 132 Evans, J. R., 154 Exotoxin, 20, 44 Expediency, 170 Exponential growth phase, 69 Exposed product (unsanitary operating practices relating to), 80, 119 Extracellular degradative enzymes, 69 Extracellular polysaccharides, 73 Extracellular spoilage factors, 75 Extrapolation, 83, 112, 151 Extrinsic factors, 106, 110–114 F Factory environment, 6, 11, 24, 84–89, 90–95, 107, 127–128, 132 Factory process, 123 Facultative anaerobes, 19, 109, 110 False ceilings (unsanitary maintenance and repair practices), 67, 81, 91f4.15, 94f4.20 False positives, 119, 141 FAO/WHO, 40, 41, 42, 43 Farber, J. M., 40, 163 Farmer, J. J., 40 Fat, 24, 28, 69 Fat bloom, 74 Fatigue, 26 Fatty residues (unsanitary operating practices relating to), 80, 127 Faust, R. E., 80, 126 FDA, 24, 42–43 FDA BAM, 142, 154 FDA plans, 144–145 Category I, 144, 151 Category II, 144 Category III, 144 Fecal, 16, 30, 37, 42, 44, 46, 133 Feng, P., 29 Fermentation, 63, 76, 107, 112 Festy, B., 39 Fetus, 12, 13 Filler heads, 76, 169 Filtration samplers, 129 Final filter (unsanitary design), 82 Finished product tests, 118, 138–139, 150, 153–154 Fish, 26, 65–67 spoilage of, 65–67 Flat sour spoilage defect, 74 Flat surface, 129

181 Flavobacterium, 68t3.2, 69 Fletcher, M., 127 Flooring, 101f4.35 Flu-like symptoms, 12 Fly, 40 Foamy, 68t 3.2 Foegeding, P. M., 27–28 Fondants, 73 Foodborne disease from selected pathogens, 7t2.1 Food contamination, cost of costs associated with selected foodborne pathogens, 2 costs associated with spoilage and foods with microbial indicators of “unacceptable quality,” 2–3 microbial foodborne illness in America, 1–2 Food poisoning, 19–22, 24, 35–37 Food process diagram, 130f7.2 Food processors (industrial), selected pathogens of concern to, 5–6 Arcobacter, emerging pathogen, 38–39 B. cereus, toxigenic pathogen, 22–25 C. botulinum, toxigenic pathogen, 25–28 C. perfringens, toxico-infectious agent, 35–37 Campylobacter, infectious invasive agent, 14–19 Cronobacter (Enterobacter sakazakii), emerging pathogen, 39–43 enterohemorrhagic E. coli, toxicoinfectious pathogen, 29–35 infectious vs. toxigenic bacterial pathogens, 6 L. monocytogenes, infectious invasive agent, 11–14 M. avium subsp. paratuberculosis, emerging pathogen, 43–47 Salmonella, infectious invasive agent, 6–11 Staphylococcus, toxigenic pathogen, 19–22 Foods associated with (the organism), 10, 16–17, 24, 34 Food service germ, 35 Forklift, 40, 80, 82, 132 wheeled vehicle traffic (unsanitary operating practices relating to), 80, 97t4.27 Forklift tires, 131 Forsythe, S., 40 Forsythe, S. J., 41 Fragment(s), 161–163 Francis, J., 31

182 Franco, D. A., 15 Frank, J. F., 19, 107, 111 Frankfurters, 13, 66, 76 Franklin, A. W., 41 Free fatty acids, 110 Freeman, H., 44 Freeze drying, 106 Frost, J. A., 160 Frozen dairy, 13 Frozen dessert, 9, 10 Fruit, 10, 34, 40, 74 spoilage of, 74 Fruit juices, 34, 72t3.3, 107t5.1 Fruit soup, 10 FTG (fluid thioglycollate medium), 109 Fuerst, J. A., 82 Fungal contaminants, 67, 70 Fungi, 103–104, 106 Furunculosis, 21 G Gabis, D. A., 80, 110, 126, 128 Gakuya, F. M., 41 Gallagher, P. G., 41 Game meat, 34 Garcia-Lopez, M. L., 65 Garments (sources), 79 Gas, 29, 64–65, 66t3.1, 70–71 excessive, 71, 108 formation, 71, 75 Gas chromatography, 75 Gaskets, 69, 76, 81, 84f4.1, 90f4.13, 93f4.19, 122, 169 valves (unsanitary operating practices relating to), 84f 4.1, 87f4.8, 93f4.19 Gas producers, 74 Gassem, M. A. A., 40 Gassing, 66t3.4, 73t3.4 Gassy fermentation, 76 Gastroenteritis, 9, 14–15 Gastrointestinal illness, 12, 23, 29, 44 Gastrointestinal tract, 9, 25–26, 29, 33 Gears, 97, 132 Generation time, 35, 104 Genes, 31, 33, 75, 163–164 virulence genes, 164 Genome, 161, 164 Georgsson, F., 18 Geotrichum, 70 Geotrichum candidum, 68t3.2 Germicidal footbaths (unsanitary operating practices relating to), 80 Gill, C. O., 7, 18

