v , . ‘ ‘ 9.2 A, ‘. a... ,.‘:,.L ixvuu wfiwfiy w...»w...,.a.a..‘...u: wfifigfixfim imbfimfi, . s 4 «gamma am , . r . 3 «3% ‘u. IR 3.: .. . , :21... , .. ‘ .ri 1 .- mats LIBRARY 2 Michigan State My?) University This is to cettify that the dissertation entitled EFFECTS OF MARINATION ON SALMONELLA PENETRATION AND MUSCLE STRUCTURE OF TURKEY BREAST presented by VAREEMON TUNTIVANICH has been accepted towards fulfillment of the requirements for the Doctoral degree in Food Science and Human Nutrition Major ProfessoFs'SignaturB 5/5/08) Date MSU is an afiinnative-action, equal-opportunity employer PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. | MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE L M ”3.1 8:110 5/08 K:IPro;/Acc&Pres/ClRC/DateDue indd EFFECTS OF MARINATION ON SALMONELLA PENETRATION AND MUSCLE STRUCTURE OF TURKEY BREAST By Vareemon Tuntivanich A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements For the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 2008 ABSTRACT EFFECTS OF MARINATION ON SALMONELLA PENETRATION AND MUSCLE STRUCTURE OF TURKEY BREAST By Vareemon Tuntivanich The effects of marinade composition, microbial load, and point of exposure on the multidirectional penetration of Salmonella into intact whole muscle turkey during marination were evaluated. Irradiated whole muscle turkey breasts were vacuum tumbled for 20 min with marinade solutions containing an 8—strain Salmonella. Each sample was dissected to collect multidirectional Salmonella concentration data. Three 1 cm-thick slices (cranial, middle, and caudal) were removed from the turkey breast longitudinal to the muscle fibers. Within these three slices, l-cm cubes were excised using a self- cauterizing electrosurgical unit and enumerated for surviving cells. The highest and lowest Salmonella counts were observed in the area immediately below the inoculated surface and in the center segment, respectively. There was a slight difference in Salmonella distribution for the cranial and caudal slices as compared to the middle slice, which may be due to the typical structure and orientation of turkey breast muscle. With higher inoculation levels (108 vs. 104 CFU/mL), higher numbers of salmonellae (P3005) were recovered from the turkey breast. In contrast, marinade containing 3.2% salt and 0.8% phosphate (typical concentration, TC) or 10.2% salt and 4.05% phosphate (high concentration, HC) showed no differences (P>0.05) in Salmonella populations after penetration into whole muscle. Greater Salmonella penetration (P3005) into marinated turkey breast was observed when the pathogen was introduced during vacuum tumbling with a contaminated marinade, compared to pre- and post—tumbling process. This may be due to the lower bacterial levels to start with in pre- and post- process treatments, as opposed to the during process treatment. The changes in turkey muscle microstructure after marination were also documented. Salt and phosphate significantly contributed to changes in muscle microstructural changes. Muscle samples immersed in HC demonstrated smaller extracellular areas as compared to control and TC samples. Swelling of myofibrils in TC and HC were observed under transmission electron microscopy. Cryostat sectioning and fluorescence microscopy were used to determine the location of marker bacteria (GFP- transformed Salmonella) in fresh turkey breast muscle. In cross-sectional areas of muscle cells, GFP-labelcd salmonellae were seen between muscle fibers and bundles and tended to be located near endomysial and perimysial supporting tissue. These results suggest the potential for Salmonellc'z to penetrate and reside at specific locations in value-added or marinated products. The results provide important information concerning product and process safety, which is necessary to develop a quantitative model for pathogen penetration into whole muscle poultry products during marination. To my family who are giving me the endless love. iv ACKNOWLEDGEMENTS I would like to express my ultimate gratitude to my major academic and research advisor, Dr. Alden Booren, for taking me as his student as well as for his guidance, encouragement and support throughout my degree. I would also like to express gratitude to Drs. Bradley Marks, Matthew Doumit, Elliot Ryser, Alicia Pastor, and Alicia Orta- Ramirez for serving on my guidance committee and for all the helpful meetings and suggestions for my research. Sincere thanks and appreciation are given to Drs. Shirley Owens and Melinda Frame for their microscopy technique expertise and Dr. Dan Guyer for his kindness to walk me through MatLab program. I would like to express my special thanks to Jennifer Dominguez and Tom F orton who helped me with several tasks in Meat Processing Lab. I am also appreciated for Dr. Gale Strasburg for his advices and suggestions through out my career at Michigan State University. I would like to express sincere thanks and appreciation to Dr. Alicia Orta-Ramirez not only for all her support and enthusiasm for this project but also for sharing time, energy and knowledge with me. Special acknowledgement is also expressed to Dr. Alicia Pastor and Dr. Ewa Danielewicz who have been mentors to me through out my program. Lastly, I would like to acknowledge all my past and present colleague in my laboratory for all their support. TABLE OF CONTENTS LIST OF TABLES ................................................................................ viii LIST OF FIGURES ................................................................................ ix CHAPTER 1 INTRODUCTION .................................................................................. 1 CHAPTER 2 LITERATURE REVIEW ................................................................................................. 4 Turkey consumption ........................................................................ 4 Skeletal muscle structure and biochemistry .............................................. 5 Non-meat ingredients and meat marination ............................................. 7 Salmonella spp .............................................................................. 10 Bacterial penetration ....................................................................... 15 Bacteria] attachment ........................................................................ 18 Bacteria] thermal inactivation ............................................................ 19 Microscopy in food research .............................................................. 22 CHAPTER 3 EFFECTS OF MARINATION ON SALMONELLA PENETRATION AND MUSCLE STRUCTURE OF TURKEY BREAST .......................................... 26 Introduction ......................................................................................... 26 Materials and methods ........................................................................... 29 Multidireetional penetration of Salmonella into turkey breast ....................... 29 Preparation of turkey breast samples ........................................... 29 Marinade preparation ............................................................. 29 Preparation of test organisms .................................................... 3O Exposure to inoculated marinade ................................................ 31 Sampling and recovery ........................................................... 31 Data analysis ....................................................................... 34 Microstructure of marinated whole muscle turkey breast .............................. 34 Preparation of turkey breast samples ........................................... 34 Sample preparation for bright-field microscopy (BFM) ..................... 34 Sample preparation for transmission electron microscopy (TEM) ......... 36 Salmonella location in fresh whole muscle turkey breast tissue ..................... 37 Preparation of fresh whole muscle turkey breast tissue ...................... 37 Preparation of green fluorescent protein (GFP)-Salm0nella inoculum. . ..38 Preparation of GFP-Salmonella inoculated marinade ........................ 38 Cryostat sampling of inoculated sample block ................................ 38 Results and discussion ............................................................................ 40 vi CHAPTER 4 SUMMARY AND OVERALL CONCLUSIONS ........................................... 73 CHAPTER 5 FUTURE RESEARCH ........................................................................... 75 APPENDICES APPENDIX A: THERMAL INACTIVATION OF SALMONELLA IN WHOLE MUSCLE AND GROUND TURKEY BREAST ............................................. 78 Abstract ...................................................................................... 78 Introduction ................................................................................. 80 Materials and methods ..................................................................... 82 Preparation of turkey breast samples ........................................... 82 Preparation of marinade solution ................................................ 83 Preparation of test organisms .................................................... 83 Preparation of inoculated marinade ............................................. 84 Exposure to inoculated marinade ................................................ 84 Thermal inactivation .............................................................. 85 Sampling and enumeration ...................................................... 86 Data analysis ....................................................................... 86 Results and discussion ..................................................................... 87 Conclusions ................................................................................. 94 APPENDIX B: SALMONELLA VIABILITY IN SALT AND PHOSPHATE MARINADES AS AFFECTED BY STORAGE TEMPERATURE ..................... 95 Abstract ....................................................................................... 95 Introduction ................................................................................. 96 Materials and methods ..................................................................... 98 Marinade preparation ................................................................................ 98 Preparation of test organisms ..................................................... 98 Preparation of inoculated marinade ............................................. 99 Sampling ........................................................................... 99 Data analysis ....................................................................... 99 Results and discussion ........................................................................................ 100 Conclusions ................................................................................ 1 04 APPENDIX C: ADDITIONAL MICROSCOPY IMAGES .............................. 106 APPENDIX D: BIBLIOGRAPHY FOR APPENDICES A AND B .................... 108 BIBLIOGRAPHY ................................................................................ 1 1 1 vii LIST OF TABLES Table 3.1 Percentage of “white” pixels representing muscle tissue pore space as determined using Matlab software program .......................................... 61 Table A] P values from ANOVA of Salmonella spp. thermal inactivation of whole and ground turkey breast muscle ........................................................... 90 Table A2 Salmonella numbers in inoculated unheated and heated turkey breast samples at 55, 60, and 62.5 °C .................................................................. 90 Table A3 First-order thermal inactivation rate constants (k, min") determined from linear regression of Salmonella survivor data (Ln N/NO = -kt) for whole and ground turkey breast muscle ........................................................... 92 viii LIST OF FIGURES Figure 2.1 Diagrams of skeletal muscle organization (a) transverse section of muscle showing fascicles, muscle fibers, myofibrils, and myofilaments (b) myofilarnents identifying specific bands and lines .................................. 6 Figure 3.1 Sampling diagram (a) 3 slices, cranial, middle, and caudal, excised from marinated whole muscle turkey breast (b) l-cm cubes dissected from those slices. Cubes labeled R1, R3, L1, L3, T1, and B1 were dissected from cranial and caudal slices and segments labeled C, R2, R4, L2, L4, T2, and B2 were dissected from middle slice ...................................................... 33 Figure 3.2 Horizontal penetration of Salmonella into (a) cranial (b) middle, and (c) caudal slices of whole muscle turkey breast, exposed to 104 (open square) and 108 Salmonella CFU/mL (solid circle). . .......42 Figure 3.3 Salmonella penetration diagram in middle slice of turkey breast showing the highest population in the cubes immediately below inoculated surface. . . . . ...44 Figure 3.4 Vertical penetration of Salmonella into the middle slice of whole muscle turkey breast, exposed to 104 (open square) and 108 Salmonella CF U/mL (solid circle) ........................................................................... 45 Figure 3.5 Horizontal penetration of Salmonella into (a) cranial (b) middle, and (c) caudal slices of whole muscle turkey breast, exposed to marinade solutions containing 3.2% salt and 0.8% phosphate (open square) and 10.2% salt and 4.05% phosphate (solid c1rc1e)48 Figure 3.6 Vertical penetration of Salmonella into the middle slice of whole muscle turkey breast, exposed to marinade solutions containing 3.2% salt and 0.8% phosphate (open square) and 10.2% salt and 4.05% phosphate (solid circle) .................................................................................... 50 Figure 3.7 Horizontal penetration of Salmonella into (a) cranial (b) middle, and (c) caudal slices of whole muscle turkey breast, exposed to contamination before (open square), during (solid circle), and after (cross) vacuum tumbling ................................................................................. 53 Figure 3.8 Vertical penetration of Salmonella into the middle slice of whole muscle turkey breast, subjected to contamination before (open square), during (solid circle), and after (cross) vacuum tumbling process ............................... 55 ix Figure 3.9 Bright-field microscopy images of transverse sections of turkey breast whole muscles exposed to (a) control with no water, salt, and phosphate, (b) 3.2% NaCl and 0.8% alkaline phosphate, and (c) 10.2% NaCl and 4.05% alkaline phosphate ................................................................................. 59 Figure 3.10 Transmission electron micrographs of longitudinal sections of whole muscle turkey breast exposed to (a) control with no water, salt, and phosphate and (b) 3.2% NaCl and 0.8% alkaline phosphate, and (c) 10.2% NaCI and 4.05% alkaline phosphate ..................................................................... 63 Figure 3.11 Enhanced fluorescence images of GFP-labeled Salmonella location in turkey breast transverse at (a) 400X and (b) lOOOX magnification ................... 68 Figure A] Survival curves of Salmonella spp. in whole (solid circle) and ground (open square) turkey muscle subjected to isothermal heating temperatures; (a) 55, (b) 60, and (c) 62.5 °C. Each point is an average of 3 observations ............ 88 Figure 8.1 Viability of the mixture of 8 Salmonella serovars in typical marinade containing 3.2% salt and 0.8% phosphate maintained at 4 °C (solid circle) and 37 °C (solid square) and high concentration marinade containing 10.2% salt and 4.05% phosphate stored at 4 °C (open triangle) and 37 °C (cross) ....... 103 Figure C. 1-3 Enhanced fluorescence images of GFP-labeled Salmonella location (circle areas) in turkey breast transverse sections taken at 400x magnification. . ....107 CHAPTER I INTRODUCTION Turkey production in the United States has more than doubled since 1970. Per capita consumption ofturkey has increased from 2.9 kilograms in 1970 to 5.9 kilograms in 2005 (USDA 2007). In 2004, turkey was the number four protein choice for American consumers because of taste and nutritional value (National Turkey Federation 2004). Additionally, consumer trends show an increasing demand for ready-to-eat products, which include marinated whole muscle products (Russell 2002). Many poultry products sold in retail markets are value added by means of marination and mechanical tenderization. Vacuum tumbling is a popular method for tenderization. During the process, whole muscle meat is tumbled either with a marinade solution or after it is injected. Salt and phosphates are commonly incorporated into the marinade to increase product yield and palatability. They function by inducing changes in muscle structure and contributing to increased water absorption. However, if the product surface and/or marinade were contaminated, bacterial pathogens may be introduced along with the water portion of the marinade into the interior of \vl'alue-added meat products (Phebus and others 1999). Once these bacteria have penetrated into meat products. they may exhibit enhanced thermal resistance, which depends on various factors, including meat species, muscle type. pH, fat content, and additives used in processing. As a result, thermal processing using proper cooking time and internal temperature is important to assure product safety. Foodborne outbreaks associated with poultry products have involved several pathogens, such as Listeria nmnoeytogenes, Salmonella and C ampylobaeter. Salmonella is the specific organism targeted in this research since it is responsible for an estimated 1.4 million cases of foodborne illness each year in the US. and is commonly found in poultry products (CDC 2005c). According to the USDA Food Safety Inspection Service (F SIS), raw turkey is a common source of Salmonella in the US. food supply (USDA-FSIS 2002). About 20.3% of ground turkey sampled in 2006 was positive for Salmonella (USDA-FSIS 2007). An infective dose as little as 10 to 100 cells can lead to symptoms of salmonellosis (nausea, abdominal cramps, diarrhea and vomiting) in susceptible persons. To improve the microbiological safety of marinated poultry products, the overall goal of this project was to determine the basic pathways by which pathogens penetrate into marinated whole muscle poultry products. The term “penetration” is used throughout the study when the microorganisms are observed or recovered from the interior of turkey breast due to the migration and/or invasion of the bacteria. The hypotheses for this project are: l. Salmonellae follow a non-random penetration pathway, and the degree of penetration depends on several factors, such as marinade composition, microbial population, and point at which the product is exposed to Sahnonefla. 2. Once salmonellae penetrate into the interior of the muscle, the pathogen preferentially locates within the muscle bundles after marination. 