Index Glass, K. A., 28 Glove (use) (unsanitary operating practices relating to), 80 GLP (good laboratory practices), 141–142, 165 Gluconobacter oxydans, 71 Glycoproteins, 80, 127 GMP (good manufacturing practices), 139, 142, 151–152 Graham, A. F., 27 Gram-negative aerobic bacteria, 65 Gram-negative cells, 110 Gram-positive lactic acid bacteria, 65 Gram-positive spores, 110 Gram-positive vegetative cells, 110 Grant, J., 34 Grapefruit, 108t5.2 Grape nuts, 68t3.2 Graves, L. M., 160 Green discoloration, 66 Greening, 65, 66 Green onion/s, 17 Greenwood, M. H., 10 Griffin, P. M., 32–33 Grimont, F., 162 Grimont, P. A. D., 162 Growth (factors influencing growth), 105f5.2 inhibitors, 126 interactions between microorganisms, 80, 126 moisture, 80, 106 nutrients, 79–80 oxidation/reduction potential, 79, 107, 109 pH, 126 temperature, 110–112 Growth niches, 80–82, 84–101, 126–128 Gude, A., 39 Guidelines (microbiological criteria), 42, 142 Guillain–Barré syndrome (GBS), 15 Gums, 40 Gurtler, J. B., 5–47 Guzewich, J., 43 H Ham, 7t2.1, 22, 26, 76, 108t5.2 Hamasaki, Y., 65 Hamilton, J. V., 40 Hammack, S., 24 Hammack, T. S., 144 Hammes, W. P., 67 Handcarts, 132 Hand samples, 119, 141 Hand washing, 22 Hanes, D., 9

Index Hansenula spp., 68t3.2 Hansenula anomala, 73 Hard cheese, 13 Harmon, S. M., 24 Harris, B. A., 44 Harris, L. M., 7, 18 Havelaar, A. H., 42 Hazard analysis, 118, 148–149 Hazard analysis critical control points plan (HACCP), 118, 139, 148–149, 150–151 Headache, 15, 21, 26 Health, C., 42 Heat, 18, 36, 65, 76, 106, 118, 127 Heat resistant, 20, 24, 27–28, 36–37, 69–70, 135 Heat resistant lipases, 69. 70 Helgason, E., 22 Hemolytic uremic syndrome (HUS), 30, 32 Hennessy, W. H., 10 Henning, W. R., 45 Herbs, 40 Hermon-Taylor, J., 44–45 Heterofermentative lactic acid bacteria, 65, 70 Heterofermentative Lactobacilli, 108 Heterofermenters, 64, 66 Hiett, K. L., 17 High acid foods, 64, 74 High pressure hoses, 123, 128 aerosolization (unsanitary operating practices relating to), 80 High pressure treatments, 112 Himelright, I., 41 Histamine, 67 Hitchens, A. D., 145 Hold and test, 153–154 Hollow structures, 95f4.22 Holzapfel, W., 67 Holzapfel, W. H., 65 Home-preserved, 26 Homofermenters, 66 Honey, 107t5.1 Hoover, D. G., 18 Hopper, W. L., 10 Horses, 34 Hoses (unsanitary maintenance and repair practices), 80, 81, 91f4.14 Host, 6, 9, 17, 19 Houf, K., 39 H2 S, 65, 66t3.1 Hugenholtz, P., 82 Hughes, R. A., 16 Hurdles, 71, 113–114

183 Hydrogen peroxide (H2 O2 ), 66, 109 Hydrolytic rancidity, 70 Hydrophilicity, 127 Hydrophobicity, 127 Hydrostatic pressure, 18 Hygiene, 17, 67 Hypotension, 9 I Ice cream, 7t2.1, 10, 13–14, 134 ICMSF, 13, 15, 30–32, 82, 148, 151 plans, 143–144 Immiscible liquids, 74 Immunocompromised adult, 12, 40 Immunological analyses, 75 Impact wrenches (effect of improper use on gaskets) (unsanitary maintenance and repair practices), 81, 93f4.19 Impingers, 129 Incubation, 6, 30, 112 In-factory risk assessment, 82–83 Infant formula, 7t2.1, 40–43 Infants, 9, 12, 40–41 Infectious dose, 9, 13–14, 16, 24, 33, 37, 41–42 Infectious invasive agents, 6 concern to food processors (industrial) Campylobacter, 14–19 L. monocytogenes, 11–14 Salmonella, 6–11 Infectious vs. toxigenic bacterial pathogens, 6 Ingredient contamination, significance of, 83 Ingredients, 6, 43, 83, 151–152 Ingredient specifications, 28 Inhibitors, 80, 126 Inhibitors (factors influencing growth), 80 Inhibitory substances, 28, 107–108 In-line (tests), 130 Insulation (unsanitary maintenance and repair practices), 81, 92f4.16 Interactions between microorganisms (factors influencing growth), 80, 126 Intermediate moisture, 72 International Commission on Microbiological Specifications for Foods (ICMSF), 15 Intoxication, 19, 21, 24, 25 Intracellular, 13, 45 Intrinsic factors, 104, 106–110, 149 Investigational sampling, 117, 139–140 Investigations, 31, 44, 67, 75, 131, 132 Irradiation, 18, 20, 63 Isolates, 20, 36, 38, 162, 163–167, 169–170 Ito, K., 75 Iversen, C., 40–42