3. Marinade ingredients, such as salt and phosphates, contribute to structural changes facilitating microbial penetration into the inner portion of muscle. Ix.) To test these hypotheses, the specific objectives were to: (I) evaluate the effects ofmarinade composition, pathogen load, and contamination stage on the multidirectional penetration of Salmonella into intact whole muscle turkey during vacuum tumbling marination; (2) document changes in turkey muscle microstructure after marination; and (3) determine the location of Salmonella in whole muscle turkey after marination with a Salmonella-inoculated marinade. In addition to these specific objectives, the effects of turkey physical structure on the thermal resistance of Salmonella were quantified and Salmonella viability in marinades containing salt and phosphate was also determined. These last studies are addressed in the appendices. (J) CHAPTER 2 LITERATURE REVIEW Turkey consumption More than 256 million turkeys were raised in the US. in 2005. According to the National Turkey Federation, 41.1% were sold to grocery stores and other retail outlets and 21.6% were distributed to foodservice outlets. From 1970 to 2005 consumption of turkey has more than doubled (USDA 2007). Nearly half of the US. population consumes turkey at least once every two weeks, with more'than a quarter eating turkey lunchmeat. Although, whole turkey continues to be the most popular, ground turkey has high appeal among all ages, genders, and economic levels. White and dark meat differ nutritionally. White meat has fewer calories and less fat than dark meat and is generally preferred in the US. rather than dark meat. A turkey typically has about 70% white meat and 30% dark meat. Today’s improvements in genetics, feed, and management practices have produced domesticated turkeys with more breast meat and meatier thighs. The "Go Lean with Protein" section of the food guide pyramid recommends eating two to three servings from the meat and poultry group each day. Therefore when comparing many other foods within this group, turkey can be a choice to provide high quality protein, low in fat, and fewer calories. A nutritional fact of boneless skinless cooked turkey breast contains 26 g of protein, which is about 8% higher than the same serving size of boneless skinless chicken breast or trimmed top loin beefsteak. In a progress report on Salmonella testing of raw poultry products from 1998 to 2006 (USDA-FSIS 2007), approximately 7.100 of2785 turkey carcasses and 20.3% of 444 ground turkey samples yielded Salmonella. Because food safety is a top priority of the US. meat and poultry industry, the pathogen reduction strategies in the workplace must coincide with the current regulations, process controls, and hazard analysis and critical control point (HACC P) programs to achieve the safest products possible for consumer. Skeletal muscle structure and biochemistry Three types of muscle tissues are distinguished based on functional and morphological characteristics: smooth muscle, cardiac muscle, and skeletal muscle. Skeletal muscle is responsible for body position and posture maintenance. Figure 2.1 a illustrates the structure of skeletal muscle, which is attached to bone via tendon and surrounded by a dense collagen called epimysium. Within a muscle bundle, muscle cells or muscle fibers are arranged in parallel and bound together by collagenous supporting tissue, the perimysium. A muscle fiber, which is surrounded by endomysium, is composed of longitudinally structural and functional subunits/organelles called myofibrils. The myofibrils are composed of myofilament polymers, referred to as thick and thin filaments. The predominant proteins in thick and thin filaments are myosin and actin, respectively and the overlapping of actin and myosin filaments gives rise to a characteristic banding called striated pattern of skeletal muscle cell and myofibril as shown in Figure 2.1 b (Ross and others 2003). epimysium sarcolemma SBTCOITICI'C myofllament muscle bundle muscle (a) thick filament . l' . 7 me A b d lb (1 thin filament an an L r——’% r—’% j ), I 7* _ ‘WT i Ly“; *—_——. .4“ T“ :) » t -_ /,.___ .._ ————— ' - — __.__—"" ‘ ‘——-"f ' :1.” __‘:> gr; _ H hand ‘— sarcomere —J M line (b) Figure 2.1 Diagrams of skeletal muscle organization (a) transverse section of muscle showing fascicles, muscle fibers, myofibrils, and myofilaments (b) myofilaments identifying specific bands and lines (adapted from Aberle and others 2001) Over 20 different proteins are associated with myofibril. The major proteins are myosin, actin, titin, tropomyosin, troponin, and nebulin (Aberle and others 2001). Actin and myosin comprise more than half ofthe protein ofskeletal muscle and are the main factors in postmortem onset and resolution of rigor, tenderness, water holding capacity, emulsification and binding properties of meat (Richardson and Mead 1999). The pH ofrnuscle during life is about 7.2. After death, the muscle acidifies to pH 6 or less through the accumulation of lactic acid, which derives from the postmortem breakdown of glycogen by glycolysis. Acidification may affect color and water holding capacity of meat. The development of rigor mortis also occurs when adenosine triphosphate (ATP) levels decline in the postmortem animal. The muscle consequently transforms from being soft and pliable to a more rigid and inextensible state, which affects textural qualities of meat after cooking (Foegeding and others 1996). By understanding the mechanism and key factors that influence muscle protein functionality, it is possible to manipulate product formulations and processing conditions to improve the quality of existing or new products. Non-meat ingredients and meat marination Other than the meat block, which is the major part ofthe product, non-meat ingredients can also be applied to achieve unique characteristics in the final product. They can be classified based upon their functional properties. These ingredients include spices and flavorings, sweeteners, salts including curing compounds, water, extenders and binders, starter cultures, alkaline phosphates, and synthetic antioxidants. Of these non-meat ingredients. salt and alkaline phosphates are of interest for this research. Many ofthe physical properties of meat and poultry, including color, texture, firmness, juicincss, and tenderness ofcooked meat, are partially dependent on the meat water-holding capacity. Higher water holding improves meat quality by contributing to consumer perception ofjuiciness. In addition, from the processor‘s viewpoint, increased water holding capacity is associated with increased product yield and therefore profits (Smith and Acton 2001). Marination is a process in which liquid and other non-meat ingredients are added to meat before cooking. Many commercial meat and poultry products are marinated by still marination, tumbling, blending and mechanical injection. Benefits from marination include increased product juicincss, tenderness, and yield. In addition, flavor is enhanced both by the addition of spices and flavorings and by reduction in rancidity that may develop during storage or after cooking. Salt and alkaline phosphates are basic non-meat ingredients that are commonly incorporated into the marinade solution. Sodium chloride is not only used for preservation and flavor enhancement but also for myofibrillar protein solubilization. Solubilized protein is important for processed meat products because it enhances binding and stabilizes water and fat in emulsified products. Salt also increases water holding capacity of processed meat products (Foegeding and others 1996). However, quality of the salt is an important concern because some salts might contain heavy metals, which accelerate lipid oxidation. Xiong and Kupski (1999) demonstrated that % salt promoted moisture absorption in chicken broiler breast. They explained that salt enhanced moisture retention because of an increase in capillary force that enabled water to be chemically bound or physically entrapped in the muscle. In addition, the effect of salt on 8 microstructure of pork was studied by Becker and other (2005). They observed expansion of muscle tissue when salt was added up to 3%. However, the structure started to shrink when salt concentration was doubled to 6%. They concluded that salt affects the secondary structure of myofibrillar protein. Alkaline phosphates also contribute to the increase in water retention in processed meat products. Alkaline pyrophosphates act as a fluidizing agent in muscle, dissociating actin and myosin, changing ionic strength and pH, which leads to increased water uptake (Xiong and Kupski 1999). The end result is less moisture loss during the cooking along with increased tenderness and juicincss. The final product yield is dependent on the type of alkaline phosphate, for instance, tetrasodium pyrophosphate, tetrapotassium pyrophosphate, sodium tripolyphosphate, and hexametaphosphate (Dziezak 1990). Alkaline pyrophosphates contribute to the highest final product yield and increase the rate of marinade absorption compared to tripolyphosphate and hexametaphosphate (Xiong and Kupski I999, Zheng and others 1999). The action of phosphates in improving water holding capacity appears to be twofold: firstly. increasing the pH, and secondly, causing an unfolding of muscle proteins, consequently creating more space among protein molecules and sites available for additional water to bind (Pearson and Gillett 1999). F roning and Sackctt (1985) suggested that phosphates increased the extractability of myofibrillar proteins, particularly the combination of tripolyphosphate and hexametaphosphate. The myofibrillar proteins have the ability to attract and bind water because ofthe increase in their negative charges and space between myofilaments. Based on USDA regulations, the total amount of phosphate added to the finished product can not exceed 0.5% (Smith and Acton 2001 ). Marinade pickup by the muscle thus determines the amount of allowable phosphate in the marinade solution. The target level of salt in the marinade or muscle is not regulated and is determined by the processor. Marinade pickup is calculated as follows: % marinade pickup = marinated weight — initial weight x 100 initial weight A combination of salts and polyphosphates in marinade solution are commonly used to enhance the tenderness of meat products (Richardson and Mead 1999). The effect of various combinations of salt (0-2.25% meat weight basis) and sodium tripolyphosphate (0-0.5% meat weight basis) was observed in restructured pork (Schwartz and Mandigo 1976). Use of these non-meat ingredients together at levels of 0.75% salt and 0.125% alkaline phosphate was recommended and desirable for water holding capacity, cooking yield, and juiciness of restructured pork. Salmonella spp. Foodborne disease is caused by consuming contaminated foods and beverages. The causative agents can range from microbial pathogens, which include bacteria, viruses, and parasites, to harmful toxins and chemicals. Campy/obaerer, Salmonella, and Escherichia coli 0157:H7 are the most commonly recognized foodborne infections caused by bacteria (CDC 2005b). Salmonella has been known to cause illness for over 100 years. It was discovered by an American bacteriologist named D.E. Salmon after whom it is named. Salmonella is widespread in the intestines of birds, reptiles, and mammals. In addition, it can also enter natural environments through human or animal excretion. Humans can acquire 10 Salmonella throuoh various meat and oultr roducts that have been mishandled or C improperly processed. Controlling and minimizing bacterial contamination is of interest among processors and food safety researchers because of the increase in foodbome outbreaks (FDA-CFSAN 1992). Characteristic o/‘Salmonella The genus Salmonella belongs to the family EnIerobacteriaeeae, known as enteric bacteria. Salmonella is a rod-shaped, nonsporeforrning, Gram negative, facultative bacterium, measuring 0.5 u x 2-3 u that exists singly, in pairs, or in short chains. Salmonella is a motile organism (with nonrnotile exceptions of S. gallinarum and S. pal/0mm) that typically possesses l-5 flagella. Over 2500 serotypes of Salmonella have been identified of which approximately 2000 serotypes are pathogenic to humans. Salmonella Typhimurium (S. Typhimurium) and Salmonella Enteritidis (S. Enteritidis) are the most common in the US. with half of all salmonellosis cases caused by these two serotypes (CDC 2005c). Salmonellae live in the intestinal tracts of warm and cold blooded animals including humans and birds and are widespread particularly in poultry and swine. Reptiles are likely to harbor Salmonella. Therefore, those who come into contact with these animals should follow good practices including handling and cleaning. Salmonella may be found in the feces ofsome pets, especially those with diarrhea. Environmental sources ofthe organism include water, soil, animal feces, insects, as well as commercial and home food preparation areas (FDA-CFSAN 1992). l l Transmission and mani/esiation ofIsa/moizellosis Salmonellosis is an infection caused by Salmonella. Transmission of the organism is through contaminated food, water, or contact with infected animals or their feces. Contaminated foods are often meat and poultry products, milk, eggs, and produce. Food may become contaminated by the unwashed hands of an infected food handler. Most persons infected with Salmonella have symptoms of diarrhea, fever, and abdominal cramps. The illness develops 12 to 72 h after infection and usually lasts 4 to 7 days. The infective dose can be as low as 15-20 cells depending on the health of the host. Most persons recover without treatment; however, the elderly, infants, and those with weakened immune systems are more likely to have a severe illness (F DA-CF SAN 1992). Those who tend to have severe diarrhea may need to be hospitalized. In these patients, Salmonella infection may spread from the gastrointestinal tract to the blood stream and then to other organs, even resulting in death unless the patients are treated with antibiotics. Salmonellosis outbreaks Approximately 1.4 million cases of salmonellosis occur per year in the US; of these about 30,000 are culture-confirmed cases reported to the Centers for Disease Control and Prevention (CDC), more than 500 are fatal cases, and 2% are complicated by chronic arthritis (CDC 2005c). Serotypes Typhimurium and Enteritidis have predominated in outbreaks with Typhimurium the most commonly isolated serotype since 1997. The three other most common serotypes are Enteritidis, Newport, and Heidelberg (cnc 2005a). Poultry is an important source of human Salmonella infections. The following are examples of several outbreaks associated with poultry consumption: - In 1982, outbreaks of foodborne salmonellosis caused by improperly cooked and stored poultry giblets for Thanksgiving occurred in Maine. One hundred-twelve culture- confirmed cases of Salmonella Enteritidis were identified in patients after consuming turkey from a restaurant. - In 1985, an estimated 351 children and staff at a Georgia elementary school developed gastroenteritis. Salmonella Enteritidis was isolated from more than 100 children. The illness was strongly associated with eating turkey salad from the school cafeteria. - In 1986, 2130 students and employees of an Oklahoma public school system developed symptoms of salmonellosis including diarrhea, nausea, vomiting, abdominal cramps, and fever. While consumption of chicken was associated with this outbreak, the exact cause remains unclear with other contributing factors including improper thawing, cooking, and/or storing processes. - In 1990, an outbreak classic salmonellosis symptoms occurred during the holiday season in a Connecticut hospital. Consumption of improperly thawed and cooked turkey used for turkey salad, turkey sandwiches, and chef‘s salad served to patients was suspected as the cause. In addition to poultry, other sources of salmonellosis include eggs, produce (e.g. tomatoes, leafy greens), other meats (e.g. ground beef, beef jerky), milk, as well as direct contact with animals and their environment. A recent rnultistate outbreak of salmonellosis in the US. emerged in a new food source, peanut butter, in 2007. The causative organism, Salmonella serotype Tennessee, was reported in 47 states since August 2006. As of May 2007, a total of 628 persons were infected after consuming peanut butter that was traced to one manufacturing facility (CDC 2007). Salmonella serotypes Typhimurium and Enteritidis have both declined substantially (28% and 34%, respectively) since 1995. The total number of Salmonella isolates has also declined during this period. though not as substantially as serotypes Typhimurium and Enteritidis (CDC 2005a). Food .S'c'i/ety guidelines and regulations The National Advisory Committee on Microbiological Criteria for Foods (NACMCF) adopted a document entitled, “HACCP Principles for Food Production,” in 1989 and defined HACCP as “a systematic approach to be used in food production as a means to assure food safety.” Therefore, to produce safe food, a management system under HACCP program has to cover the analysis and control of biological, chemical, and physical hazards from raw material production, procurement and handling, to manufacturing, distribution and consumption of the finished product (Scott and Stevenson 2006). A final Food Safety and Inspection Service (FSIS) rule for sanitation perfomrance standards was established in 1999, which is applicable to all meat and poultry plants. Establishments that follow this guide can be certain that they are meeting the sanitation performance standards. However. it is not a requirement to perform the specific sanitary practices described in this document (USDA-FSIS I999). In 1999, FSIS required that certain ready-to-eat (RTE) products (cooked/roast beef products, fully cooked, uncured meat patties. and certain fully cooked poultry products) meet three performance standards: lethality, stabilization, and handling. For the lethality standard, the F SIS did not require that any specific technique or process be used to meet the standard; however, for cooked products, the FSIS did propose to require a heat treatment (USDA-FSIS 1999). In 2005, the FSIS documented time and temperature tables used for cooking ready—to-eat poultry products that the industry may follow (USDA-FSIS 2005). Pathogen reduction performance standards are based on the prevalence of Salmonella found by the agency's nationwide microbiological baseline studies. FSIS selected Salmonella as the target organism because, firstly, it is one of the most common causes of foodborne illness associated with meat and poultry products. Secondly, Salmonella lives in the intestinal tract of many animals destined for human consumption. Thirdly, current methods can recover Salmonella from various meat and poultry products, and finally, reducing the presence of Salmonella will also eliminate or adequately reduce other vegetative pathogenic microorganisms in the product. In August 2006, the F SIS published the first “Compliance Guideline for Controlling Salmonella in Poultry”. This guideline, which targets small and very small poultry plants, describes concerns and validates controls for each step in the broiler slaughter process (USDA-FSIS 2006). Bacterial penetration It is often assumed that the interior ofintact, undamaged whole muscle is free from microorganisms. Sources ofcontamination ofmeat are through evisccration, slaughtering instruments, and harvest environment, animal hide and intestines, water, and subsequent handling (Elmossalami and Wassef 1971). Several studies have demonstrated bacterial migration into the inner portion of whole muscle during processing. Penetration of microorganisms into muscle depends on various factors such as humidity, temperature, histological structure of meat, and characteristics ofthe bacteria. Gill and Penney (1977, 1982) concluded that significant migration was due to proteolytic activity and mobility of organisms and incubation temperature with the rate of penetration increasing with incubation temperature. S. Typhimurium, a motile and proteolytic pathogen, was found to travel further into beef muscle at 37 °C as opposed to 20 °C. S. Enteritidis also migrated greater distances below the surface as time increased (Elmossalami and Wassef 1971). Work done by Thomas and others (I 987) supported these results and they also indicated that water availability in the extracellular space contributed to an increase in penetration rate. Bacterial migration into the interior portion of meat is related to the water holding capacity of meat proteins. Moreover, meat that has undergone freeze-thaw cycles tends to be more srrsceptible to bacterial penetration (Sikes and Maxcy 1980). Fiber orientation is another factor that plays an important role in the depth of penetration (Sikes and Maxcy 1980). The penetration is more extensive when the inoculum was applied vertically to the meat grain as opposed to horizontal apphcafion. Ifa mechanical means is used to tenderize and/or infuse flavor into meat products, a potential risk may be posed through this physical action by carrying contaminants from the surface into the interior. The transferred bacteria may proliferate and increase in number inside the tenderized meat products. Phcbus and others (1999) showed that up to 16 4% of an E. eoli 0157:H7 inoculum was carried by the blades into the interior of surface- inoculated beef sirloins during tenderization. Interior portions of blade tenderized beef roasts have shown measurable aerobic plate counts with greater contamination on the exterior portions (Raccach and Henrickson 1979). Aerobic and anaerobic