184 J Jacobs-Reitsma, W. F., 17 Jam, 73, 106 Janicka, G., 41 Jay, J., 109 Jay, M. T., 168 Jenson, I., 23 Joerger, R. D., 18 Johansen, K. A., 46 Johne’s disease, 44, 46 Johnson, E. A., 25, 28 Johnson, J., 29 Johnson, J. L., 133 Johnson, K., 160 Jöker, R. N., 41 Jones, P. H., 17 Jones, R. L., 44 Joshua, G. W., 18 Juneja, V. K., 28 Jung, M., 40 K Kalinowski, R. M., 67 Kandhai, M. C., 40 Kang, D. H., 40 Keener, K. M., 15–16 Kennedy, D., 45 Ketley, J. M., 16 Keys, C., 164 Kiehlbauch, J. A., 38 Kim, J., 27–28 Kimura, A. C., 168 Kinderlerer, J. L., 74 Kleiman, M. B., 41 Kluyveromyces marxianus, 68t3.2 Knill, M. J., 16 Kollee, L. A. A., 41 Kornacki, J. L., 1–3, 5–47, 70–71, 79–101, 103–114, 117–123, 125–135, 137–145, 147–154 Kozempel, M. F., 134, 151 Kramer, J. M., 17 Kruiningen, H. J., 44 Kuzina, L. V., 40 L L. monocytogenes, 2, 6, 7t2.1, 11–14, 82, 110, 112, 133, 134, 145, 158, 162, 165 cost, 12 disease syndromes, 12–13 food processing environments, 14 growth in high-temperature, 111f5.5 infectious dose, 13–14

Index infectious process, 13 key reactions of Listeria, 12t2.3 organism, 11–12 reservoirs and implicated foods, 14 L. perolens, 71, 72t3.3 Laboratory cross contamination, 120, 166, 173 Laboratory environment, 119, 141 Lactate salts, 28 Lactic acid, 19, 28, 64–67, 70, 71, 76, 107, 169–170 Lactic acid bacteria (LAB), 28, 64–67, 72t3.3, 107, 135, 169–170 Lactobacilli, 65, 71, 75, 107t5.1, 108t5.2, 135 Lactobacillus spp., 72t3.3, 108 Lactobacillus jensenii, 66 Lactobacillus plantarum, 75 Lactobacillus viridescens, 66t3.1, 76 Lactococci, 69 Lactococcus lactis, 65, 68t3.2 subsp. Diacetylactis, 68t3.2 Lai, E., 162 Lai, K. K., 41 Lamb, 17 Lammerding, A. M., 39 Lappi, V. R., 170 Lawsuit, 2, 8, 15 Lb. buchneri, 75 Lc. lactis var maltigenes (Lc. lactis var. maltigenes), 68t3.2, 69, 70 Leclercq, A., 40 Lehmacher, A., 10 Lehner, A., 38–39 Lenati, R., 42 Lerner, J., 38 Lethal, 25, 83 Lettuce, 17, 34, 40 Leuchner, R., 67 Leuconostoc spp., 65, 66t3.1, 71 Leuconostoc citreum, 65 Leuconostoc mesenteriodes, 65 Leukocytes, 9, 13 Leuschner, R. G. K., 40 Lindsay, W. M., 154 Line, J. E., 17–18 Lipases, 69, 70 Lipolytic bacteria, 110 Liquefy, 80, 127 Listeria, 11, 12t2.3, 14, 107t5.1, 133, 139 Listeria-like organisms, 133 Listeria monocytogenes, 2, 6, 7t2.1, 82, 110, 133, 145, 158 Listeria seeligeri, 11, 12t2.3 Listeriosis, 12–14, 158

Index Liver pate, 10 L(+)-lactate, 71 Log linear, 111 Log normal, 143 Log phase, 31, 104 Lot contamination, 120 Lots, 33, 121, 154 Lovell, R., 46 Lowman, R., 28 Low moisture baked products, 72 Low water activity, 10, 13, 67, 74 Lozano-Leon, A., 44 Luber, P., 17 Lucas, H. L., 134 Luchansky, J. B., 140 Lui, Y., 23 Lund, D., 112 Lynt, R. K., 26 M M. avium subsp. paratuberculosis, 43–47 costs, 44–45 disease, 44 food processing issues heat resistance, 45–46 reason for lack of information about MAP, 46–47 organism, 43–44 reservoirs, 45 Ma, L., 83, 134, 151 Madden, R. H., 18 Mafart, P., 28 Maintenance/repair practices creating growth niches/transmit microbes, 90–95 cealing leaks, bowed water-soaked ceiling tile, 94f4.20 deformed gasket, 93f4.19 failure to maintain seal on electrical box cover, 95f4.23 failure to replace torn hoses, 91f4.14 failure to replace torn or leaking gaskets, 90f4.13 penetrated hollow structures, 95f4.22 poorly sealed roof, 93f4.18 rusted opened back motor fan cover, 92f4.17 torn insulation, 92f4.16 unsealed, pen-sized wall penetration, 94f4.21 wet ceiling tiles, 91f4.15 Maltodextrin, 43 Malty, 68–70