plate counts demonstrated the translocation of bacteria after four passes through a needle tenderizer (Boyd and others 1978). However, without blade tenderization, a significant microorganism penetration was also observed. Warsow and others (2007) evaluated the potential for Salmonella migration into intact whole muscle turkey breasts during still marination. Salmonella counts in samples increased with application of vacuum and decreased with a below the inoculated surface increased. The hydrodyne process (patent number 5,273,766 and 5,328,403) is an emerging, non-thermal process for improving meat tenderness with this process reducing shear force as much as 72% (Solomon and others 1997). While hydrostatic pressure (100-1000 MPa) from high pressure processing (HPP) is used as a non-thermal pasteurization approach (NACMCF 2004), the hydrodyne process is a hydrodynamic pressure processing that uses only a small amount of a high energy explosive to generate supersonic shock waves upon detonation in a liquid medium. The mechanical force of the shock waves that produce tenderization may also cause mechanical stress on bacteria. However, only a slight reduction in the E. eoli 0157:H7 population was found in ground beef when this process was applied (Podolak and others 2006). Moreover, the tenderization achieved by the hydrodyne process allowed E. eoli to move from the outer inoculated surface into the beef muscle due to disruption of muscle fibers by the wave 17 force (Lorca and others 2002, 2003). This technique is now commercially used to eliminate pathogens from processed RTE meat products (Hayrnan and others 2004). Bacterial attachment The tissues under the hide ofhealthy animals are essentially sterile prior to slaughter (Elmonssalami and Wassef I971; Niven 1987); however, contamination often occurs during processing. During slaughter, surface tissue can be exposed to microorganisms, including pathogens, and become contaminated. Cut surfaces are susceptible to bacterial penetration because they have no membranous surface to hold its cells together and protect the interior (Anderson and others 1992). There are several processing procedures that facilitate bacterial migration. The attachment of bacteria to meat after migration or contamination is a very complex process (Notermans and Kampelmacher 1983). Bacterial structures involved in attachment include extracellular polymers, fimbriae (pilli) and flagellae. In addition to bacterial structures, attachment also depends on bacterial strain, bacterial concentration, type of meat surface or microtopography of the surface, and temperature (Firstenberg-Eden 1981). Gill and Penney (1977) indicated that, after surface inoculation, Salmonella likely penetrated into tissue between the muscle fibers and later confirmed that organisms are usually located between the muscle fibers and the surrounding endomysiurn (collagen fiber layers) (Gill and Penney 1982). In addition, some bacteria were also observed within the endomysium. Lorca and others (2002) located bacterial cells in muscle after hydrodynamic treatment by using a green fluorescent protein (GFP)-labcled E. eoli strain coupled with laser scanning confocal microscopy (LSCM). These GFP-labclcd cells were seen in crevices between muscle bundle fibers. Their work is consistent with the work of Prachaiyo and McLandsborough (2000), who indicated that, after inoculated meat surface for 5 min and washed 3 times with phosphate-buffered saline, GFP-labelcd E. coli associated within the surface or sarcolemma of individual muscle fibers or within crevices of raw beef muscle (in the spaces between and on the surface of muscle fibers). Bacterial thermal inactivation Thermal inactivation ofa microorganism has been previously expressed in terms ofD and 2 values. These values are calculated assuming microbial inactivation follows first order kinetics. D value or decimal reduction time is defined as the time required to destroy 90% ofthe population, which indicates the thermal stability/resistance ofa specific organism in a specific medium or food at a constant temperature. A 2 value is the temperature increase required to reduce the thermal death time by 90% (Doyle and others 2001). In general, D values for Salmonella at 60°C range from 5-6 min in chicken, 5-13 min in turkey, and 3-5 min in beef. In most studies, the 2 values for Salmonella ranged from 5.0-6.5°C (Orta-Ramirez and Smith 2002). Previously, the USDA established time/temperature protocols with specific endpoint temperatures for commercial thermal processes. In 1999, USDA-FSIS finalized the regulations for the production ofcertain meat and poultry products. The USDA—FSIS lethality performance standards for RTE products are based on Salmonella. Although E. coli 0157:H7 is ofgreater concern. especially due to the higher number of outbreaks associated with consumption ofground meat products. it is less heat resistant than Salmonella. In contrast, Listeria monoer'togenes is typically more heat resistant than Salmonella; however, contamination occurs during post-processing steps. The lethality performance standards for thermal processing are expressed in terms of specific decimal reductions or logm reductions (USDA-FSIS 1999). The USDA requires a process to achieve a 6.5-logm reduction in RTE cooked beef, roasted beef, and cooked corned beef and a 7-logm reduction in RTE poultry products (USDA-F SIS 1999). A cocktail of Salmonella serotypes consisting of pathogenic strains that exhibit relatively high heat resistance and have been previously implicated in foodbome outbreaks is recommended in pathogen thermal inactivation studies used to validate a thermal processes (USDA-FSIS 1999). Various factors affecting bacterial thermal resistance have been documented, including meat species, product composition (e.g., carbohydrates, fat content), water activity, pH, salt and other additives, pathogen species and strains, growth temperature, stage of bacterial growth, initial population load, heat shock, and methodology used for detection of survivors. Generally, thermal resistance of bacteria is higher in meat products than in other model systems, such as buffer solutions, peptone, and agar (Juneja and others 1995, 2001). Murphy and others (2000) reported increased thermal resistance of Salmonella in food products compared to laboratory media. Physical arrangement of various components within the food matrix might also cause differences in bacterial thermal resistance. When Orta-Ramirez and others (2005) and Velasquez and others (2005) studied Salmonella heat resistance, beef and pork (whole or ground), respectively, a significantly greater heat resistance of Salmonella was seen when present in the whole muscle, as opposed to ground muscle products. 20 Doyle and Mazzotta (2000) suggested that bacteria location in a food (surface attachment vs. interior dispersion) may also affect the resistance of Salmonella. Bacteria attached to muscle tissue are more heat resistant than bacteria suspended in liquid media (Murphy and others 2000). When a cocktail of Sahnonella was inoculated and heated in meat, including beef, pork, turkey, and chicken, D values were significantly higher than those in chicken broth. According to Juneja and others (2001), this effect may be attributed to poor heat transfer through the heating menstrum or other factors present in the process before cells were subjected to heat injury. Murphy and others (2000) compared the D-values in meat to those in a semi-liquid medium and found that the D- values were higher in ground chicken breast than in a 0.1% peptone—agar solution using temperatures ranging from 55 to 70 0C in a controlled temperature water bath. They also demonstrated that the sample size and shape affected themral inactivation of Salmonella in ground chicken breast meat. Higher levels of fat (Juneja and others 1997, Ahmed and others 1995, Veeramuthu and others 1998, Smith and others 2001), as well as the addition of additives such as salts, lactates, and phosphates (Kotrola and Conner I997; Maurer 2001), may enhance thermal resistance ofpathogens. In a study by Juneja and others (2000), asymptotic D values (D values for large times) were determined in ground beef containing different fat levels. Increased heat resistance of an 8-strain Salmonella was positively correlated to the fat content. According to Ahmed and others (1995), these higher D values were due to the reduction of water activity because bacteria suspended in fat are more difficult to destroy than those in aqueous media. The effect of water activity itself on thermal inactivation of Salmonella in ground turkey was also determined. As water activity in the meat was reduced, thermal resistance of Salmonella significantly increased (Carlson and others 2005). Salt is a typical ingredient frequently added to meat products. Many studies have shown that high concentrations of salt increase thermal resistance of Salmonella. When heated at less than 63.5 °C. the lethality rate for Salmonella increased with increasing concentrations ofsalt and pyrophosphatc. However, heat resistance of Salmonella was unaffected by sodium lactate (Juneja and other 2003). The effect of sublethal stress (heat- , cold-, and starvation-injured) on Salmonella thermal inactivation kinetics was evaluated in ground turkey (Wesche and others 2005). The results indicated that heat-shocked Salmonella (54 °C for 30 min) were more thermal resistance; however, no effects were observed when cells were subjected to cold shock or starvation stress. Microscopy in food research The microscope is a powerful tool in food science research revealing structural details beyond what the naked eye can see. Microscopic examination most often employs the use of either light microscope or the electron microscope. For most routine work, the light microscope is helpful. For special research purposes, especially in studies on internal cell structure, an advanced microscopy technique such as electron microscopy (i.e. transmission electron microscopy and scanning electron microscopy), is required in addition to the light microscope (i.e. fluorescence microscopy and confocal scanning laser microscopy) (Madigan and others 2000). Transmission electron microscopy (TEM) has been used to study the ultrastructural changes in muscle foods because high resolution of 0.2 nm is achieved Ix) IQ compared to 200 nm using comientional light microscopy (Flegler and others 1993). Electromagnets function as lenses and electrons are used instead of light rays, in a high vacuum system. However, to be able to view internal cell structures, special techniques of thin sectioning are required to allow the electron beam to penetrate the sample. TEM of thin sections is appropriate for viewing inter- and intracellular structures and for measuring sarcomere length and spacing to intermyofibrillar connections (Silva and others 1992). Birkhold and Sams (1995) studied the ultrastructural changes induced by high voltage post-mortem electrical stimulation and muscle tension in broiler chickens using TEM. This technique was also used to study changes in chicken skin morphology after specific processing stages, such as scalding, picking, and chilling (Kim and others 1993). Using the same technique, Pohlman and others (1997) demonstrated microstructure differences in beefrnuscle subjected to ultrasound and conventional cooking treatments. Even though, high resolution is achieved using TEM, the technique requires extensive sample preparation. Light microscopy, in contrast, involves simple sample preparation and specimens can be thicker compared to electron microscopy. Fluorescence microscopy is most often in studying the location and movement of molecules and subcellular components in the cell. This technique is used to study specimens that can be made to fluorescc. Usually, cellular components do not flrroresce themselves. Fluorescent markers are therefore introduced so that the labeled material will emit energy detectable as visible light when irradiated with the light at a specific wavelength. The use of fluorescent dyes, immunofluorcsccnce, and tagging of proteins can be applied to allow the sample to fluoresce and be viewed under the fluorescence microscope. I‘d DJ In 1994, the use of green fluorescent protein (GFP) as a cell and tissue marker was developed (Chalfie and other 1994). GPF is a small protein (27 kDa) found in the jellyfish Aequorea VlClOl'la. This protein is fluorescent with absorption and emission peaks at 395 nm and 509 nm, respectively. Therefore, GFP-transformed organisms can be easily seen under the conventional fluorescence microscope and confocal filter sets (Cubitt and others 1995). High resolution optical techniques can be used non-invasively to monitor dynamic activities, three-dimensional arrangement, and behavior of living cells (Paddock 1999). GF P has been expressed in several bacterial pathogens including E. coli 0157:H7 (Fratamico and others 1997), Listeria monocytogenes (Freitag and others 1999), and Salmonella (C be and Kim 1999). Prachaiyo and McLandsborough (2000) observed no difference in behavior of GFP-transformed E. coli 0157:H7 and parental strains. The growth characteristics and surface properties including hydrophobicity and cell surface charge of organisms were not affected by introducing GF P into the plasmid. Noah and others (2005) also observed comparable growth characteristics of GFP- transformed Salmonella to the parental strains. In addition, the transformed strains maintained their fluorescence following a series of culture transfers and after refrigeration (4 °C) for up to 12 days. Other than the conventional fluorescence microscopy, the tagged protein can also be seen under confocal scanning laser microscopy (CSLM). CSLM has been used to study the location and viability of bacteria in food products. The location of Salmonella attached to poultry skin was detemrined using C SLM after staining with Pyronin Y (Kim and others 1996). Using the same technique, Takeuchi and other (2000) evaluated attachment ofS. Typhimurium to lettuce stained with fluorescein isothiocyanate. Wong- Liong and others (1997) stained eggshell membranes with fluorescein isothiocyanate to observe the interaction of S. Enteritidis with eggshell and the migration of Salmonella into the egg membrane. Microscopy is a valuable tool and has been extensively applied into a wide range of scientific fields. There are a variety of microscopes and techniques available. Since new techniques are being developed, the suitable tool for individual food research must be wisely chosen in order to support the finding in the chosen medium. h) (J: CHAPTER 3 EFFECTS OF MARINATION ON SALMONELLA PENETRATION AND MUSCLE STRUCTURE OF TURKEY BREAST INTRODUCTION Deep muscle tissue from healthy animals is generally sterile indicating that the interior ofintact, undamaged whole muscle should be free from microorganisms (Elmossalami and Wassef 1971). However, contamination of the meat can occur through many routes including contact with slaughtering instruments, the environment. animal hide, fecal materials, and water. Many meat and poultry products sold in retail markets are value added by means of marination and/or mechanical tenderization. Tenderizing techniques, including blade and needle tenderization can carry bacteria from the meat surface into the interior (Phebus and others 1999; Raccach and Henrickson I979; Boyd and others 1978). Warsow (2007) evaluated the potential for Salmonella penetration into vacuum tumbled and still-marinated intact whole muscle turkey breasts. Numbers of Salmonellc'i increased with vacuum processing with lower numbers seen in the center of the samples. Cut tissue (Anderson and others 1992) and meat that has undergone multiple freeze-thaw cycles (Sikes and Maxcy 1980) are more prone to bacterial penetration. Such newly introduced bacteria may proliferate in the interior ofthe muscle tissues as the nutrients and environment permit (Raccach and Henrickson 1979), thus leading to concerns regarding subsequent thermal inactivation during cooking. Marination is one ofthe Value-added processes that affords convenience and variety for consumers while increasing profits for the processor by improving product yield. To control fluid loss. salt and alkaline phosphates are commonly dissolved in the marinade. Tumbling is a process used to increase marinade uptake and promote tenderization ofmeat (Chen 1982, Xiong and Kupski 1999). In addition, a vacuum is often applied during tumbling to improve marinade distribution and speed of penetration into the product. However, the increased water content between muscle fibers has been reported to increase bacterial penetration (Thomas and others 1987; Sikes and Maxcy 1980). Various hypotheses have been proposed for the mechanisms by which bacteria attach to meat surfaces. Several microscopic and sample preparation techniques are available to observe these mechanisms. Use ofa marker protein (or tagging protein) attached to the bacteria of interest coupled with fluorescence microscopy is one useful method. Green fluorescence protein (GFP) is a fluorescent protein with absorption and emission peaks at 395 nm and 509 nm, respectively. Therefore, GFP-transformed microorganisms can be seen using fluorescence microscopy. GFP has been successfully expressed in several bacterial pathogens, including E. eoli 0157:H7, Listeria monoei'togenes, and Salmonella ( F ratamico and others 1997; Freitag and others 1999; Cho and Kim 1999). The GFP expressing strains have been confirmed to have behaviors that are not significantly different from parental strains (Prachaiyo and McLandsborough 2000). Consumer trends show an increasing demand for ready-to-cat products, which include marinated whole muscle foods (Russell 2002). One concern when the marinade is absorbed into the interior of value-added meat and poultry products is the potential migration of bacterial pathogens from the meat surface into the interior. In response to these concerns, the specific objectives for this study were to (I) evaluate the effects of marinade composition, microbial load, and contamination stage on multidirectional migration of Salmonella into intact whole muscle turkey during marination; (2) assess changes in turkey muscle microstructure after marination; and (3) determine the location of GFP-labeled Salmonella in fresh whole muscle turkey tissue after marination in an inoculated marinade. MATERIALS AND METHODS Multidirectional penetration of Salmonella into intact whole muscle turkey breast Preparation of turkey breast samples Fresh, whole muscle. boneless, skinless turkey breasts (approximately 1.5-2 kg) were obtained from a local grower and packer in a single large lot to eliminate lot-to—lot variability. The muscles were vacuum packaged individually in plastic bags (Cryovac Sealed Air Corp, Duncan, SC, USA), frozen at -20 °C and irradiated (~10 kGy, CFC Logistics, Quakertown, PA, USA) to eliminate the natural microflora. When the irradiated whole muscle turkey breasts were used in the study, they were assessed for background flora after thawing by diluting l-g samples 1:5 in 0.1% peptone water followed by plating in duplicate on PetrifilrnTM aerobic count plates (3M Corp, St. Paul, MN, USA) followed by incubation at 37 °C for 48 h. Marinade preparation Two marinade formulations were typical marinade solution containing 3.2% NaCl (EMD Chemicals Inc., Gibbstown, NJ, USA) and 0.8% phosphate solution (Butcher and Packer Supply Company, Detroit, MI, USA) calculated to target 0.4% NaCl and 0.06% phosphate in the marinated product and a high concentration marinade containing 10.2% NaCl and 4.05% phosphate calculated to target 1.25% NaCl and 0.5% phosphate in marinated product. A 520 mL aliquot of marinade was poured into glass bottles with screw caps and autoclaved for 15 min at 121 °C to ensure sterility. All autoclaved rnarinades were stored at room temperature (25 0C) until used. The pH of low (7.39) and high (7.65) marinade concentrations were measured using a Corning I45 pH meter (Corning, Medficld. MA, USA). Preparation of test organisms The following eight serovars of Salmonella were obtained from Dr. V.K. Juneja (USDA-ARS, Eastern Regional Research Center, Philadelphia, PA, USA): S. Thompson FSIS 120 (chicken isolate), S. Enteriditis H3527 and H3502 (clinical isolates phage types 13A and 4, respectively), S. Typhimurium DT 104 H3380 (human isolate), S. Hadar MF60404 (turkey isolate), S. Copenhagen 8457 (pork isolate), S. Montevideo FSIS 051 (beefisolate), and S. Heidelberg F5038BG1 (human isolate). These serovars have shown moderate to high thermal resistance and been implicated in outbreaks (Juneja and others 2001). All serovars were stored at -80 °C in tryptic soy broth (TSB) (Difco, Becton Dickinson, Sparks, MD, USA) containing 20% glycerol. The cultures were propagated by weekly transferring one loop of frozen culture to 9 mL of sterile TSB. The cultures were maintained by daily transfer to fresh TSB followed by incubation for 18-24 h at 37 °C, with two consecutive transfers prior to use. All strains were maintained separately and then combined on the day ofexperirnent. After centrifugation at 6000 x g for 20 min at 4 °C to obtain an 8-servovar cocktail, the supematant was decanted and the cell pellet was suspended in 500 mL of sterile marinade . . . . . x 4 ... to give a final bacterial concentration of approximately 10 or 10 CFUlmL. 