185 Malty aroma, 69 Malty flavor, 70 Marinescu, M., 38–39 Marth, E. H., 14, 82, 106–107, 112–113, 134, 148 Marzipan, 73 Masaki, H., 41 Mass spectroscopy, 75 Mastitis, 21 Mathur, R. P., 41 Maximal temperature, 110 Mayonnaise, 10, 34 McClelland, M., 163 McFadden, J. J., 44 McIngvale, S. C., 31 Mead, P. S., 1, 8, 9, 12, 23, 29, 33 Meat, 11, 14, 16, 26, 33, 37, 65–67 Meat, spoilage of, 65–67 Meng, J., 34 Meningitis, 12–13, 40–41 Mesophiles, 64 Mesophilic aerobic spores, 134 Metabisulfite, 28 Metal scrapers, 129 Metaxopoulos, J., 66 Methicillin-resistant S. aureus (MRSA), 20 3-Methylbutanal, 69 Methyl ketones, 69 Methylmercaptan, 65 Metz, H., 40 Microbial growth, survival, death, factors are required for, 103–106 extrinsic factors impact of temperature on microbial death, 110–112 temperature, 110 interactions between intrinsic and extrinsic factors, 112 hurdles, 113–114 intrinsic factors acidity, 106–107 competition, 110 inhibitory substances, 107–108 oxygen and oxidation/reduction potential, 108–110 water activity (aw ), 106 Microbial growth niche, 45, 80, 81, 92f4.16, 99f4.30, 126 Microbial growth to oxygen, relationship, 109f5.4 Microbial population growth, 105f5.2 Microbial size relationships, 104f5.1

186 Microbial spoilage problems in processed foods, solving, 63–65 emulsions, 74–75 isolation and identification of spoilage organisms, 75 spoilage of bakery products, 72–74 spoilage of beverage products, 71–72 spoilage of canned foods, 74 spoilage of dairy products, 68–71 spoilage of fruit and confectionery products, 74 spoilage of processed meat, poultry, and fish, 65–67 Microbiological criteria guidelines, 42 microbiological limit, 142, 151 specifications, 142 establishing specifications, 142 standards, 142 Microbiological limit (microbiological criteria), 142, 151 Micrococci, 69 Microns (“µ”), 104 Midura, T. F., 145 Milk, 14, 22, 40, 69–70, 112 Milk powder, 10, 112 Millson, M., 17 Mincemeat, 73 Mitscherlich, E., 82, 112 MLST (multilocus sequence typing), 163–164 MLVA (multilocus variable-number tandem repeat analysis), 164 Modified atmosphere packaged (MAP), 64, 65 Modified Oxford Medium, 133 Moir, C. J., 23 Moisture, 27, 68, 72, 106, 126, 167 Moisture (factors influencing growth), 80, 106 Molds, 28, 66t3.1, 72t3.3 Molecular subtyping, 158 DNA sequencing-based subtyping and characterization methods, 163–164 PCR methods generating fragment length polymorphisms, 163 pulsed field gel electrophoresis (PFGE), 161–162 ribotyping, 162–163 Molecular subtyping of bacteria, value and methods for principles of molecular subtyping methods, 160 overview of molecular subtyping genetic techniques, 160–161

Index overview of molecular subtyping methods, 161–164 users of molecular subtyping methods, 165–166 company response to molecular or subtyping data, 170–173 foodborne illness outbreak investigation, 166–168 in-plant investigation, 168–170 value of molecular subtyping methods for food industry, 157–158 evolution of molecular subtyping methods, 159–160 limitations of traditional cultural methods, 158–159 Molin, N., 28 N -Monoalkyl maleates and fumarates, 28 Monroe, P. W., 41 Moore, G., 17–18, 129–130 Moore, J. E., 18 Moraxella, 65, 67 Moreno, G. S., 16 Morgan, D., 17 Mortimore, S., 29, 148 Mosso, M. D. L. A., 40 Motile, 8, 15, 38 Motor fan cover, 92f4.17 Mucor, 68t3.2, 70 Mucosal inflammation, 9 Muir, D. D., 70 Multilocus enzyme electrophoresis, 159 Multilocus sequence typing, 159, 163 Mung bean sprouts, 40 Murphy, C., 17 Murrell, W. G., 28 Muscle cramping, 21 Mustard, 10, 24 Mustard dressing, 10 Mutton, 10 Muytjens, H., 41 Muytjens, H. L., 41 Mycobacteria, 43, 44 Mycobacterium bovis, 43, 45 Mycobacterium tuberculosis, 43–44, 46 N Naser, S. A., 44 Nausea, 9, 21, 23, 37 Nazarowec-White, M., 40 Necrotic enteritis, 36 Necrotizing enterocolitis, 40 Neelam, M., 41 Negative controls, 119 Neonatal sepsis and meningitis, 40

Index Neosartorya glabra, 74 Nephropathy, 32 Neurological impairment, 13, 41 Neurotoxins, 25–27 Newell, D. G., 160 Ng, H., 134 Niches (microbial growth niche), 45, 80, 81, 92f4.16, 99f4.30, 126 Nisin, 28 Nitrites, 28 Noble, M., 44 Non-motile, 65 Non-proteolytic, 25–27 Non-starter lactic acid bacteria (NSLAB), 70 Noriega, F. R., 41 Nutrients (factors influencing growth), 43, 68, 114, 127 O Obligate aerobes, 109 Obligate anaerobes, 109 Odors, 26, 67, 71, 73 Oenococcus oenos, 72t 3.3 Off-flavors, 67, 69, 72t3.3 Oh, S. W., 40 Oliguria, 9 Olive, D. M., 160 Olsen, S. J., 168 One class (sampling plan), 143 Open sores, 22 Operating practices creating growth niches/transmit microbes, 84–89 failure to adequately disassemble, clean, and inspect behind diaphragm of pump, 87f4.8 failure to adequately disassemble horizontal agitator shaft coupling for cleaning, 87f4.9 failure to clean underside of product conveyor and thoroughly dry afterward, 88f4.11 failure to control dust accumulation, 86f4.6 failure to disassemble pumps and gaskets adequately for cleaning, 84f4.1 failure to inspect and clean air line boxes, 86f4.7 failure to maintain segregation of work boots from street shoes, 85f4.5 floor scrubber, 85f4.4 inappropriate use of broom to move standing water, 84f4.2 storage of cleaned and dried equipment near wet, 89f4.12

187 storage of pallets, boxes, and packaging material on wet floor, 88f4.10 uncleaned areas underside of floorassociated materials, 85f4.3 Operational samples, 123 Optimal temperature, 20, 27, 114 Orange juice, 75 Organic acid preservatives, 107 Organic acids, 35, 107 Organoleptic, 2, 64, 65 Osmophilic yeasts, 73, 107t5.1 Osmotolerant, 73 Osteomyelitis, 21 Oxidation/reduction potential (ORP/ Redox, E h ), 28, 106, 108–110, 112 factors influencing growth, 79, 107, 126 Oxygen, 18, 28, 64 Ozsan, K., 41 P P. guillermondii, 68t3.2 P. mephitica, 70 P. nigrifacines, 70 P. norvegensis, 68t3.2 Packaging, 22, 65, 67 Packaging material (unsanitary operating practices relating to), 67, 72, 88t4.10 Pallets (unsanitary operating practices relating to), 81, 88f4.10 Palop, A., 74 Palumbo, S. A., 29, 31, 35 Paprika chips, 10 Parabens, 28 Paralysis, 15, 26 Parasites, 103 Park, C. E., 17 Park, J., 40 Park, S. F., 18 Parsley, 17 Particulates, 71 Passover season, 76 Pasta, 23, 40 Pasteurization, 18, 28, 45–46, 69, 70, 134 Pasteurized milk, 7t2.1, 10, 14, 45, 69 Pathogen, enhanced detection of, 140 Pathogen of concern, 148, 154 Paton, A. W., 33 Paton, J. C., 33 Patton, C. M., 31 PCA pour plates, 129 PCA spread plates, 129

188 PCR (polymerase chain reaction), 38, 47, 75, 140, 163 Peanut butter, 10, 164 Peck, M. W., 28 Pediococcus, 66, 71, 134 Peeling paint and rust (unsanitary maintenance and repair practices), 81, 91f4.15, 92f4.17 Peleg, M., 28 Penetrations, 11, 81, 94f4.21, 128 Penicillium, 67, 68t3.2, 70, 73t3.4 Peppers, 26 Percent moisture content, 106 Personnel, 2, 97, 117, 122, 128, 158 Pet food, 10 PetrifilmTM , 129 pH, 106–107 factors influencing growth, 126 limiting, 107, 108t5.2 Phage, 10, 159 Phage typing, 159 Phagocytic cells, 9, 13 Phantom TDT curve, 111 Phenolic antioxidant/s, 28 Phlebitis, 21 Phosphates, 28, 114 Phospholipase, 69 Photobacterium, 67 Physicochemical properties, 127 Pichia burtonii, 73 Pichia fermentans, 68t3.2 Pierson, M. D., 145 Pipe, 40, 76, 81 Pipe brushes (unsanitary operating practices relating to), 81 Pipe interiors, 76 Plate count agar (PCA), 129 Plowman, J., 28 Plumbs, 108t 5.2 Pneumonia, 21 Polymorphism, 161, 163–164 Polyphosphates, 28 Polysaccharides, 73, 80, 127 Pork, 14, 17 Portioning, 65 Positive controls, 119 Post-cook, 11 Post operational (sampling/testing), 117 Post-process contamination, 6, 67, 69, 83 Post-processing environment, 125 Potatoes, 17, 23, 108t5.2 Potato flour, 40 Potato salad, 10