3 0 Exposure to inoculated marinade Frozen turkey breast was thawed at 4 °C for 2 days prior to the experiment. Whole turkey breasts were transferred to a sterile stornacher bag (Fisher Scientific, Pittsburgh, PA, USA), and inoculated marinade was added at 14% w/w. The bag was tied, placed in another bag to ensure sealing, and tied again before placing into a laboratory-scale vacuum tumbler. After vacuum (100 kPa) was drawn, the tumbler vessel was rotated for 20 min at 4 °C. Alternatively to simulate contamination during processing, 20 mL of inoculated marinade was added drop wise onto a sterile tray, after which the turkey was placed on the contaminated surface and immediately transferred to a sterile bag. After adding sterile marinade at 14% w/w, the bag was tied, double bagged, and vacuum tumbled for 20 min. To simulate contamination after marination, breast muscle was tumbled with sterile marinade (added at 14% w/w), removed from double bags after tumbling, and placed with the cut surface and the opposite side down on a tray contaminated with 20 mL of inoculated marinade to simulate a contaminated food contact surface. Sampling and recovery Three 1 cm-thick slices, defined as cranial, middle, and caudal, were removed from turkey breast muscle (Figure 3.1-a). The middle slice was taken at the thickest part of the breast. Cranial and caudal slices were sectioned at approximately 5 and 10 cm, respectively, measured from both ends. Then l-cm cubes were excised from those slices as illustrated in Figure 3.1b. Horizontal cubes were labeled L4, L3, L2, L1, R1, R2, R3, and R4, whereas vertical cubes were labeled T1, T2, BI, and B2 The sampling plan 31 design was used to generate a quantitative 3—D Salmonella migration pattern. Salmonella was recovered by serially diluting the cubes in 0.1% peptone water, plating in duplicate on PetrifilmTM aerobic plates and incubating at 37 °C for 48 h before counting. The minimum detection level was 10 CFU/g due to the actual bacterial count on 1:10 dilution plating. A self—sterilizing electrosurgical unit (ESU) (SurgistatTM 11-20, ValleylabTM, Boulder, CO, USA) with a setting at 120 W coagulation and 40 W in a pure cut mode was used for al dissections as reported by Warsow and others (2003). An advantage of this innovative ESU protocol over traditional excision methods, confirmed by a preliminary study, is that bacterial transfer from the outer to inner surfaces of the sample is eliminated by the tissue cauterizing blade. However, the heat generated by the ESU may inactivate some surface and near-surface organisms, resulting in a slight underestimation of microbial penetration during marination. Thus, bacterial counts from two excision methods (ESU vs. traditional) were compared to ensure that heat generated by ESU did not destroy the target pathogen. Warsow (2003) demonstrated that although the ESU maintained a sterile cutting surface, the instrument did not supply sufficient heat to significantly affect bacterial counts within the sample. 32 V ’p‘ Cranial ¢ / Middle . Vertical profile 1 T2 Tl /L4 L3 L2 L1 c R1 R2R3 R <——— Honzoma' profile / :2 1 (b) Figure 3.1 Sampling protocol (a) 3 slices, cranial, middle, and caudal, excised from marinated whole muscle turkey breast (b) l-cm cubes dissected from the slices. Segments labeled R1, R3, Ll , L3, T1, and B1 were dissected for cranial and caudal slices because cubes T2, B2, R4, L4 did not exist since these pieces are smaller and segments labeled C, R2, R4, L2, L4, T2, and 82 were dissected for middle slice. 33 Data analysis Each treatment, from microbial load, marinade concentration, and contamination stage treatments, was replicated three times. The Salmonella migration pattem was determined by plotting the logarithm ofthe surviving cells vs. cube identification. Analysis of Variance (ANOVA) was run to evaluate the effect oftreatmcnt factors and sample location on the migration of Salmonella. Tukey-Kramer‘s test (or=0.05) was performed to compare means. Statistical analyses were performed using JMP statistical package (Version 3.2.2, SAS Institute, Inc., Cary, NC, USA). Microstructure of marinated whole muscle turkey breast Preparation of turkey breast samples Irradiated frozen turkey breasts (boneless and skinless, approximately 1.5-2 kg) were thawed 48h at 4 °C. A Wamer-Bratzler hand-coring device (1.27-cm diameter, G.R. Electrical Mfg. Co., Manhattan, KS, USA) was used to aseptically remove cores from the top of the turkey breast (2-3 cores/treatment) (Warsow and others 2008). Cores were submerged into the two marinades (3.2% NaCl + O.( % phosphate solution and 10.2% NaCl + 4.05% phosphate) for 20 min at 4 °C with the remaining cores serving as untreated controls, placed in Petri dishes, and sectioned, using a sterile surgical blade into smaller pieces measuring approximately 2 x 4 x 2 mm (W x L x H). The treated samples were then placed into small vials (5—6 pieces/treatment) in preparation for microscopic examination. Sample preparation for bright-field microscopy (BF M) Fixation: Treated samples were fixed in 0.1 M cacodylate buffer (pH 7.4) containing 2.5% (v/v) glutaraldehyde and 2.0% (v/v) paraformaldehyde and stored at 4 °C for 5 (1. Samples were then rinsed in cacodylate buffer (0.1 M, pH 7.4) and post-fixed in 0.1 M cacodylate buffer containing 2% (v/v) osmium tetroxide to preserve the lipids and maintain the three-dimensional structure. After 24 h of dehydration, the sections were subjected to three 15-min rinses in cacodylate buffer and dehydrated using 30 to 100% acetone. Infiltration and embedding: Samples were infiltrated with Poly/Bed 812 resin and acetone at ratios of 1:3, 2:2, and 3:1, followed by the pure resin. Samples were rotated for each step at least for 2 h and up to l d. To provide support during ultramicrotomy and retain spatial organization of the specimen sections, samples were placed in a mold, embedded in Poly/Bed 812 resin, and polymerized in an oven at 60 °C for 24 h. Thick sectioning and staining: Thick sections were used to determine how the tissue samples were affected by different salt and phosphate concentrations in the marinade solutions. After trimming blocks surfaces with a razor blade to expose the tissue, the embedded tissue was sectioned with a MTX ultramicrotome (Boeckeler Instruments, Tucson, AZ, USA) to 500-1000 nm using a glass knife. These thick sections (5-6 pieces) were mounted on 35 glass slides and stained with Epoxy Tissue Stain (ready-to-use Toluidinc Blue and Basic Fuchsin, Electron Microscopy Science, Hatfield, PA, USA) on a hot plate for approximately 20-30 s. Observation of the stained sections was done under a confocal laser scanning microscope (Zeiss LSM5 Pascal, Thornwood, NY, USA) using a 633 nm laser to obtain a transmitted image. Image analysis: Digital images of muscle structure obtained from bright-field microscopy were processed using Matlab Image Processing Toolbox. The threshold value was calculated by using the “Otsu method", a function available in the Toolbox. Subsequently this threshold was used to determine “white” and “black” pixels, i.e., when the values are below or above the threshold, respectively. “White” space was converted into percentage of extracellular muscle tissue space by dividing the number of“white” pixels by the total number ofpixels in the image. The extracellular spaces of treated samples (TC and HC) were then compared to the untreated samples (control). Five images per treatment were taken digitally. Significance was determined at the 95% confidence level for all mean comparisons. In addition. the muscle cell diameters from each treatment were averaged and compared to that ofthe control sample. Only muscle cells that were shown in full structure were determined for diameters using a ruler. Sample preparation for transmission electron microscopy (TEM) TEM sample preparation involved the following two extra steps from that of BFM. Thin sectioning: After re-trimrning the blocks to reduce the sectioning area and improve thin section quality due to compression and wrinkle, sections were obtained with an MTX ultramicrotome using a diamond knife (90-1 10 nm thickness, 0.7 mm/sec cutting speed, DDK, Delaware Diamond Knives, USA) and collected on 300-mesh copper grids. Positive staining: Uranyl acetate (2% (w/v) in 50% ethanol) and lead citrate (Reynolds formulation) were used as positive stains. Grids were placed section side down on a drop of uranyl acetate (1 drop/grid) placed on a piece of Parafilm and covered with a Petri dish. Grids were incubated for 10 min, rinsed with distilled water, and allowed to air dry. On another piece of Parafilm, sodium hydroxide pellets were placed inside a Petri dish to create a COg-free environment. Grids were placed section side down on the lead citrate (l drop/grid) and incubated for 15 min before being rinsed with distilled water, dried, and viewed under a transmission electron microscope (JEOL 100C X, Japan) at an accelerating voltage of 100 kV. Salmonella location in fresh whole muscle turkey breast tissue Preparation of fresh whole muscle turkey breast tissue Fresh, commercially processed, whole muscle turkey breast (skinless and boneless) was obtained from a local grower and packer and used immediately. To visualize a cross section, turkey breast tissue was excised to measure approximately 1 cm x 1 cm x 0.3 mm (W x L x H) using a scalpel. 37 Preparation of green fluorescent protein (GFP)-Salmonella inoculum GFP- labeled Salmonella (S. Typhimurium DT104 ATCC 700408/ISSAGFP) was obtained from Dr. .lolm Sofos, Department of Animal Science, Colorado State University (Fort Collins, CO, USA) and stored frozen at -80 °C in TSB containing 10% (vol/vol) glycerol. The culture was propagated by transferring one loop of frozen culture to 9 mL of sterile TSB supplemented with yeast extract (TSB-YE). The cultures were maintained by daily transfer (100 rrL) to fresh TSB-YE followed by incubation for 18-24 h at 37 °C, with a minimum oftwo consecutive transfers prior to use. Preparation of GFP-Salmonella inoculated marinade After centrifuging the TSB-YE culture at 6000 .r g for 20 min at 4 °C, the cell pellet was suspended in 10 mL of marinade containing 3.2% NaCl and 0.8% phosphate. Subsequently, turkey samples were spot-inoculated on the upper side with 5 1.1L of marinade and incubated for I h at 4 C’C to allow migration of Salmonella. Cryostat sampling of inoculated sample block Inoculated meat was placed in a mold at room temperature. An embedding medium (Tissue-Tekim OCT compound, Sakura Finctek USA Inc., Torrance, CA, USA) for frozen tissue specimens was used to support the tissue in the specimen block. After immersing the mold into liquid nitrogen for 5 min, the frozen tissue samples were transferred to a -20 °C cryostat chamber (Leica CM3050, Leica lnstrurnents GmbH, Wetzlar, Germany) and allowed to equilibrate for at least 2 h. The sample was 38 temperature equilibrated to that of the cryostat chamber. The frozen tissue was then mounted onto the specimen disc using OCT compound before being returned to the cryostat chamber. After the OCT hardened (approximately 20 min), the specimen disc was attached to the chuck holder and sectioned from the uninoculated side toward the inoculated surface to a thickness of 8 pm using disposable Teflon® coated microtorne blades (DuraEdge Low Profile Blades, Source Medical Products, Lake Forest, IL, USA). The sections were transferred by contacting a dry microscope slide and mounted with PermaFluor mounting medium (Thermo Sandon, Pittsburgh, PA, USA) and a cover slip. The location of bacterial cells in turkey breast tissue was determined by fluorescent microscopy (Leica DMLB, Leica, Wetzlar, Germany) at 400X and 1000X magnification. 39 RESULTS AND DISCUSSION Multidireetional penetration of Salmonella into intact whole muscle turkey breast The overall results of this study demonstrated bacterial penetration into whole muscle turkey breast (Figures 3.2 to 3.8). Earlier research with bacteria penetration was limited to individual serovar’s proteolytic properties (Gill and Penney 1977, 1982; Gupta and others 1983; Sike and Maxcy 1980). Proteolytic enzymes indigenous to the bacteria were hypothesized to break down the connective tissue (endomysium) between muscle fibers and consequently facilitate penetration into tissue. However, protease production secreted by the bacteria does not occur until the meat starts to deteriorate (Gill and Penney 1977). Therefore, this circumstance does not apply because uncontaminated meat was used and no time was allowed for proteolytic enzymes to be produced and secreted by Salmonella serovars. Salmonella distribution as a result of migration was significantly (P<0.0001) impacted by microbial load and excised cube location (or cube ID) (Figures 3.2 and 3.3). However, no interactions (P=0.7786) were seen between microbial load and cube location. With the higher microbial load (108 CFU/mL as opposed to 104 CFU/mL), an increase in bacterial number was observed within the turkey breast. Figures 3.2 a, b, and c illustrate the penetration pattern in the cranial, middle, and caudal slices, respectively. The highest Salmonella counts were generally observed in the cubes L4 and R4, immediately below the inoculated outside surface (Figures 3.2 b and 3.3). Cube C at center yielded fewer salmonellae (P3005) compared to cubes R4 and L4. As clearly seen in the middle slices (Figures 3.2 b) Saln‘zonel/a penetrated toward the center of the turkey 40 breast. The cranial and caudal slices showed no significant movement of Salmonella towards the middle section with numbers of Salmonella seen horizontally (Figures 3.2 a and c). This may be due to the structure, thickness variation, and orientation ofthe turkey breast at both ends. Populations of Salmonella, Figure 3.4, recovered from cubes T2 and B2 next to the inoculated surface were significantly (P3005) greater compared to cube C at the center. Both inoculum levels showed a decrease in Salmonella from the outside surface toward the center. 41 Figure 3.2 Horizontal profile of Salmonella in (a) cranial slice (b) middle slice, and (c) caudal slices of whole muscle turkey breast, subjected to 104 CFU/mL (open square) and 108 CFU/mL (solid circle) inoculum levels. Segment 1D corresponds to the sampling diagram as illustrated in Figure 3.1 in materials and methods section. Each data point is an average of three replications with standard deviation. * and I” indicate significant difference (P3005) between and within treatment levels, respectively. r * r d a m 4 4 R Q m. R .x. r *3 .I T d a a a 3 3 Tel. Imim llll. ..1m:11_ R * R - c. l i a b 2 m 2 R 7.; TImrIIl R * a 1 a 1 1 T... TEL. R T01 ImllR * R m c.. - m - m b b m c C m 1.1 n”. C m 9 9 f r r e r e a a a a S S . pp] L 1 1' TI.I+ ImIIlU i “ 44E L L r :0» b r r a a 2 .2.. . to m l u L *9 I d r I a a a 3 _ b j _ 3 . T a :L i. 11.... L ‘L Tml L *, r r a a 4 m. .1 a 14. L T a _ _ _ _ _ _ _ 1 q _ _ _ _ _ _ 7 a a a a _ _ . _ 920 8.. 998 m3 925 8.. (a) (b) (C) Segment ID 43 Log CFU/g Segment 1D Figure 3.3 Salmonella penetration diagram in middle slice of turkey breast showing the highest population in the cubes immediately below inoculated surface. ~ g b.* a Q a.b * LogCFU/g O—bNQJbUlmwm l—Q—l T2 T1 C 81 82 Segment ID Figure 3.4 Vertical profile of Salmonella in the middle slice of whole muscle turkey breast, subjected to 104 CFU/mL (open square) and 108 CFU/mL (solid circle) inoculum levels. Segment ID corresponds to the sampling diagram as illustrated in Figure 3.1 in materials and methods section. Each data point is an average ofthree replications with standard deviation. * and ”'b indicate significant difference (P3005) between and within treatment levels, respectively. Inoculum level previously contributed to microbial penetration (Kim and Doores 1993). In this study, penetration was highest for marinade inoculated at 108 CFU/mL. This high bacterial load was needed to quantify the extent of penetration (Butler and others 1979). Muscle fiber orientation also impacts the depth of penetration with Gill and Penney (1982) showing that penetration is more extensive when the inoculum is applied vertical to the meat fibers as opposed to horizontal. In addition, bacteria not only move downward from the point of inoculation (as affected by gravity) but also ascend from the point ofinoculation (Sikes and Maxcy 1980). Although, fiber orientation was not considered in this study, a penetration gradient in multiple directions was seen toward the middle portion ofturkey breast. In addition to inocrrlurn level and fiber orientation, bacterial motility can also impact penetration with organisms moving between the endomysial sheath and associated muscle fibers (Thomas and others 1987). Flagellated bacteria more readily attach to surfaces compared to nonmotile bacteria (Butler and others 1979; Gill and Penney 1982). While Salmonella is motile, this motility is not the only factor facilitating the migration of the organism into the meat. Incubation temperature is also important when determining extent of penetration (Gill and Penney 1977, 1982). As the temperature increases, the distance bacteria penetrate increases. Gill and Penney found that motile proteolytic strain of S. Typhimurium migrated further into beef muscle at 37 °C as opposed to 20 oC. However, in order to mimic industry conditions, all experiments were performed at 4 °C. Multidireetional penetration of Salmonella serovars into whole muscle turkey / breast, subjected to typical (TC: 3.2% salt and 0.8% phosphate) and high (HC: 102% salt 46 and 4.05% phosphate) marinades was also examined as shown in Figures 3.5 and 3.6. Marinade concentrations did not significantly (P=0096) affect Salmonella migration into whole muscle. while cube location showed a significant (P<0.0001) effect. The interaction between these two factors was not significant (P208189). Fewer salmonellae were observed near the center cube within middle slices for both marinade concentrations (Figure 3.5 b). There was no difference in Salmonella recovery among cubes from cranial and caudal slices (Figures 3.5 a and c). Figure 3.6 illustrates a vertical sectioning profile of Salmonella in the middle slice of the turkey breast. Significantly (P3005) higher numbers of salmonellae were recovered in the segments immediately below the surface (T2 and B2) compared to the C cube. 47 Figure 3.5 Horizontal profile of Salmonella in (a) cranial slice (b) middle slice, and (c) caudal slices of whole muscle turkey breast, subjected to marinade solutions containing 3.2% salt and 0.8% phosphate (open square) and 10.2% salt and 4.05% phosphate (solid circle). Segment ID corresponds to the sampling diagram as illustrated in Figure 3.1. Each data point is an average of three replications with standard deviation. a‘c indicate significant difference (P3005) within treatment levels. There were no significant (P>0.05) difference between treatment levels. 48 and IaIeI. 1 .l 32.. ImLTI .I and r l m 0... 33 i111 TIH_—__qd 876543210 9988.. ) a ( R2 R3 R4 R1 L3 L2 L1 L4 Segment ID bb a,c ab and i. 7d__._a1 876543210 99.0 8.. (1)) R2 R3 R4 R1 L3 L2 L1 L4 Segment ID (I (I 21 8 _.a 987 (C) I- — q a — 4 654321 296 8.. 0 R2 R3 R4 R1 L3 L2 L1 L4 Segment ID 49 8 7 6 O) S 5 ~ u. o 4 - b or 3 3- b 2 .. 