Index Poultry, 10, 14, 17–19, 39, 65–67 Poultry, spoilage of, 65–67 Poultry products, causative spoilage agents and manifestations of microbial deterioration in, 66t3.1 Powdered milk, 40 Pregnant, 13 Pre-operational (sampling/testing), 117, 118, 132 Pre-operational swabs, 132 Preventative maintenance, 69 Primers, 163 Principle investigator (PI), 121 Probabilities, 138, 142 Probiotics, 17 Process cheese, 13 Process failure events, significance of, 83 Product contact, 43, 121, 166 Production schedule, 120 Productivity, 2 Product stream, 82, 118, 122, 134 Prostration, 21 Protease activity, 28 Proteases, 26, 28, 70 Proteinaceous, 127, 130, 132 Protein detection kits, 129 Proteins, 10, 25, 44 Proteolysis, 112 Proteus, 70 Pseudomonas, 65, 66t3.1, 67, 68t3.2, 69, 70, 72t3.3, 109 Pseudomonas fragi, 69 Psychrobacter, 65 Psychrotrophic, 23, 65, 69 Psychrotrophs, 64, 110 Puddings, 23 Pulsed electric fields, 112 Pulsed field gel electrophoresis (PFGE), 161–162 Pulsed light, 112 PulseNet, 158, 159 Pulsotypes, 161 Pumpkin, 2, 108t5 Pumps (unsanitary operating practices relating to), 76 Putrid, 68t3.2, 69 Pyruvic acid, 70 Q Quality assurance, 122, 169 Quality systems, 119–120 Quantitative analyses, 128 Questionable results, 141–142

Index R Racemase, 71 Radcliff, R., 5–47 Radish, 17 Radish sprout/s, 34 Rajmohan, S., 69 Ralyea, R. D., 170 Ranch dressing, 34 Rancidity, 69, 70 Randazzo, C. L., 159–160 RAPD (random amplified polymorphic DNA), 161, 163 Rats, 41 Ray, B., 15 Read, T. D., 161 Ready-to-eat/heat and serve bakery products, causative spoilage agents and manifestations of microbial defects in, 73t3.4 Ready-to-eat (RTE) meat, 6, 11, 13, 65, 66t3.1, 73t3.4, 140 causative spoilage agents and manifestations of microbial deterioration in, 66t3.1 Recalls (product), 2, 14, 167, 173 Record-keeping, 118 Red meat, 39 Redox, see Oxidation/reduction potential (ORP/Redox, E h ) Refrigeration, 25, 28, 35, 63, 106 Reij, M. W., 6, 7 Renal failure, 32 Rendered animal proteins, 10 REP-PCR (Repetitive Extragenic Palindromic Sequences, PCR), 159, 161, 163, 169–170 Reproducibility, 159, 163 Reservoirs, 10, 14, 16, 22, 24, 26, 33–34, 37, 45 Resistant microbial populations, 153 Restaino, L., 40 Restriction digestion, 161 enzymes, 161–162 Restrooms (unsanitary design), 82 Rhizopus, 70 Rhodehammel, E. J., 24 Rhodotorula, 67, 69t3.2 Rhodotorula mucilaginosa, 66t3.1 Ribotyping, 162–163 Rice, 23–24 Rice flour, 40 Rice starch, 40

189 Rickettsia, 103 Riordan, T., 17 Risk assessment, 13, 24, 132 Risk of product contamination, 82 Risk rank (ingredients), 151 Ritchie, J. M., 31, 33 Roasting, 135 Roberts, T., 2, 8, 15, 30 Robertson, L. F., 40 Robinson, J. E., 41 RodacTM , 129 Rodgers, S., 28 Rollers, 132 Roman, D. J., 18 Roof, 11, 93f4.18 Roof penetrations (unsanitary maintenance and repair practices), 81 Ropey bread, 73 Rossiter, C. A., 45 Rotation of sanitizers, 80, 127, 153 Ryser, E. T., 5–47, 112 S Saccharomyces, 73t3.4 Saccharomyces bailii, 132 Saccharomyces cerevisiae, 68t3.2, 71, 73t3.4, 75 Sacks, J. J., 17 Sahin, O., 16–17 Salad, 10, 34, 76–77, 168–170 Salad base, 10 Salad dressings, 76–77, 168 Salami, 34, 67 Salkin, D., 47 Salmonella, 6–11, 18, 109, 112, 133 cost, 8 disease syndromes, 8–9 dry foods, 10 food processing environments, 11 foods associated with human infection, 10 infectious dose, 9 organism, 8 reservoirs, 10 Salmonella berta, 8 Salmonella serotype Enteritidis, 9, 10 Salmonellosis, 8, 10, 12, 166–168 Salt, 11, 22, 27 Salting, 63 Salt/moisture (S/M), 70 Sample collection, 121, 131t7.2 Samples, 3, 36, 119–122, 128–129 Sampling, 3, 121–122, 123 investigational, 139–140 and resampling, statistics, 137–139

190 Sampling plans category I (FDA plans), 144 category II (FDA plans), 144 category III (FDA plans), 144 one class, 143 two class, 143 types of attributes plans, 143 FDA plans, 144–145 ICMSF plans, 143–144 variables plans, 145 Sanders, G. W., 17 Sandwiches, 34 Sandwiched regions, 129 Sandwich fillings, 22 Sanitation, 22, 64, 122–123 Sanitation breakpoint, 120 Sanitation standard operating procedures (SSOP), 148 Sanitization, 18, 19, 35, 127, 129, 132 Sanitizer, 80, 127, 153 Santos, C., 41 Sauces, 23, 75 Sausages, 26, 34, 40, 67, 135 Scherer, K., 18–19 Schindler, P. R., 40 Schoeder-Tucker, L., 39 Schoeni, J. L., 23, 25, 29 Scoops, 129, 131t7.2 Scott, W. J., 28 Scribe, 121, 122 Seafood, 14, 17 Seizures, 32 Septicemia, 9, 12 Sequencing, 164 Serologically defined, 8 Serological reactivity, 20 Serotypes, 8, 11, 38, 158 Setlow, P., 25 Shapiro, R. L., 25–26 Sharma, S. K., 27 Shelf-life, 69, 73 Shelf-stable, 66t3.1, 67 Shellfish, 14, 17, 29 Shewanella, 67 Shewanella putrefaciens, 66t3.1 Shigella, 8t2.2, 33, 168 Shoulder, 111 Shreeve, J. E., 17 Siegel, L. S., 28 Sieve-type impactors (air sampler), 129 Silliker, J. H., 128 Simmons, B. P., 41

Index Simmons, N. A., 45 Skelly, C., 16 Skin lesions, 21 Skinning, 65 Skunk-like aroma, 70 Slicing, 65 Slicing machines, 132 Slime, 64, 65, 67, 76 Slimy gelatinous, 68t3.2 Slit-type impactors (air sampler), 129 Smeets, L. C., 41 Smith, J. L., 15–16 Smoked meat, 46 Smoke house treatments, 46 Smoot, L. M., 145 Snygg, B. G., 28 Sodium benzoate, 71 Sodium chloride (NaCl), 18, 28 Soils, 14, 24, 26, 130 Solubilize, 127 Sorbates, 28, 70, 71 Sorbitol, 29 Soriano, J. M., 40 Soups, 10, 23 Sources (contamination), 11, 39 air, 80 garments, 79 ingredients, 79, 82, 83 skin, 79, 81 Souring, 66t3.1 Sour milk, 14 Sour tea, 40 Southern blot, 162 Soy, 24, 40 Soy protein, 43 Speck, M. L., 134 Sperber, W. H., 151 Spices, 24, 40, 73, 76 Spinach, 17, 34 Split curd, 68t3.2 Split defects, 71 Spoilage, 2–3, 63–77, 168–169 Spoilage organisms, isolation and identification of, 75 Spoilage yeasts, 67, 71, 107f5.1 Sponge, 128 Spores, 23, 28 Sporotrichum carnis, 67 Spratt, B. G., 164 16S rDNA gene, 159–160 Stabel, J. R., 44–47 Stable, J., 46 Stale, 69

Index Stanbridge, L. H., 65 Standard operating practices (SOP), 123 Standards (microbiological criteria), 142 Staphylococcal enterotoxins (SE), 19–20 Staphylococcal food poisoning (SFP), 19–22 Staphylococcus, 19–22 costs, 21 disease syndromes, 21 organism, 19–20 reservoirs, 22 staphylococcal enterotoxin, 20 toxic dose, 21 Staphylococcus aureus, 2, 6, 7t2.1, 19–22, 107t5.1, 127 culture age on heat resistance, 105f5.3 Staphylococcus epidermidis, 19–20 Starter, 70, 71 Stationary phase, 27, 31 Statistically based finished product testing, 150 Statistical process control (SPC), 154, 155 Statistics, 130, 137–139 Steam, 35 Steam capper, 74 Steels, H., 71 Stern, N. J., 17–18 Stiles, M. E., 65 Stoll, B., 41 Storage, 81 Stratford, M., 71 Streptococci, 19, 109 Strockbine, N. A., 160 Styes, 21 Subtyping, molecular, 75, 120, 157–173 Sucrose, 36, 43, 76 Sugars, 26, 40, 64, 107t5.1 Sulfur compounds, 67 Sulliman, S. M. A., 41 Sung, N., 43, 45 Superantigens, 21 Suppliers, 152 Support bases, 98 Suppurative (pus-forming) infections, 21 Surface, 18, 22, 35, 73–76 Surface charges, 127 Surface taint, 70 Surfactant, 74 Surrogate microorganisms, 83, 119, 151 Surrogate organisms (CCPs), 47 Surveillance, 23, 158 Swabbing, 169 Swabs, 34, 76, 128–129, 132

191 Swaminathan, B., 158–159, 162 Symptoms, 9, 15, 23, 32, 37 onset of, 15, 21 T Tail (heat resistant), 111 Taints, 71 Tamplin, M. L., 32 Tamura, A., 40 Tanaka, N., 114 Taurine, 67 Tauxe, R. V., 17 Team/sampling team, 117, 121–122 Temperature, 18–20, 27–30, 74, 82–83 factors influencing growth, 110–112 Tenover, F. C., 162, 165 Testing laboratory, 119, 140, 165 Thermal death time (TDT), 11, 76 Thermal processes, 106 Thermoduric bacteria, 110 Thermoduric micrococci and lactococci, 69 Thermophiles, 110 Thermotolerant Streptococcus, 69 Thickening agent, 75 Thompson, J. M., 73 Threlfall, E., 160 Thrombocytopenia, 32 Tift, W. L., 41 Time (factors influencing growth), 111–112 Tines, 132 Tompkin, B., 67 Tongue blade, 129 Top, J., 164 Torn insulation, 81, 92f4.16 Torulaspora delbrueckii, 68t3.2 Torulopsis, 68t3.2 Touring (the plant), 121 Townsend, S., 41 Toxic dose, 21 Toxic metabolites, 109, 110 Toxico-infectious agents, 5, 6 Toxico-infectious pathogen, concern to food processors (industrial) C. perfringens, 35–37 enterohemorrhagic E. coli, 29–35 Toxigenic pathogen, concern to food processors (industrial) B. cereus, 22–25 C. botulinum, 25–28 Staphylococcus, 19–22 Toxigenic pathogens, 6 Toxigenic vs. infectious bacterial pathogens, 6 Toxin, 6, 15, 19–21, 24