1 .. O I I l 7 1 T2 T1 C B1 82 Segment ID Figure 3.6 Vertical profile of Salmonella in the middle slice of whole muscle turkey breast, subjected to marinade solutions containing 3.2% salt and 0.8% phosphate (open square) and 10.2% salt and 4.05% phosphate (solid circle). Segment ID corresponds to the sampling diagram as illustrated in Figure 3.1. Each data point is an average of three replications with standard deviation. a'b indicate significant difference (P3005) within treatment levels. There were no significant (P>0.05) difference between treatment levels. 50 Non-meat ingredients such as salt and polyphosphate are commonly used to enhance muscle tenderness (Lyon and Hamm 1986) and are usually incorporated into marinades during tenderization. Tumbling turkey breast muscle with these non-meat ingredients reduces fluid water loss during cooking and improves meat quality (F roning and Sackett 1985, Maki and Froning I987). Alkaline pyrophosphates act as a fluidizing agent in muscle, dissociating actin and myosin, changing ionic strength and pH, which leads to increased water uptake (Xiong and Kupski 1999). High water content was shown to increase the rate of bacterial penetration, presumably because water uptake by muscle increases the inter-fiber distance of muscle and thereby decreasing bacterial resistance to movement within the tissue. In addition, increased inter-fiber gaps from high water content in muscle may contribute to increase microorganism penetration rate (Thomas and others 1987). However, Sikes and Maxcy (1980) have found that the addition of sodium tripolyphosphate to comminuted pork reduced the penetration depth. In our study, no significant increase or decrease in Salmonella penetration was seen using HC as opposed to TC marinade. This may be because bacterial migration occurred before water was absorbed by the individual muscle cells. The points of Salmonella exposure; before, during, and after marination, were also assessed. The point at which contamination occurred significantly (P<0.0001) affected the extent of Salmonella penetration (Figures 3.7 and 3.8). There were no interactions (P=04939) observed. Migration of Salmonella was observed in samples subjected to these treatments. For all treatments examined (Figures 3.7 b and 3.8), Salmonella counts in the middle slice significantly decreased (P3005) as the depth below the inoculated surface increased. However, the cranial and caudal slices (Figures 3.7 a 51 and c) did not show any specific penetration pattern. Greater recovery was observed when the turkey was vacuum tumbled in contaminated marinade. Less Salmonella migration into marinated turkey breast was observed in pre- and post- marinade treatments. This may be due to the lower Salmonella levels, as opposed to during marination when the pathogen was suspended in the marinade. The C segment from the middle slice had the lowest bacterial count among all cubes. 52 Figure 3.7 Horizontal profile of Salmonella in (a) cranial slice (b) middle slice, and (c) caudal slices of whole muscle turkey breast, subjected to contamination at pre- (open square), during (solid circle), and post- (cross) vacuum tumbling process. Segment ID corresponds to the sampling diagram as illustrated in Figure 3.1. Each data point is an average ofthree replications with standard deviation. A43 and ““1 indicate significant difference (P3005) between and within treatment factors, respectively. LII [ca-J (a) (b) (C) LogCFU/g O—le-bU'ICDVCD Log CFUIg O—INOJAUIO'JNCD LogCFU/g o-smeobororxioo a,A a.A a,A a,A a,A a,A a A a.A a,A . ’ a.A a.A d,A L4 L3 L2 L1 C R1 R2 R3 R4 Segment 10 E a,d,A g d,A 1'11 a,A a A i a,b,d,B é a,A Z a,A a,c,A ' a,A,B b,c,A b.A b.A ‘ L4 L3 L2 L1 C R1 R2 R3 R4 Segment ID a,A { a,A a A a,A a,A d’A a,A a.A a,A,B 3~A d‘A ’ a,B L4 L3 L2 L1 C R1 R2 R3 R4 Segment ID LogCFU/g O—wameNm o- T2 T1 C 81 82 Segment ID Figure 3.8 Vertical profile of Salmonella in the middle slice of whole muscle turkey breast, subjected to contamination at pre- (open square), during (solid circle), and post- (cross) vacuum tumbling process. Segment ID corresponds to the sampling diagram as illustrated in Figure 3.1. Each data point is an average of three replications with standard deviation. a'b indicate significant difference (P3005) within treatment factors. There was no significant difference between treatment factors. Tenderization is widely used in the meat industry to enhance both marketability and profits. Mechanical tenderization breaks down the connective tissue and fragments the myofibrillar structure leading to increased tenderness (Boyd and other 1978). These physical actions carry bacteria from the outer surface ofintact muscle into the interior portion and allow for potential proliferation (Raccach and Henrickson 1999). Significant bacterial penetration has also been observed without any invasive procedures that would enhance mechanical damage to muscle tissue. Gupta and others (1983) demonstrated that S. Typhimurium penetrated at least 3 cm into poultry muscle at atmospheric pressures after 20 h of incubation at 37°C. In addition, Warsow and others (2008) evaluated Salmonella migration into intact whole muscle turkey breasts during a still marination process. Salmonella counts significantly increased in vacuum-marinated samples (3 cm into the turkey breast) and decreased with depth below the inoculated surface. The results from Warsow‘s study are similar to the results in this study indicating that pathogens can penetrate into the interior of turkey breast without using an invasive marination process involving blade tenderization or needle injection. However, this study, which presents novel information about Salmonella penetration, assessed the effects of marinade ingredients (salt and phosphate), microbial load, and contamination stage (prc-, during. and post-marination). In addition, multiple directional penetration was studied, as opposed to a single directional penetration, using improved sampling methods. Marination is a traditional process used to tenderize and improve meat flavor (Richardson and Mead 1999). This process involves liquid with dissolved non-meat ingredients that are incorporated into the products before cooking. Many commercially produced meat products are marinated by still marination, tumbling, blending or 56 mechanical injection. Overall, this study showed that contamination ofthe marinade solution and/or the poultry surfaces resulted in bacterial penetration into the interior of the product during vacuum tumbling. Consequently, while marination improves meat product quality and yield, such products may pose greater health hazard if an internal cooking temperature of 71 °C. or other microbial interyention techniques are not used. Microstructure of marinated whole muscle turkey breast To obtain an overall view of the muscle fiber, bright-field microscopy images of muscle bundle cross sections were taken from thick sections. Figure 3.8 shows a series of transverse sections of turkey breast subjected to various treatments, including control (turkey breast tissue with no additional water), TC, and HC marinades. The images roughly indicate the difference in extracellular space between muscle fibers. In the absence of salt and phosphate (Figure 3.9 a), the extracellular spaces were wider with the muscle fibers, showing a sharp polygonal shape. After non-meat ingredient addition (Figures 3.9 b and c), muscle fiber expansion or swelling was evident along with a significant (P3005) increase in cell diameter compared to the control. However, no significant differences were seen between muscle cell diameters for samples submerged in TC and HC (P>0.05). When salt and alkaline phosphates were added to the marinade solution, the circumference of the myofibers tended to smooth out resulting in a more rounded fiber structure. In addition, a decrease in extracellular space was observed with the presence ofnon-meat ingredients. The changes in tissue organization occurred as a result of water uptake, which caused these fibers to swell and expand within the connective tissue framework. Figure 3.9 Bright-field microscopy images oftransverse sections of turkey breast whole muscles subjected to (a) control with no water, salt, and phosphate, (b) 3.2% aCl and 0.8% alkaline phosphate, and (c) 10.2% NaCl and 4.05% alkaline phosphate. Magnification 200x; bars indicate 50 pm. \I a ( (b) (C) 60 The Matlab sofiware program was used to calculate muscle tissue pore space represented by “white” pixels (Table 3.1). The extracellular space in turkey muscle samples treated with salt and phosphate, TC and HC, was less than the control (P3005). In addition, increasing 3.2% salt and 0.8% phosphate to 10.2% salt and 4.05% phosphate in marinade solution significantly (P<0.05) decreased the extracellular area of muscle cells. Table 3.1 Percentage of “white” pixels or muscle tissue pore space as determined using Matlab software program. Treatment White area (%)' Control 26.14 i 6.5a Typical marinate concentration 13.62 i 29" (3.2% salt and 0.8% phosphate) High marinade concentration 7. 57 i 0.50 (10.2% salt and 4.05% phosphate) 3” indicate significant difference (P 3 0.05) of white area percentage between treatments. * Each value is an average of 5 representative images i standard deviation. 61 Marinade solutions containing inorganic phosphates and NaCl are used to enhance the quality of poultry and other muscle foods. Such ingredients improve water holding capacity, reduce shrinkage and cooking loss, and increase product tenderness (Froning and Sackett 1985). Most previous reports have focused on the effects of salt and phosphate on water holding capacity. Several studies have shown that a combination of alkaline pyrophosphates and/or tripolyphosphatcs with NaCl improve water holding capacity in meat and poultry products as compared to NaCl alone (Shults and others 1972; Wierbicki and others 1976). However, Xiong and Kupski (1999) reported that a concentration of 8% salt tended to diminish the water-binding effects of phosphate. In this study TEM was used to observe the ultrastructural changes that occur in turkey breast tissue after marination. Figure 3.10 illustrates representative micrographs of turkey samples subjected to TC and HC marinades as well as the control (no water, salt, and phosphate added). The TEM images of longitudinal sections show less interspace between myofibrils as well as larger myofibrils (width) as the concentration of salt and phosphate increased. These changes in muscle fiber size may be due to the contribution of salt and phosphate incorporated in the marinade. In addition, as shown in Figure 3.10 c, increasing the salt and phosphate concentration to 10.2% and 4.05%, respectively, eliminated the striated pattern in muscle structure. This observation may be due to the high salt concentration in marinade solution that not only causes muscle fibers to swell and dissociate but also solubilizes and extracts the myofibrillar protein within the basic myofibril structure. Figure 3.10 Transmission electron micrographs of longitudinal sections of whole muscle turkey breast subjected to (a) control with no water, salt, and phosphate and (b) typical concentration marinade with 3.2% NaCl and 0.8% alkaline phosphate, and (c) high concentration marination with 10.2% NaCl and 4.05% alkaline phosphate. Magnification 10,000X; bars indicate 2 pm. 63 ) a ( (b) (C) 64 The water holding capacity ofmeat usually increases at pH values above the isoelectric point of myofibrillar protein. At the molecular level, alkaline phosphates break down crosslinks between myofibrillar proteins (actornyosin), increase the number of negatively charged sites on proteins available for water binding, and extend the interfibrillar spaces in which water is immobilized in the myofibril lattices. This action consequently contributes to myofibrillar expansion (Offer and Trinick 1983). Release of monophosphates from phosphate compounds is associated with additional changes in muscle pH. Li and others (2001) reported that meat samples treated with pyrophosphate and tripolyphosphate had increased levels of monophosphates and therefore increased water absorption. The muscle structure also changes in response to salt concentration. Muscle that was immersed in a 3% salt solution exhibited swollen of fibers (as the diameter increases), which shrank at salt concentrations 2 6% (Bocker and others 2005). The microstructural changes depend on existing NaCl in solution and the degree of swelling or dehydration of the myofibrils. Changes at the molecular level correspond to the protein secondary structure. The NaCl concentrations above 2% create an ionic strength sufficiently high to solubilize chicken myofibrillar proteins (Liu and Xiong 1997). Overall, this study demonstrated the effect of salt and alkaline phosphate on the microstructure of turkey breast meat by increasing the muscle fiber size, which has not been previously reported in the literature. Although, the application of salt and alkaline phosphate has been widely used in the industry to improve the quality and palatability of products, the ionic characteristics of the two ingredients are not the only factors that contribute to changes of muscle microstructure due to water absorption. Post mortem 65 status ofthe meat, type of muscle and fibers, and changes in the aqueous pH surrounding the muscle fibers to a more alkaline condition also affect the extent of swelling due to increased water binding to the muscle fibers (Offer and others 1989; Egelandsdal and others 1995). 66 Salmonella location in fresh whole muscle turkey breast tissue Wet-mount preparations of GFP-labeled Salmonella were examined using fluorescence microscopy. In order to obtain an overall view of bacterial penetration/attachment, the fluorescence images were superimposed over the transmitted images of muscle fibers at the corresponding areas. Figure 3.1 1 illustrates some representative images of 8-pm thickness transverse sections of muscle tissue. The images indicate that Salmonellc'i cells are likely to locate between turkey muscle fibers and muscle bundles and reside near endomysial and perimysial connective tissue. 67 Figure 3.1 1 Enhanced fluorescence images of GFP-labeled Salmonella location in turkey breast transverse sections taken at (a) 400X magnification; (b) enlargement of circle area in (a) taken at IOOOX magnification. Bars indicatelO rim and arrows point to GFP- Saln-zonella, which has penetrated the muscle. lrnagcs in this dissertation are presented in color. 68 (a) — 10 pm 0?) 10 11111 69 GFP-transformed Salmonella is visibly green and fluoresces when exposed to long wave UV light, which can be detected by fluorescence microscopy. One limitation is that the GFP—transformed Salmonella might have different characteristics when compared to the parental strain. However, Prachaiyo and McLandsborough (2000) observed no difference in behavior of GFP-transformed E. coli and the parental strains. The surface properties, including hydrophobicity and cell surface charge of the organisms, were not affected by introducing a GFP-containing plasmid. In addition, Noah and others (2005) indicated the ability of GFP-transformed Salmonella to maintain their fluorescence following 8-15 transfers and during refrigerated (4 °C) storage for up to 12 (1. Therefore, the GFP-transformed Salmonella in this study should behave similarly to the parental strain. The fluorescence images were also taken longitudinally to the muscle fibers (not shown). However, the specific location of Salmonella was questioned when they were found near muscle fibers. Whether they were positioned inside or on top of the fiber could not be visually determined. The location of Salmonella could consequently be easily misinterpreted. Therefore, cross sectional areas were examined. Gill and Penney (1977) indicated that salmonellae are likely to penetrate the tissue between muscle fibers. This may be due to changes in poultry muscle post-slaughter, where spaces between muscle fibers are created by radial shrinkage of the muscle fibers that allow bacteria to enter the deeper layers of the muscle tissue (Thomas and others 1987). Collagen fibers were suggested as one point of attachment for S. Typhimurium (Kim and Doores 1993; Woody and others 2000). In addition, microorganisms are also present within the 70 sarcolemma (or surface) of individual muscle fibers or within crevices of raw meat muscle (Prachaiyo and McLandsborough 2000; Lorca and others 2002). The mechanism by which bacteria attach to meat surfaces involves two stages. The primary stage is a weak adhesion due to physical forces and the second is a firmer adhesion by fimbriae and pili, involving formation of hydrogen or ionic bonds and a formation of extracellular polysaccharides (or extracellular polymers) (F irstenberg-Eden 1981; Lillard 1989). Dickson and Koohmaraie (1989) indicated that cell surface charges and hydrophobicity influenced bacterial attachment. A net negative charge on the bacterial cell wall aides in adhesion to positively charged collagen fibers. However, the magnitude of attachment varies from strain to strain based upon the net negative charge ofthe strain. Changes in the microtopography of tissue caused by various processing procedures also contribute to bacterial attachment. Water and saline alter the muscle integrity by causing individual fibers to separate (Thomas and others 1987). Slow freezing and thawing also disrupt the integrity of the sarcolemma. Slow freezing draws intrafiber water across the sarcolemma, which results in the formation of large ice crystals external to the muscle fiber (up to 100 um cross-sectional size) that can tear the membrane (Sikes and Maxcy 1980; Do and others 2004). These conditions contribute to structural changes and damage in the intercellular space and consequently serve as ports of entry for bacteria, thus allowing attachment at those crevices. Because of these concerns, fresh turkey breast muscle (post-rigor) that has never been frozen was used in this study to avoid the ice crystal damages from the freeze-thaw process. 71 The use of green fluorescence protein—transformed Salmonella coupled with the fluorescence microscope successfully demonstrated the target pathogen in supporting connective tissue and crevices within the muscle. This information has not been previously reported. CHAPTER 4 SUMMARY AND OVERALL CONCLUSIONS Marination is one of the processing methods used to increase product yield and palatability. These results suggest that Salmonella may be able to enter and survive in value-added marinated products. The degree of bacterial penetration into whole muscle turkey breast was associated with microbial load and time of exposure. Cube location within the middle slice had a significant effect on Salmonella penetration into muscle. When comparing treatment levels within each factor, inoculum level had the largest impact but vacuum tumbling also increased bacterial penetration. Marinade concentration had little effect on Salmonella penetration. This may be due to migration that occurred before the muscle cells absorbed water. The photomicrographs obtained from this study indicate that the muscle fiber expanded and decreased in extracellular space after subjecting turkey breast muscle to marinades containing salt and phosphate. Marinades diffused into the fibers and myofibril matrices. TEM was used to observe the effect of marinade composition on muscle tissue. The swelling of turkey myofibrils indicated moisture absorption during marination. These findings support that non-meat ingredients, especially salt and alkaline phosphate, contribute significantly to marinade penetration and absorption in muscle fibers. Salmonella present in inoculated marinade solution migrate into the interior of whole muscle turkey breast and locate in the extracellular space. The organisms tend to be associated with or located near endomysial and perimysial connective tissue. 73 Microbial contamination is associated not only with the marination step during meat processing but also with slaughtering equipment, processing environment, animal hide, and gastrointestinal contents. Water and subsequent handling procedures also play important roles. Meat processing that lead to changes in microtopography may increase the opportunity for pathogens to reside and attach firmly to meat and poultry products. Therefore, postmortem hygiene during processing is of extreme importance. If the process encourages or permits the movement of surface bacteria deep into the beef muscle, normal cooking time/temperatures may not eliminate microorganisms or pathogens protected within the tissue, subsequently creating a microbial safety hazard. In addition, improper storage temperatures and atmospheres may allow the introduced bacteria to survive and proliferate in these products. Once they have penetrated into meat products, these bacteria may attach to a preferential area or location and fail to be inactivated during thermal processing. Therefore, thermal processing with proper cooking time and temperature is important to assure product safety for consumers. 74 CHAPTER 5 FUTURE RESEARCH This study demonstrated that bacterial multidirectional penetration into intact whole muscle turkey breast can occur during marination and depends on the microbial load and time ofexposure. Although unlikely, penetration may also be related to marinade composition, since only salt and phosphate were studied and other non-meat ingredients may influence penetration. Recommended future research in this area includes: (1) development ofa 3-dimcnsional profile of pathogen level in sample that will be useful to pararneterize mathematical models in order to verify and improve the safety of marinated whole muscle poultry products. (2) evaluation of other treatment factors (i.e. marination time, other non-meat ingredients, partial vacuum treatment, non-meat intervention techniques such as sodium lactate, packaging and storage time after marination) and incorporation of these factors into the mathematical model developed in this dissertation that will aid in the understanding of bacterial penetration in ready-to-eat poultry products. (3) development of an extensive model by including information of other microorganisms and meat species. Vacuum tumbling is generally known to tenderize meat and distribute the marinade throughout the product. This method of marination facilitates the penetration of Salt-nonella the interior of intact whole muscle turkey breasts. Various tenderization and marination techniques are available for commercial use; therefore, evaluation ofthe effect of other tenderization techniques (i.e. still marination. needle and blade tenderization) on bacterial multidirectional penetration would be useful to generate additional migration mapping since internalized bacteria might be distributed differently as a result ofother value-added processes. Microbiological data showed that Salmonella could survive in marinade solutions containing salt and alkaline phosphate stored at 4 °C for up to 2 months. Concern arises when whole muscle products have been vacuum tumbled with contaminated marinades, as they might pose a risk due to survival pathogen in the product ifthe cooking is not sufficient. Therefore, the determination of microbial survival or multiplication in marinated whole muscle products stored in refrigerated retail cases should be further evaluated. Bright-field and transmission electron microscopy images revealed the effect of both salt and alkaline phosphates on muscle structure. Further studies are needed to determine the effect of individual non-meat ingredients at different concentrations and marination times on muscle structure changes, as this will assist food technologists in incorporating appropriate non-meat ingredient to attain desired product attributes. Results from this study also demonstrated that the location of Salmonella in extracellular areas (between muscle fibers and bundles) with the pathogen tending to be associated with supporting connective tissues. The effect of tenderization techniques on muscle structure (creating tears to the membrane and causing mechanical disruption of muscle tissue integrity and therefore creating space for potential contaminating pathogen to reside) should also be studied to assist in achieving sufficient internal temperatures to inactivate pathogens that may be embedded and protected within the interior of whole 76 muscle products. Bacterial attachment is a complex phenomenon involving several factors (such as microtopography of the attachment area and physicochemical properties of the exposed surface). These properties should also be considered to understand the mechanism of Salmonella adhesion. Thermal inactivation work was done only on small size cores, rather than on full sized turkey breasts or breast roasts. Small core samples are necessary to run a simple isothermal inactivation process and to estimate the inactivation parameters. Applying these estimated thermal inactivation parameters to industrial cooking processes for whole turkey breast might be another focus area for future projects. Results suggest that internalization of Salmimella in whole muscle turkey breast leads to enhanced thermal resistance. Additional studies are needed to elucidate the mechanisms by which bacterial pathogens are protected in whole muscle compared to ground muscle when subjected to thermal inactivation. Future research should include product physical state (whole muscle or ground muscle) when developing predictive thermal inactivation models for different cooking processes. 77 APPENDICES APPENDIX A THERMAL INACTIVATION OF SALMONELLA IN WHOLE MUSCLE AND GROUND TURKEY BREAST ABSTRACT Previous research has demonstrated the potential for Salmonella penetration into whole muscle turkey breast when processed with inoculated marinade. Moreover, when internalized, this organism may exhibit enhanced thermal resistance. In this study, the effect of turkey physical structure (ground and whole muscle) on the thermal resistance of Salmonella was evaluated. Irradiated whole and ground turkey breasts were exposed to a salt and phosphate marinade containing 8 serovars of Salmonella (~108 CFU/mL) for 20 min. Samples were subjected to isothermal heating treatments (55, 60, and 62.5 °C) in triplicate for predetermined times after which surviving cells were enumerated. Core temperatures were recorded to determine the thermal lag time (time to reach the target temperatures). The differences in Salmonella counts before and after the thermal lag time were not significantly different (P>0.05). The first-order kinetic rate constants (k) for whole muscle were approximately 50% lower than those for ground muscle, of same composition, at each temperature (P<0.05), indicating that the Salmonella inactivation rate was greater (P<0.05) in ground than in whole muscle samples. These results suggest that internalization of Salmonella in whole muscle products leads to enhanced thermal resistance. Therefore, the cooking times and temperatures currently recommended for 78 pathogen inactivation, especially for marinated value-added meat products, may need to be reevaluated. 79 INTRODUCTION Salmonella continues to be one of the most common foodborne pathogens implicated in gastroenteritis cases in humans and continues to be a major public health concern to the food industry. Salmonellosis can be fatal and the cost associated with infection can be very high (Mattick and others 2001). Many outbreaks of salmonellosis have resulted from the consumption of contaminated meat, eggs, or dairy products. Poultry and poultry products remain one of the leading sources of Salmonella infection in F humans (Tietjen and Fung 1995). Microorganisms that contaminate meat surfaces during processing may migrate into the products where they may survive and/or propagate (Raccach and Henriekson 1979). In addition, these microorganisms may exhibit enhanced thermal resistance during cooking or thermal processing (Orta-Ramirez and others 2005). Bacterial thermal resistance can be affected by many intrinsic and environmental parameters including meat species, muscle type, product composition (e.g., carbohydrates, fat content), suspending medium, water activity, pH, and salt. Generally, thermal resistance of bacteria is higher in meat products than in other model systems, such as buffer solutions, peptone, and agar (Juneja and others 1995, 2001). Vegetative bacteria are more susceptible to heat in foods with higher water activities typically because of the better heat transfer capability. Increasing the fat content in the substrate decreases moisture content and therefore, accounts for increased pathogen survivability because of poor heat penetration through the heating menstruum (Ahmed and others 1995; J uneja and Eblen 2000). Non- meat additives, such as salts, Iactates, and phosphates, may also enhance thermal 80 resistance of pathogens (Kotrola and Conner 1997, Maurer 2001). Doyle and Mazzotta (2000) suggested that bacterial location in the food system (surface attachment vs. interior dispersion) may also affect the resistance of Salmonella. Physical arrangement of various components within the food matrix might also cause differences in bacterial thermal resistance. Salmonella was significantly more heat resistant when present in whole muscle beef and pork as opposed to ground products (Orta-Ramirez and others 2005; Velasquez and others 2005). In addition, the cooking/heating conditions and methods (Murphy and others 2001), and bacterial recovery methodologies (Wesche and others 2005) also dictate the results obtained from thermal inactivation experiments. Under commercial conditions, thermal processing with proper cooking time and temperature is necessary for both product quality and safety. Recently, mathematical models have been used to validate various thermal processes for better food safety assurance (Leaper and Richardson 1999). Therefore, a knowledge of thermal death (inactivation) kinetics is essential to eliminate or reduce target organisms. Several factors other than processing time and temperature also influence thermal death rates of bacteria. It is important that an adequate model that includes the impact of product and process variables on the inactivation kinetics of microbial contaminants be developed. However, the validity of current thermal inactivation data from meat models needs to be reexamined for marinated products along with thermal resistance data for whole and ground muscle products to ensure product safety. Because of this concern, the objective of this study was to determine the effects of turkey physical structure, whole muscle and ground product, on the thermal resistance of Salmonella. 81 MATERIALS AND METHODS Preparation of turkey breast samples Fresh, whole muscle, boneless, skinless turkey breasts were obtained from a local packer within 12 h of slaughter and chilled in a single large lot to eliminate lot-to-lot variability. On the receiving day, turkey breasts were cored using a hand-coring device to create cylinder-shaped whole muscle samples measuring 1.2 cm in diameter and 5-7 cm in length. Ground turkey samples were obtained by grinding whole muscle turkey breasts twice through a 4-mm diameter plate using a Kitchen Aid grinder (Model kS-A, Hobart, Troy, OH, USA). Subsequently, whole and ground muscle samples were vacuum packaged in double plastic bags (Cryovac Sealed Air Corp, Duncan, SC, USA), frozen at -20°C and transported on dry ice to CFC Logistics (Quakertown, PA, USA) for irradiation. All samples were irradiated with approximately lOkGy to eliminate indigenous microflora. To determine microbial populations after irradiation, l-g samples from whole muscle turkey breast were diluting 1:5 in 0.1% sterile peptone water (Difco, Becton Dickinson, Sparks, MD, USA), homogenized in a masticator (Model 0410, IUL Instruments USA, Inc. Cincinnati, OH, USA), plated in duplicate on PetrifilmTM aerobic count plates (3M Corp, St. Paul, MN, USA), and incubated at 37 °C for 48 h. Moisture, fat, and protein levels were determined in triplicate using AOAC methods 950.46B, 991.36, and 981.1, respectively (AOAC 1996). To determine the pH, 10 g of turkey was homogenized in 90 mL of distilled water using a Polytron homogenizer (Model PT 10/3 5, Brinkmann Instruments, Westbury, NJ, USA) at speed 82 setting 3 and measured directly using a pH meter (Model 145, Coming, Medfield, MA, USA). Preparation of marinade solution The marinade solution contained 96% water (filtered and deionized), 3.2% NaCl (EMD Chemicals Inc., Gibbstown, NJ, USA) and 0.8% phosphate (50% food grade liquid potassium phosphates from Butcher and Packer Supply Company, Detroit, MI, USA). A 520 mL-aliquot of the marinade was poured into 650 mL-glass bottles with screw caps and autoclaved for 15 min at 121 °C to ensure sterility. The marinades were stored at room temperature (25°C) until used. Preparation of test organisms The following eight strains of Salmonella were obtained from Dr. V.K. Juneja (USDA-ARS, Eastern Regional Research Center, Philidelphia, PA, USA): S. Thompson FSIS 120 (chicken isolate), S. Enteriditis H3 527 and H3502 (clinical isolates phage types 13A and 4, respectively), S. Typhimurium DT 104 H3380 (human isolate), S. Hadar MF60404 (turkey isolate), S. Copenhagen 8457 (pork isolate), S. Montevideo FSIS 051 (beef isolate), and S. Heidelberg F5038BG1 (human isolate). These serovars have shown moderate to high thermal resistance and been implicated in outbreaks (J uneja and others 2001). All serovars were stored frozen at -80 °C in a tryptic soy broth (TSB) (Difco, Becton Dickinson, Sparks, MD, USA) containing 20% glycerol. Prior to use, each serovar of Salmonella was propagated weekly by transferring one loop of frozen culture to a 9-mL TSB tube. The cultures were maintained separately 83 by daily transfer to fresh TSB followed by incubation for 18-24 h at 37 °C, with a minimum of two consecutive transfers to obtain cells in late lag phase (Smith and others 2002). Preparation of inoculated marinade On the day of experiment, the 8 serovars of Salmonella were transferred to a centrifuge bottle and centrifuged at 6000 x g for 20 min at 4 °C, the supernatant was decanted off and the pellet was resuspended in 500 mL of sterile marinade to give a target Salmonella cocktail concentration of approximately 108 CFU/mL. The concentration was confirmed by platting on PetrifilmTM. The final concentration was confirmed by serial dilution in 0.1% peptone water followed by duplicate plating on aerobic PetrifilmTM count plates and incubating at 37 °C for 24 h before enumeration. Exposure to inoculated marinade Frozen whole muscle cores and ground muscle were thawed overnight at 4 °C. The core samples were transferred to sterile Whirl PakTM bags (N asco, Fort Atkinson, WI, USA) and Salmonella-inoculated marinade was added to cover the samples. Core weights were recorded before and after a 20 min of marination at 4°C to determine marinade uptake. The resulting average marinade uptake, 0.14 g marinade/ g turkey, was used to determine the amount of marinade needed for ground turkey samples. Ground turkey was manually mixed in a sterile plastic container with inoculated marinade that was aseptically added dropwise. Uniformity of inoculation was verified by randomly plating inoculated unheated whole muscle and ground turkey. Samples were diluted with 0.1% 84 peptone water and plated on aerobic PetrifilmTM count plates. Inoculated whole and ground muscle samples were subsequently packed into sterile brass tubes (1.27 cm diameter and 10 cm length), sealed with sterile rubber stoppers, and both ends of the tubes were wrapped with Teflon tape. All tubes were stored at 4°C before subjected to the thermal treatment (within 2 h). Thermal inactivation Brass tubes containing whole and ground samples were placed into a temperature controlled water bath (N ESLAB Instruments Inc., Newington, NH, USA) and heated isotherrnally at 55, 60, and 62.5°C. Internal temperature was monitored using a l-mm type T thermocouple (Omega Engineering, Stamford, CT, USA) inserted into the geometric center of a turkey sample and connected to a DuaLogR data logger (model 91 100-50, Cole Parmer Instrument Company, Vernon Hills, IL, USA). The thermal lag time, or time required for the internal temperature to reach within 05°C of target temperature, was recorded. After the internal temperature reached the target temperature, the first tube was removed from the water bath and directly placed in an ice-water bath to stop additional heat processing until enumeration. For whole muscle, samples heated at 55, 60, and 62.5 °C, tubes were removed every 5 min, 45 s, and 7 s, respectively. For ground muscle, samples heated at 55, 60, and 62.5 °C, tubes were removed every 4 min, 30 s, and 7 s, respectively. All whole and ground muscle experiments were performed in triplicate. 85 Sampling and enumeration After heating, samples were aseptically removed from the brass tubes, placed in sterile Whirl PakTM bags, diluted 1:10 in 0.1% peptone water, and homogenized for 3 min in a masticator (Model 0410, IUL Instruments, Inc., Cincinnati, OH, USA). Appropriate serial dilutions were plated on PetrifilmTM aerobic plates in duplicate. Salmonella survivors were enumerated after 48 h of incubation at 37 °C. The minimum detection level was 10 CFU/g. Data analysis Salmonella survivor curves were generated for whole muscle and ground turkey by plotting the logarithm of the survival ratio vs. holding time at each temperature. Analysis of variance (ANOVA) was used to determine the effects of muscle structure (whole vs. ground), heating temperature, and holding time on Salmonella survival. Mean values were compared using Tukey-Kramer’s test (0t=0.05). Thermal inactivation rate constants (k, min") were determined for first-order kinetic response by plotting the natural log of the Salmonella survivors vs. time. An Arrhenius-type equation (k = Boe'B‘n) was parameterized by plotting Ln( k) vs. (l/T). ANOVA was run to evaluate the effects of temperature and grinding on k. Statistical analyses were performed using JMP (Version 3.2.2, SAS Institute, Inc., Cary, NC, USA). 86 (a) 0 Log (NINO) r'o 0 1 0 20 30 40 Time (min) (b) 3 E 5 or o .r 0 1 2 3 4 5 6 Time (min) (C) S .2. E. a o .r _3 ., , , , ., . - . . 0 0.2 0.4 0.6 0.8 1 1.2 Time (m in) Figure A] Survival curves for Salmonella in whole (solid circle) and ground (open square) turkey muscle subjected to isothermal heating of; (a) 55, (b) 60, and (c) 62.5 °C. Each point is an average of 3 observations. 88 RESULTS AND DISCUSSION Raw whole and ground turkey muscle contained 72.5 i 0.2% moisture, 0.95 i 0.56% fat, and 26.7 i 0.59% protein. After receiving lOkGy irradiation, whole muscle turkey was randomly plated using PetrifilmTM to determine background flora. The control had no Salmonella initially present in the turkey samples. Figure A] shows the survivor curves generated by plotting the logarithm of the survivors (Log N/No) versus the heating time at each isothermal temperature. N/No is the fraction of the survivors, where N and No are the Salmonella count at a specific holding time and at time zero, respectively. Time zero is the time immediately after thermal lag time; the time required for the internal temperature to reach within O.5°C of the target temperature. As expected, thermal inactivation was affected by heating temperature and time. As heating temperature and holding time periods increased, a decrease in Salmonella counts was observed. At all isothermal heating temperatures, both ground and whole muscle samples exhibited log-linear decline in the number of survivors as heating time increased. No lag periods, shouldering, or tailing were evident in any of the Salmonella survivor curves suggesting a homogeneous response to heat resistance and heat transfer through the turkey (J uneja and others 2001). However, some survival curves in the literature exhibit shouldering, which may be caused by poor heat transfer through the heating medium (Murphy and others 2000; Juneja and others 2001 ). 87 ANOVA was used to assess the effect of heating time, holding temperature, and type of meat (whole and ground) on Salmonella survival. The P values are presented in Table A. 1. The analysis indicates that all of the factors and 2-way interactions affected the survival of Salmonella (P S 0.05). Table A2 shows the initial Salmonella counts in inoculated samples before heating and initial counts at time Zero (after thermal lag time). The results indicated no differences (P>0.05) in initial Salmonella counts between muscle types, which means they started with a comparable microbial load. In addition, the come- up times for both types of turkey meat were not significantly different (P>0.05). Several factors including initial cell numbers (Smith and others 2001), bacterial strain and growth phase (Heddleson and other 1991), heating medium composition (Ahmed and others 1995; Lorca and others 2003, sample size and shape (Smith and others 2001), ice crystal formation (Smith and others 2001; Doyle and Cliver 1990), and additives (Murphy and others 2000), can account for increased thermal stability of pathogens. However, the factors mentioned above were experimentally controlled in this study. 