192 Toxin shock syndrome (TSS), 20 TPP, 32 Trachoo, N., 18–19 Tradition, 83, 118, 151 Traffic patterns, 123 Training, 43, 122 Transplacental migration, 13 Trend analyses, 118 Trichosporon spp., 67 Trimethylamine, 67, 75 Trimethylamine oxide, 67 Trisodium phosphate, 18 Truncated Listeria assay, 133 Tuncer, I., 41 Tuttle, J., 30, 32–33, 35 Two class (sampling plan), 143 U Ultraviolet light, 112 Unclean utensils, 73 Undercooked, 17, 29 Undissociated acid, 107, 108 United States Department of Agriculture (USDA), 8, 13, 37, 83, 119, 151 Unsanitary design of factory or its equipment air flow, 82 air handling units (AHUs), 81, 82 ball–valves, 82, 100f4.32 catch pans, 81, 82 common areas for raw and finished product, 82 cooling coils, 82 disassembly, 81, 84f4.1 electrical charging rooms for electrically powered vehicles, 82 elevators, 82, 97f4.27 final filter, 82 loading and receiving docks, 82 restrooms, 82 untrapped catch pans, sinks, lubrication, etc. water to drains, 81, 96f4.24 Unsanitary maintenance and repair practices broken gasket seals on product pumps, 81 condensate catch pan drains, 81 duct tape, 81 dust, 81, 86f4.6 exposed product, 80 false ceilings, 81, 91f4.15, 94f4.20 fatty residues, 80 forklift, wheeled vehicle traffic, 80, 82 gaskets, valves, 80–81, 84f4.1, 93f4.19 germicidal footbaths, 80

Index glove (use), 80 high pressure hoses, aerosolization, 80 hoses, 80–81, 91f4.14 impact wrenches (effect of improper use on gaskets), 81, 93f4.19 insulation, 81, 92f4.16 maintenance worker attire, 81 packaging material, 81, 884.10 pallets, 81, 88f4.10 peeling paint and rust, 81 pipe brushes, 81 point of use filters on compressed air-lines, 81, 99f4.31 walls and ceilings, 81, 94f4.21 water (inadequate temperature), 80 welds, 81 Uremia, 9 Uric acid, 67 Urinary tract infections, 21 Urmenyi, A. M. C., 41 V Validation, 83, 123 Validation (re-validation), 150–151 Valliant, V., 41 Valves, 76 Van Acker, J., 41 Van Belkum, A., 160 Vandamme, P., 38 Van Kruiningen, H. J., 44 Van Os, M., 41 Variables plans, 143, 145 Vegetable oil, 43 Vegetables, 14, 17, 23, 34 Verification procedures, 118, 148, 150 Vibrio, 67 Vibrio parahemolyticus, 108t5.2 Viruses, 103–104 Vitamin, 43 Volatile amines, 67 Vomiting, 9, 12, 21, 23 Vought, K. J., 9 W Wagenaar, J. A., 18 Wallace, C., 148 Wallace, F. M., 140 Wallemia sebi, 73 Walls and ceilings (unsanitary maintenance and repair practices), 81, 94f4.21 Wang, W. L., 18 Water, 34, 39, 72t3.3

Index contaminated/unchlorinated, 11, 16–17, 20, 24, 167 inadequate temperature (unsanitary operating practices relating to), 80 Water activity, see aw (water activity) Water-in-oil emulsion, 70 Wayne, L. G., 46 Weinstein, P., 16 Weir, E., 41 Weissella viridescens, 65, 66t3.1 Weissman, J. B., 10 Welds (unsanitary maintenance and repair practices), 81 Wells, S. J., 44 Welsh, J., 163 Wesley, I. V., 38–39 Wet heat, 106 Wetzel, R. G., 41 Wheeled containers, 132 Whey protein, 43 Whiskers, 66t3.1 White, P. L., 18 Whiting, R. C., 27 Wiedmann, M., 160–161, 163–164 Williams, C. E., 15 Williams, J. G., 163 Willis, J., 41 Wilson, I. G., 17 Wolter, H., 67

193 Wong, A. C., 23 Wong, A. C. L., 23, 25 Woodford, N., 160 Workers, 80, 131 X Xerophilic fungi, 74 Xerophilic molds, 73, 107t5.1 Xerotolerant, 73 X-ray, 112 Y Yan, Z., 5–47 Yanmin, H., 46 Yarrowia lipolytica, 68t3.2 Yeasts, 7t2.1, 67, 73 Yersinia enterocolitica, 7t2.1, 82, 110 Yogurt, 7t2.1, 68t3.2, 107 Z Zink, D., 42 Zogaj, X., 41 Zones (proximity to product), 82–83 Z -value, 111 Zwietering, M., 42 Zygosaccharomyces, 73t3.4 Zygosaccharomyces bailii, 71, 75 Zygosaccharomyces lentus, 71 Zygosaccharomyces rouxii, 71, 73, 75