89 Table A.1 P values from ANOVA for thermal inactivation of Salmonella spp. in whole and ground turkey breast muscle. Test factors P values Grinding 0009 Temperature 0.0072 Time <0.0001 Grinding*Temperature 0.0034 Grinding*Time 0.0019 Temperature*Time <0_0001 Table A2 Salmonella numbers in inoculated unheated and heated turkey breast samples after thermal lag time at 55, 60, and 62.5 °C. Salmonella survival count‘ (Log CF U/ g) Heating Unheated sample Heated sample at time zero temperature Whole muscle Ground muscle Whole muscle Ground muscle 55 7.38 i 0.1a 7.28 i 0.1a 7.09 i 0.1d 7.43 i 0.6d 60 6.72 i 0.7b 7.29 i 0.1b 6.48 : 02° 6.85 i 0.1c 62-5 7.30 i o.1° 7.27 i 0.0c 6.62 i 0.1f 5.33 i 1.2f * Salmonella numbers shown are the means of three replications and expressed as mean d: standard deviation. “'f indicate significant difference (P S 0.05) of mean Salmonella survival number between muscle types within heating temperature. 90 Assuming log-linear inactivation kinetics, the first-order kinetic rate constant or thermal inactivation rate constant (k, time'l) was obtained from linear regression of Ln N/NO = -kt; where t is the heating time. Within the type of muscle, the calculated k values increased as the heating temperature increased (Table A3). These findings are supported by research done by Orta-Ramirez and others (1997), Veeramuthu and others (1998), and Murphy and others (2000). Veeramuthu and others (1998) and Orta-Ramirez and others (1997) who evaluated themal inactivation of S. Senftenberg in turkey thigh meat at 55- 65 °C and ground beef at 53-68 °C, respectively. They reported increased kinetic constants with increased heating temperatures. In the study performed by Murphy and others (2000) which determined thermal kinetics in chicken breast meat at 55-70 °C the kinetic rate constants for Salmonella at 55, 60 and 62.5 °C were 0.08, 0.39, and 0.92, respectively, which are approximately 40-70% lower than ours. The difference in bacterial thermal resistance is likely from variation in Salmonella serovars and heating medium composition. 91 Table A3 F irst-order thermal inactivation rate constants (k, min") determined from linear regression of Salmonella survivor data (Ln N/NO = -kt) for whole and ground turkey breast muscle. k values}. Muscle type 55 °C 60 °C 62.5 °C Whole muscle 0.14 i: 0.01a 1.25 :t 0.16a 3.43 i 1.093 Ground muscle 0.23 a: 0.05b 2.54 :l: 0.04b 6.25 :1: 0.62b * k values shown are the means of three replications and expressed as mean i standard deviation. a‘b indicate significant difference (P s 0.05) of mean k values within similar heating temperature. 92 When the kinetic rate constants between muscle structure were compared, k values for whole muscle were approximately 50% lower (P<0.05) than those for ground muscle at all heating process (Table A3). In these experiments, both muscle types contained equivalent raw meat composition, bacterial counts before and after thermal lag time, and were subjected to the same thermal process. As a result, the only difference accounting for higher Salmonella heat resistance in whole muscle was due to its muscle structure. Salmonella seemed to be protected from heat penetration within intact whole muscle as opposed to ground turkey. The grinding process affects microstructure and physical characteristics of the product. The organized tissue structure of whole muscle will be damaged during grinding, thus creating larger inter-spaces within the muscle tissue and a more porous structure. This porous structure is an important factor that influences heat transfer within the meat (N gadi and others 2001). Moreover, these structural changes might expel sarcoplasmic fluid from muscle fibers, resulting in increased water availability between muscle cells for heat transfer. These data will make a positive contribution to food safety for manufacturers whose processes involve a heat treatment step. Understanding these variations is necessary in order to design adequate thermal inactivation to eliminate Salmonella in thermally processed foods. However, others factors (such as meat composition, thermal history of bacteria, heating medium and condition, and additional additives) also need to be considered when developing a thermal inactivation model in order to verify processing adequacy for meat and poultry products. 93 CONCLUSIONS In this chapter, Salmonella was inoculated into whole and ground muscle and the thermal resistance was determined. The thermal inactivation rate constants or k values were affected by heating time and temperature. Results suggest that internalization of Salmonella in whole muscle turkey breast leads to enhanced thermal resistance. These findings should assist the food industry in designing Hazard Analysis Critical Control Point (HACCP) plans to effectively eliminate the pathogen in thermally processed turkey, including whole and ground muscle. The cooking times and temperatures currently recommended for pathogen inactivation may need significant adjustment to cover all aspects and parameters as the products and processes change. Accountability for product physical state will reinforce regulatory standards. If more detailed predictive model is developed, ready-to-eat poultry processors will be able to better meet lethality performance standards, and ensure the safety and quality of their products. 94 APPENDIX B SALMONELLA VIABILITY IN SALT AND PHOSPHATE MARINADES AS AFFECTED BY STORAGE TEMPERATURE ABSTRACT The effect of salt and phosphate concentration and storage temperature on Salmonella viability in marinade was investigated. An eight-strain Salmonella cocktail (~ 8 log CFU/mL) was inoculated into two different marinade formulations; “typical” and “high” concentrations, which contained 3.2% NaCl and 0.8% phosphate, and 10.2% salt and 4.05% phosphate, respectively. The marinade solutions were stored at 4 and 37 °C and sampled for surviving Salmonella for up to 63 d. The Salmonella population in a typical marinade solution held at 37 °C decreased 2.85 logs compared to 0.8 log after 62 days at 4 °C. After 63 days, a reduction of 1.4 logs was observed with the high concentration marinade stored at 4 °C. In contrast, rapid death of Salmonella was observed when marinade containing high salt and phosphate was kept at 37 °C. Consequently, Salmonella can potentially persist in marinades during long- term storage at 4 °C. 95 INTRODUCTION Tenderness and juiciness are the two sensory attributes that consumers use to evaluate meat products to determine satisfaction (Smith and Acton 2001). To enhance these factors, the meat industry injects non-meat ingredients into meat products, which can work synergistically to enhance their functionality. Water, salt and phosphate are often incorporated into marinade solutions to achieve this purpose. Several food processing procedures may introduce unexpected pathogens into the finished meat product. One of those steps is the preparation of brine solutions that may be contaminated with pathogenic organisms. Yet, marination procedures are used to reduce and prevent the growth of spoilage organisms. Marinades can support the survival of selected known pathogens. Perko-Makela (2000) studied the survival rate of seven Campylobacterjejum' strains in a marinade sauce and demonstrated that a commercial marination procedure did not eliminate this microorganism. Factors affecting microbial growth in foods include water activity (aw), pH, and temperature. Minimum growth conditions for Salmonella are: 6.5 °C, pH 4.5, and aw >0.95 (FDA-CFSAN 1992). Values that deviate from these conditions will affect the growth and survival of Salmonella. These bacteria tolerate many stressful conditions and can survive sub-optimal conditions for a period of time. While the original pH of a food may prevent growth of Salmonella, a shift in pH by other microbes may permit bacterial growth. Water activity varies with the change of temperature and decreases with the addition of solute. Mattick and others (2000) reported low aw values when supplementing growth media with several humectants that affected the survival of Salmonella Enteritidis 96 and Typhimurium. The survival of Salmonella at a specific temperature depends on the specific Salmonella serovar, their growth phase in the media, their physical conditions, food composition or testing media, and competing microflora (Doyle and Mazzotta, 2000). Airoldi and Zottola (1988) found that Salmonella Typhimurium was viable at low temperatures for extended incubation times. Moreover, the temperature associated with the propagation period affected the growth of organisms in the first several days. Since Salmonella is recognized as a significant foodborne pathogen that has been involved in major outbreaks, the ability of this organism to survive in marinade solutions is a safety concern. The objective of this study was to monitor survival of Salmonella in marinade solution containing different salt and phosphate concentrations stored at 4 and 37 °C for up to 63 days. 97 MATERIALS AND METHODS Marinade preparation Two marinades were prepared: a typical marinade containing 96% water (filtered and deionized), 3.2% NaCl (EMD Chemicals Inc., Gibbstown, NJ, USA), and 0.8% phosphate (50% food grade liquid potassium phosphates from Butcher and Packer Supply Company, Detroit, MI, USA) and a high concentration marinade containing 10.2% NaCl and 4.05% phosphate. A 520 mL-aliquot of each marinade was poured into glass bottles with screw caps and autoclaved for 15 min at 121 °C to ensure sterility. The pH of the marinade was measured using Corning 340 pH meter (Corning, Suffolk, UK). Preparation of test organisms The following eight strains of Salmonella were obtained from Dr. V.K. J uneja (USDA-ARS, Eastern Regional Research Center, Philidelphia, PA, USA): S. Thompson FSIS 120 (chicken isolate), S. Enteriditis H3527 and H3 502 (clinical isolates phage types 13A and 4, respectively), S. Typhimurium DT 104 H3380 (human isolate), S. Hadar MF60404 (turkey isolate), S. Copenhagen 8457 (pork isolate), S. Montevideo FSIS 051 (beef isolate), and S. Heidelberg F503 8BG1 (human isolate). These serovars have shown moderate to high thermal resistance and been implicated in outbreaks (J uneja and others 2001). All serovars were stored frozen at —80 °C in a tryptic soy broth (TSB) (Difco, Becton Dickinson, Sparks, MD, USA) containing 20% glycerol. Each strain of Salmonella was propagated by transferring one loop of frozen culture to a 9-mL TSB tube. The cultures were maintained by daily transfer to fresh TSB 98 followed by incubation for 18-24 h at 37 °C, with a minimum of two consecutive transfers prior to use. Preparation of inoculated marinade On the day of the experiment, the eight serovars of Salmonella were combined in a centrifuge bottle and centrifuged at 6000 x g for 20 min at 4°C, the supernatant was decanted off and the cells were washed with 50 mL of sterile peptone water, pelleted by centrifugation and resuspended in 500 mL of sterile marinade to give a final Salmonella cocktail containing ~108 CFU/mL. Three bottles of each marinade concentration were capped and stored at 4°C and 37°C for subsequent microbial analyses. Sampling Salmonella populations were determined by serially diluting aliquots in sterile 0.1% peptone water, plating in duplicate on PetrifilmTM aerobic plates and incubating at 37°C for 48 h. All marinades were sampled for Salmonella for up to 63 d or until population decreased below the detection limit of 10 CFU/mL. Data analysis All treatments for each marinade concentration and temperature combination were conducted in triplicate on every sampling day. Numbers of Salmonella were plotted as log CFU/mL against incubation time on a semi-log scale. 99 RESULTS AND DISCUSSION Salmonella populations in marinades stored at 4 and 37 °C illustrated in Figure B 1. Initial Salmonella populations in the inoculated marinade ranged from 7.53 to 8.23 log CFU/mL with numbers of Salmonella decreasing as storage time progressed. Viability of Salmonella was highest in the typical marinade stored at 4 °C. Exposure to high salt and phosphate concentrations and a temperature of 37 °C dramatically reduced the survival of Salmonella with populations decreasing to less than 10 CFU/mL after 5 days. Regardless of storage temperature, the high concentration marinade was more detrimental to survival compared to the typical marinade. The more rapid decline of Salmonella in the high concentration marinade may relate to the increased levels of solutes that decrease the water activity to the point where the organism are no longer able to survive. The minimum aw for survival of Salmonella is reportedly 0.95 (FDA 1992). Mattick and others (2000) reported the optimal growth of Salmonella strains at aw of 0.99. High salt concentrations are associated with low water activity. A 3.2 % salt concentration is equivalent to an aW of 0.98, which would allow greater survival of Salmonella in marinade. In contrast, the high concentration marinade containing 10.2% salt had a survival, which limits aw to approximately 0.92. Fletcher and Csonka (1998) reported that the addition of solutes (i.e. salt and sugar) can raise the upper limit growth temperature and viability of microorganism at lethal temperature. Their results showed that addition of 0.9-1 .7% of NaCl could enhance the survival of S. Typhimurium at 50 °C. 100 Within the same concentration of salt and phosphate in the marinade, lower storage temperature maintained higher viability of Salmonella. For the typical marinade solution, only a 0.8-log reduction was observed at 4 °C compared to a 2.8-log reduction at 37 °C. In the marinade solution containing high salt and phosphate, the Salmonella population decreased 1.4 logs during refrigerated storage, as opposed to about 8 logs when the marinade was stored at 37 °C. Previous work by Warsow (2008) who examined the survival of a Salmonella cocktail in a specific marinade supports our findings that the microorganism survived longer at 4 °C. Warsow’s marinade solution contained 7% salt and 3% phosphate. This study revealed greater viability of Salmonella when stored at 4 °C compared to 37 °C. Airoldi and Zottola (1988) examined the growth and survival of S. Typhimurium in nutrient deficient media incubated at low temperature. They found that Salmonella could survive in 0.1% peptone water for more than 1 year at 7 °C. Mattick and others (2000) demonstrated similar results to our study, describing improved survival in the presence of high salt concentrations at lower temperatures. At a low aw, a rapid decrease in Salmonella survival was observed at 37 °C compared to 21 °C. It was concluded that survival does not depend solely on water activity in growing media but also on a combination of growth condition factors. This same research group also studied long-terrn survival of Salmonella strains at low water activity. Salmonella survived at low aw for extended periods; however, an 8% NaCl (~ aW 0.95) was bactericidal. In addition, pH was an important factor influencing the survival of Salmonella at low aW when exposed to heat (Mattick and others 2001). Rather than having typical short cell morphology, filamentous phenotypes have been found in organisms that are under stress conditions. Mattick and others (2000) demonstrated that, 101 in response to sub-optimal growth conditions, filaments were observed when S. Typhimurium was incubated at both 21 and 37 °C in tryptic soy broth (TSB) supplemented with NaCl. With NaCl at aw values of 0.95-0.98 (approximately 3.5-8 % NaCl), the filaments appeared to be longer and more numerous. They explained that the formation of filaments was due to the low water activity that may affect the regulation of cell division genes. However, rehydration with fresh TSB containing no salt revealed rapid filamentous division into large numbers of viable daughter typical cells. 102 :0 D D. D. D. D. O O O LogCFUImL DAMN-#U'IOINCDCO o . ..__l___zn_.‘. - L ' >‘<’x><'x Xxxxxx‘ A o . D 20 30 40 50 60 70 nmew) Figure B.l Viability of an 8-strain Salmonella cocktail in typical marinade containing 3.2% salt and 0.8% phosphate maintained at 4 °C (solid circle) and 37 °C (solid square) and high concentration marinade containing 10.2% salt and 4.05% phosphate stored at 4 °C (open triangle) and 37 °C (cross). Each point is an average of 3 observations. 103 CONCLUSIONS Most foods contain sufficient nutrients to support microbial growth. Several factors support, prevent, or limit the growth of organisms in foods. Various solutes are incorporated into foods in order to reduce the water activity and maintain a reasonable safety margin before growth of microorganisms can occur. In addition to water activity of the culture medium, other endogenous and exogenous factors also affect the survival of Salmonella. The interplay of factors will significantly determine whether a specific microorganism will survive or grow in a given food. A synergism or antagonism might occur altering the survival of a particular microbe from one product to another. Therefore bacterial survival prediction can only be achieved through experimentation. This study assessed the viability of Salmonella in a marinade solution containing different salt and phosphate concentrations that was stored at 4 and 37 °C. Salmonella survived sub-optimal grth conditions for extended periods of time at refrigeration temperature. The properties of marinade did impact the survival of Salmonella. Highest survival was observed when both marinades (3.2% salt + 0.8% phosphate and 10.2 % salt + 4.05% phosphate) were kept at 4 °C. A combination of high storage temperature and high salt and phosphate concentration in marinade was most detrimental to the survival. However, extended survival was seen in the high concentration marinade at 4 C°. One goal of industrial marination procedures is to rapidly achieve marinade equilibrium in the products. Consequently, if a Salmonella contaminated marinade is incorporated into meat, it is possible that Salmonella could survive during refiigerated 104 storage of marinated poultry products. Therefore, it is critical that these marinated products be adequately cooked to eliminate any potentially internalized salmonellae. 105 (1) (2) APPENDIX C ADDITIONAL MICROSCOPY IMAGES O O 106 (3) O Figure C. 1 -3 Enhanced fluorescence images of GFP-labeled Salmonella location (circle areas) in turkey breast transverse sections taken at 400K magnification. Images in this dissertation are presented in color. 107 APPENDIX D BIBLIOGRAPHY FOR APPENDICES A AND B Ahmed NM, Conner DE, Huffman DL. 1995. Heat-resistance of Escherichia coli 0157:H7 in meat and poultry as affected by product composition. J Food Sci 60:606-610. Airoldi AA, Zottola EA. 1988. Growth and survival of Salmonella typhimurium at low temperature in nutrient deficient media. J Food Sci 53: 151 1-1 513. AOAC. 1996. Official Methods of Analysis. 16th ed. Virginia: Assn. of Official Anal. Chem. Doyle ME, Mazzotta AS. 2000. Review of studies on the thermal resistance of , Salmonella. J Food Prot 63:779-795. Doyle MP, Beuchat LR, Montville TJ. 2001. Food microbiology: Fundamentals and frontiers. 2nd Ed. American Society for Microbiology Press. 872 p. Doyle MP, Cliver DO. 1990. Salmonella. In: Oliver DO, editor. Foodborne diseases. Academic Press. p 186-204. FDA-CF SAN. 1992. Foodborne pathogenic microorganisms and natural toxins handbook: bad bug book: Salmonella spp. Food and Drug Administration. Center for Food Safety and Applied Nutrition. http://www.cfsan.fda.gov/~mow/chap1.html. Last accessed October 17, 2007. Fletcher SA, Csonka LN. 1998. Characterization of the induction of increased thermotolerance by high osmolarity in Salmonella. Food Microbiol 15:307-317. Heddleson RA, Doores S, Anantheswaran RC, Kuhn GD, Mast MG. 1991. Survival of Salmonella species heated by microwave energy in a liquid menstruum containing food components. J Food Prot 54:637-642. Juneja VK, Eblen BS. 2000. Heat inactivation of Salmonella Typhimurium DT104 in beef as affected by fat content. Lett Appl Microbiol 30:461-467. Juneja VK, Eblen BS, Marmer BS, Williams AC, Palumbo SA, Miller AJ. 1995. Thermal resistance of nonproteolytic type B and type B Clostridium botulinum spores in phosphate buffer and turkey slurry. J Food Prot 58:75 8-763. 108 Juneja VK, Eblen BS, Ransom GM. 2001. Thermal inactivation of Salmonella spp. in chicken broth, beef, pork, turkey, and chicken: determination of D- and Z-values. J Food Sci 66:146-152. Kotrola J S, Conner DE. 1997. Heat inactivation of Escherichia coli 0157:H7 in turkey meat as affected by sodium chloride, sodium lactate, polyphosphate, and fat content. J Food Prot 60:898-902. Leaper S, Richardson P. 1999. Validation of thermal process control for the assurance of food safety. Food Control 10:281-283. Lorca TA, Claus JR, Eifert JD, Marcy J E, Sumner SS. 2003. Penetration of surface- inoculated bacteria as a result of electrically generated hydrodynamic shock wave treatment of boneless skinless chicken breasts. Poult Science 82:1205-1210. Mattick KL, Jorgensen F, Legan JD, Cole MB, Porter J, Lappin-Scott HM, Humphrey TJ. 2000. Survival and filamentation of Salmonella enterica serovar Enteritidis PT4 and Salmonella enterica serovar Typhimurium DT104 at low water activity. Appl Environ Microbiol 66: 1274-1279. Mattick KL, J orgensen F, Wang P, Pound J, Vandeven MH, Ward LR, Legan JD, Lappin-Scott HM, Humphrey TJ. 2001. Effect of challenge temperature and solute type on heat tolerance of Salmonella serovars at low water activity. Appl Environ Microbiol 67:4128-4136. Maurer J L. 2001. Environmental effects on the thermal resistance of Salmonella, Escherichia coli 0157:H7, and triose phosphate isomerase in ground turkey and beef. Michigan: MS. Thesis Michigan State University. Murphy RY, Johnson ER, Marks BP, Johnson MG, Marcy JA. 2001. Thermal inactivation of Salmonella senftenberg and Listeria innocua in ground chicken breast patties processed in an air convection oven. Poult Sci 80:515-521. Murphy RY, Marks BP, Johnson ER, Johnson MG. 2000. Thermal inactivation kinetics of Salmonella and Listeria in ground chicken breast meat and liquid medium. J Food Sci 65:706-710. Ngadi MO, Kassama LS, Raghavan GSV. 2001. Porosity and pore size distribution in cooked meat patties containing soy protein. Can Biosyst Eng 43 :3.17-1 1.24. Orta-Ramirez A, Marks BP, Warsow CR, Booren AM, Ryser ET. 2005. Enhanced thermal resistance of Salmonella in whole muscle compared to ground beef. J Food Sci 70:359-362. 109 Orta-Ramirez A, Price JF, Hsu Y, Veeramuthu GJ, Cherry-Merritt J S, Smith DM. 1997. Thermal inactivation of Escherichia coli 0157:H7, Salmonella senftenberg, and enzymes potential as time-temperature indicators in ground beef. J Food Prot 60:471-475. Perko-Makela P, Koljonen M, Miettinen M, Hanninen ML. 2000. Survival of Campylobacter jejuni in marinated and nonmarinated chicken products. J Food Safety 20:209-216. Raccach M, Henrickson RL. 1979. Microbial aspects of mechanical tenderization of beef. J Food Prot 42:971-973. Smith SE, Maurer JL, Orta-Ramirez A, Ryser ET, Smith DM. 2001. Thermal inactivation of Salmonella spp., Salmonella lyphimurium DT104, and Escherichia coli 0157:H7 in ground beef. J Food Sci 66:1164-1168. Veeramuthu GJ, Price J F, Davis CE, Booren AM. 1998. Thermal inactivation of Escherichia coli 0157:H7, Salmonella senftenberg, and enzymes with potential as time-temperature indicators in ground turkey thigh meat. J Food Prot 61:171-175. Velasquez A, Tuntivanich V, Orta—Ramirez A, Booren AM, Marks ET, Ryser ET. 2005. Enhanced thermal resistance of Salmonella in marinated whole-muscle vs. ground pork. Abstr Inst Food Tech Ann Mtg. Abstract No. 89E-14. Warsow CR, Orta-Ramirez A, Marks BP, Ryser ET, Booren AM. 2008. Single directional migration of Salmonella into marinated whole muscle turkey breast. J Food Prot 71 :153-156. Wesche AM, Marks BP, Ryser ET. 2005. Thermal resistance of heat-, cold-, and starvation-injured Salmonella in irradiated comminuted turkey. J Food Prot 68:942-948. 110 BIBLIOGRAPHY Aberle ED, Forrest JC, Gerrard DE, Mills EW. 2001. Principles of Meat Science. 4th ed. Iowa: Kendall/Hunt Publishing Co. 354 p. Ahmed NM, Conner DE, Huffman DL. 1995. Heat-resistance of Escherichia coli 0157:H7 in meat and poultry as affected by product composition. J Food Sci 60:606-610. Anderson ME, Marshall RT, Dickson J S. 1992. Estimating depths of bacterial penetration into post-rigor carcass tissue during washing. J Food Safety 12:191-198. Birkhold SG, Sams AR. 1995. Comparative ultrastructure of pectoralis fibers from electrically stimulated and muscle-tensioned broiler carcasses. Poult Sci 74:194- 200. Bocker U, Kohler A, Bertram HC, Anderson HJ, Ofstad R. 2005. Effects of salting on microstructure and protein secondary structure as measured by FT-IR microspectroscopy in porcine M. semitendinosus of different quality. 5 1 st International congress of meat science and technology 607-612. Boyd KJ, Ockerrnan HW, Plimpton RF. 1978. Sensory characteristics and microbiological evaluation of stored mechanically tenderized beef semimembranosus muscle. J Food Sci 43:670-672, 676. Butler JL, Stewart JC, Vanderzant C, Carpenter ZL, Smith GC. 1979. Attachment of microorganisms to pork skin and surfaces of beef and lamb carcasses. J. Food Prot 42:401-406. Carlson TR, Marks BP, Booren AM, Ryser ET, Orta-Ramirez A. 2005. Efffect of water activity on thermal inactivation of Salmonella in ground turkey. J Food Sci 702M363-366. CDC. 2005a. Annual summary: Salmonella. Centers for disease control and prevention. 94 p. CDC. 2005b. Foodborne illness. Centers for disease control and prevention. Division of Bacterial and Mycotic Diseases. http://www.ch.gov/ncidod/dbmd/diseaseinfo/foodbomeinfections_g.htrn#mostco mmon. Last accessed October 16, 2007. CDC. 2005c. Salmonellosis. Centers for disease control and prevention. Division of Bacterial and Mycotic Diseases. 111 http://www.cdc.gov/ncidod/dbmd/diseaseinfo/salmonellosis_t.htm. Last accessed October 15, 2007. CDC. 2007. Salmonellosis-outbreak investigation, February 2007. Centers for disease control and prevention. Division of Bacterial and Mycotic Diseases. http://www.cdc.gov/ncidod/dbmd/diseaseinfo/salmonellosis_2007/outbreak_notic e.htm. Last accessed October 16, 2007. Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC. 1994. Green fluorescent protein as a marker for gene expression. Science 263:802-805. Cho JC, Kim SJ. 1999. Viable, but non-culturable, state of a green fluorescence protein- tagged environmental isolate of Salmonella typhi in groundwater and pond water. FEMS Micro Biol Lett 170:257-264. Cubitt AB, Heim R, Adams SR, Boyd AE, Gross LA, Tsien RY. 1995. Understanding, improving and using green fluorescent proteins. Trends Biochem Sci 20:448-455. Dickson J S, Koohmaraie M. 1989. Cell surface charge characteristics and their relationship to bacterial attachment to meat surfaces. Appl Environ Microbiol 55:832-836. Do G-S, Sagara Y, Tabata M, Kudoh K-i, Higuchi T. 2004. Three-dimensional measurement of ice crystals in frozen beef with a micro-slicer image processing system. Int J Refrig 27:184-190. Doyle ME, Mazzotta AS. 2000. Review of studies on the thermal resistance of Salmonella. J. Food Prot. 63:779-795. Doyle MP, Beuchat LR, Montville TJ. 2001. Food microbiology: Fundamentals and frontiers. 2nd Ed. Washington: American Society for Microbiology Press. 872 p. Dziezak JD. 1990. Phosphates improve many foods. Food Technol 44:80-92. Egelandsdal B, Martinsen BK, Autio K. 1995. Rheological parameters of protein functionality: a medel study using myofibrils from red and white bovine muscles. Meat Sci 66:915-924. Elmossalami E, Wassef N. 1971. Penetration of some microorganisms in meat. Zentralb Veterinaerrned Reihe B 18:329-336. FDA-CFSAN. 1992. F oodbome pathogenic microorganisms and natural toxins handbook: bad bug book: Salmonella spp. Food and Drug Administration. Center for Food Safety and Applied Nutrition. http://www.cfsan.fda.gov/~mow/chap1.html. Last accessed October 17, 2007. 112 Firstenberg-Eden R. 1981. Attachment of bacteria to meat surfaces: a review. J Food Prot 44:602-607. Flegler SL, Heckman, Jr., J .W., and Klomparens, KL. 1993. Scanning and transmission electron microscopy: an introduction. New York: Oxford University Press. 225 p. Fratamico PM, Deng MY, Strobaugh TP, Palumbo SA. 1997. Construction and characterization of Escherichia coli 0157:H7 strains expressing firefly luciferase and green fluorescent protein and their use in survival studies. J Food Prot 60:1167-1173. F reitag NE, Jacobs KE. 1999. Examination of Listeria monocytogenes intracellular gene expression by using the green fluorescent protein of Aequorea Victoria. Infect Immun 67:1844-1852. Froning GW, Sackett B. 1985. Effect of salt and phosphates during tumbling of turkey breast muscle on meat characteristics. Poult Sci 64:1328-1333. Gill CO, Penney N. 1977. Penetration of bacteria into meat. Appl Environ Microbiol 33:1284-1286. Gill CO, Penney N. 1982. Bacterial penetration of muscle tissue. J Food Sci 47:690-691. Gupta LK, Kalra MS, Singh A, Singh B. 1983. Penetration of bacteria into muscles of goat, pork, and poultry meat. J Food Sci and Tech 20:13-16. Hayman MM, Baxter I, Oriordan PJ, Stewart CM. 2004. Effects of high-pressure processing on the safety, quality, and shelf life of ready-to-eat meats. J Food Prot 67:1709-1718. Juneja VK, Eblen BS. 2000. Heat inactivation of Salmonella typhimuriurn DT104 in beef as affected by fat content. Lett Appl Microbiol 30:461-467. Juneja VK, Eblen BS, Manner BS, Williams AC, Palumbo SA, Miller AJ. 1995. Thermal resistance of nonproteolytic type B and type B Clostridium botulinum spores in phosphate buffer and turkey slurry. J Food Prot 58:758-763. J uneja VK, Eblen BS, Ransom GM. 2001. Thermal inactivation of Salmonella spp. in chicken broth, beef, pork, turkey, and chicken: determination of D- and Z-values. J Food Sci 66:146-152. J uneja VK, Marks HM, Mohr T. 2003. Predictive thermal inactivation model for effects of temperature, sodium lactate, NaCl, and sodium pyrophosphate on Salmonella serotypes in ground beef. Appl Environ Microbiol 69:5138-5156. 113 Juneja VK, Snyder OP, Marmer BS. 1997. Thermal destruction of Escherichia coli 0157:H7 in beef and chicken: Determination of D and 2 values. Int J Food Microbiol 35:231-237. Kim J -W, Doores S. 1993. Influence of three defeathering systems on microtopography of turkey skin and adhesion of Salmonella Typhimurium. J Food Prot 56:286-291, 305. Kim J, Slavik MF, Griffis CL, Walker JT. 1993. Attachment of Salmonella Typhimurium to skins of chicken scalded at various temperatures. J Food Prot 56:661-671. Kim KY, Lillard HS, Frank JF. 1996. Three-dimensional visualization of Salmonella attachment to poultry skin using confocal scanning laser microscopy. Lett Appl Microbiol 22:280-282. Kotrola J S, Conner DE. 1997. Heat inactivation of Escherichia coli 0157:H7 in turkey meat as affected by sodium chloride, sodium lactate, polyphosphate, and fat content. J Food Prot 60:898-902. Li R, Kerr WL, Toledo RT, Teng Q. 2001. 31P NMR analysis of chicken breast meat ‘ vacuum tumbled with NaCl and various phosphates. J Sci Food Agric 81 :576- 582. Lillard HS. 1989. Factors affecting the persistence of Salmonella during the processing of poultry. J Food Prot 52:829-832. Liu G, Xiong YL. 1997. Gelation of chicken muscle myofibrillar proteins treated with protease inhibitors and phosphates. J Agric Food Chem 45:3437-3442. Lorca TA, Claus JR, Eifert JD, Marcy JE, Sumner SS. 2003. Penetration of surface- inoculated bacteria as a result of electrically generated hydrodynamic shock wave treatment of boneless skinless chicken breasts. Poult Sci 82:1205-1210. Lorca TA, Pierson MD, Claus JR, Eifert JD, Marcy JE, Summer SS. 2002. Penetration of surface-inoculated bacteria as a result of hydrodynamic shock wave treatment of beef steaks. J Food Prot 65:616-620. Lyon BG, Hamm D. 1986. Effects of mechanical tenderization with sodium chloride and polyphosphates on sensory attributes and shear values of hot-stripped broiler breast meat. Poult Sci 65: 1702-1707. Madigan MT, Martinko JM, Parker J. 2000. Brock: Biology of microorganisms. 9th ed. New Jersey: Prentice hall. 114 Maurer J L. 2001. Environmental effects on the thermal resistance of Salmonella, Escherichia coli 0157:H7, and triose phosphate isomerase in ground turkey and beef. Michigan: MS. Thesis Michigan State University. Murphy RY, Marks BP, Johnson ER, Johnson MG. 2000. Thermal inactivation kinetics of Salmonella and Listeria in ground chicken breast meat and liquid medium. J Food Sci 65:706-710. NACMCF. 2004. Requisite scientific parameters for establishing the equivalence of alternative methods of pasteurization. National Advisory Committee on microbiological criteria for foods. August 27, 2004. Washington. 66 p. National Turkey Federation. 2004. Turkey nutrition. http://eatturkey.com/consumer/nutritionhtml. Last accessed October 15, 2007. Niven C. 1987. Microbiology and parasitology of meat; part l-microbiology. In: Price JF, Schweigert BS, editors. The science of meat and meat products. 3rd ed. Connecticut: Food & Nutrition Press. p 217-263. Noah CW, Shaw CI, Ikeda J S, Kreuzer KS, Sofos JN. 2005. Development of green fluorescent protein-expressing bacterial strains and evaluation for potential use as positive controls in sample analyses. J Food Prot 68:680-686. Notermans S, Kampelmacher EH. 1983. Attachment of bacteria in meat processing. Fleischwirtsch 63:72-73, 76-78. Offer OW, Knight P, Jeacocke R, Almond R, Cousins T, Elsey J, Parsons N, Sharp A, Starr R, Purslow P. 1989. The structural basis of the water-holding, appearance and toughness of meat and meat-products. Food Microstruct 8: 151-170. Offer OW, Trinick J. 1983. On the mechanism of water holding in meat: the swelling and shrinking of myofibrils. Meat Sci 8:245-281. Orta-Ramirez A, Marks BP, Warsow CR, Booren AM, Ryser ET. 2005. Enhanced thermal resistance of Salmonella in whole muscle compared to ground beef. J Food Sci 70:359-362. Orta-Ramirez A, Smith DM. 2002. Thermal inactivation of pathogens and verification of adequate cooking in meat and poultry products. Adv Food Nutr Res 44:148-194. Paddock SW. 1999. Confocal microscopy: methods and protocals. Humana Press. Totowa, NJ. 446 p. Pearson AM, Gillett TA. 1999. Processed meat. 3rd ed. Maryland: Aspen Publisher, Inc. 448 p. 115 Phebus RK, Marsden J L, Thippareddi H, Sporing S, Kastner C. 1999. Blade tenderization and food safety. Proc Recip Meat Conf 52:71-72. Podolak R, Solomon MB, Patel JR, Liu MN. 2006. Effect of hydrodynamic pressure processing on the survival of Escherichia coli 0157:H7 in ground beef. Innovative Food Science and Emerging Technologies 7:28-31. Pohlman FW, Dikeman ME, Zayas J F, Unruh JA. 1997. Effects of ultrasound and convection cooking to different end point temperatures on cooking characteristics, shear force and sensory properties, composition, and microscopic morphology of beef longissimus and pectoralis muscles. J Anim Sci 75:386-401. Prachaiyo P, McLandsborough LA. 2000. A microscopic method to visualize Escherichia coli interaction with beef muscle. J Food Prot 63:427-433. Raccach M, Henrickson RL. 1979. Microbial aspects of mechanical tenderization of beef. J Food Prot 422971-973. Richardson RI, Mead GC. 1999. Poultry meat science. Vol 25. New York: CABI Publishing. 444 p. Ross MH, Kaye GI, Pawlina W. 2003. Histology: a text and atlas. Philadelphia: Lippincott Williams & Wilkins. 875 p. Russell J. 2002. Annual deli report: fresh start. Oct. 2002. Natl Provis 216: 14. Schwartz WC, Mandigo RW. 1976. Effect of salt, sodium tripolyphosphate and storage on restructed pork. J Food Sci 41 :1266-1269. Scott VN, Stevenson KE. 2006. HACCP a systematic approach to food safety. 4rd ed. Washington: Food Products Association. Shults GW, Russell DR, Wierbicki E. 1972. Effect of condensed phosphates on pH, swelling and water-holding capacity of beef. Food Res 37:860-864. Sikes A, Maxcy RB. 1980. Postmortem invasion of muscle food by a proteolytic bacterium. J food Sci 45:293-296. Silva TJP, Bracker CE, Orcutt MW, Forrest JC, Judge MD. 1992. Preparation of muscle samples for comparative electron microscopy. J Anim Sci 70:2452-2461. Smith DP, Acton JC. 2001. Marination, cooking, and curing of poultry products. In: Sams AR, editor. Poultry meat processing. New York: CRC Press LLC. p 257- 279. 116 Smith SE, Maurer JL, Orta-Ramirez A, Ryser ET, Smith DM. 2001. Thermal inactivation of Salmonella spp., Salmonella Typhimurium DT104, and Escherichia coli 0157:H7 in ground beef. J Food Sci 66:1164-1168. Solomon MB, Long JB, Eastridge J S. 1997. The hydrodyne: A new process to improve beef tenderness. J Anim Sci 75:1534-1537. Takeuchi K, Matute CM, Hassan AN, Frank JF. 2000. Comparison of the attachment of Escherichia coli 0157:H7, Listeria monocytogenes, Salmonella Typhimurium, and Pseudomonasfluorescens to lettuce leaves. J Food Prot 63:1433-1437. Thomas CJ, 0' Rourke RD, McMeekin TA. 1987. Bacterial penetration of chicken breast muscle. Food Microbiol 4:87-95. USDA. 2007. Red meat, poultry, and fish (boneless, trimmed equivalent): per capita availability. US. Department of Agriculture/Economic Research Service. http://www.ers.usda.gov/Data/FoodConsumption/spreadsheets/mtpcc.xls. Last accessed October 15, 2007. USDA-F SIS. 1999. Performance standards for the production of certain meat and poultry products final rule. 9 CFR Parts 301, 317, 318, 320, and 381. US. Department of Agriculture Food Safety and Inspection Service, Washington, DC. January 6. USDA-F SIS. 2002. FSIS sets new procedures for plants that fail Salmonella tests. Food Safety and Inspection Service-US. Department of Agriculture. http://www.fsis.usda.gov/oa/background/newsalmproc.htm. Last accessed October 15, 2007. USDA-FSIS. 2005. Time-temperature tables for cooking ready-to-eat poultry products. US. Department of Agriculture Food Safety Inspection Service. USDA-FSIS. 2006. Compliance Guideline for Controlling Salmonella in Poultry lst ed. US. Department of Agriculture Food Safety and Inspection Service. August 2006. USDA-FSIS. 2007. Progress report on Salmonella testing of raw meat and poultry products, 1998—2006. Food Safety and Inspection Service-US. Department of Agriculture. http://www.fsis.usda.gov/Science/Progress_Report_Salmonel1a_Testing/index.asp . Last accessed October 15, 2007. Veeramuthu GJ, Price J F, Davis CE, Booren AM. 1998. Thermal inactivation of Escherichia coli 0157:H7, Salmonella senftenberg, and enzymes with potential as time-temperature indicators in ground turkey thigh meat. J Food Prot 61 :171-175. 117 Velasquez A, Tuntivanich V, Orta-Ramirez A, Booren AM, Marks ET, Ryser ET. 2005. Enhanced thermal resistance of Salmonella in marinated whole-muscle vs. ground pork. Abstr Inst Food Tech Ann Mtg Abstract No. 89E-14. Warsow CR. 2003. Penetration of Salmonella spp. into whole muscle turkey breast during vacuum marination. Michigan: MS. Thesis Michigan State University. Warsow CR, Orta-Ramirez A, Marks BP, Ryser ET, Booren AM. 2008. Single directional migration of Salmonella into marinated whole muscle turkey breast. J Food Prot 71:153-156. Wesche AM, Marks BP, Ryser ET. 2005. Thermal resistance of heat-, cold-, and starvation-injured Salmonella in irradiated comminuted turkey. J Food Prot 68:942-948. Wierbicki E, Howker JJ, Shults GW. 1976. Effect of salt, phosphates and other curing ingredients on shrinkage of lean pork meat and the quality of smoked processed ham. J Food Sci 41:1116-1121. Wong-Liong J W, Frank JF, Bailey S. 1997. Visualization of eggshell membranes and their interaction with Salmonella Enteritidis using confocal scanning laser microscopy. J Food Prot 60: 1022-1028. Woody J M, Walsh RA, Doores S, Henning WR, Wilson RA, Knabel SJ. 2000. Role of bacterial association and penetration on destruction of Escherichia coli 0157:H7 in beef tissue by high pH. J Food Prot 6323-11. Xiong YL, Kupski DR. 1999. Time-dependent marinade absorption and retention, cooking yield, and palatability of chicken filets marinated in various phosphate solutions. Poult Sci 78:1053-1059. Zheng M, Toledo R, Wicker L. 1999. Effect of phosphate and pectin on quality and shelf- life of marinated chicken breast. J Food Qual 22:553-564. 118 11311111111131131111