- THFQ'?

\ . ' UBRARY
200 7 Michig’ State

Un:\ 1‘ was:“ i.
ruvmauy

 

This is to certify that the
thesis entitled

EFFECT OF BEEF PRODUCT STRUCTURE AND SUBLETHAL
COOKING HISTORY ON SALMONELLA THERMAL
INACTIVATION

presented by

Maria Avelina Mogollon Jijon

has been accepted towards fulfillment
of the requirements for the

MS. degree in Food Science

 

 

 

 

 

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

._.-.c---o-»—-.—---—.—.-.—.-.-—.-.-u—‘—-—-_--~_.—.-.-.--.-—4--.--.—.—

. c~-.-.-.—~—--.-.-.-.--.—.-r—-.—.-.-o--—.-

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5108 K:lProj/Aoc&Pres/ClRC/DateDueindd

EFFECT OF BEEF PRODUCT STRUCTURE AND SUBLETHAL COOKING
HISTORY ON SALMONELLA THERMAL INACTIVATION

By

Maria Avelina Mogollon Jijon

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Food Science and Human Nutrition

2007

ABSTRACT

EFFECT OF BEEF PRODUCT STRUCTURE AND SUBLETHAL COOKING
HISTORY ON SALMONELLA THERMAL INACTIVATION

By
Maria Avelina Mogollén Jijén

The effect of beef product structure and sublethal heating during cooking
on Salmonella thermal resistance has not been reported, and its importance is
not known. Therefore, the objectives of this study were: 1) To determine the
relationship between thermal resistance of Salmonella and degree of grinding,
and 2) To determine the effect of cooking profiles and sublethal heating on
Salmonella thermal inactivation in whole-muscle beef.

All samples were inoculated with a marinade containing an eight-serovar
Salmonella cocktail (~108 CFU/mL). For objective 1, samples (whole-muscle,
coarse-ground, fine-ground, or pureed) were packed into sterile brass tubes,
sealed, and held in a water bath at 60°C for 8-11 durations (at 30 5 intervals). For
objective 2, small, sterile beef steaks were subjected to 6 different cooking
schedules in a moist-air convection oven until 6.5 log reductions in Salmonella
were predicted by traditional log-linear inactivation kinetics applied to real-time
internal temperature data.

Results showed that thermal resistance of Salmonella was higher
(P<0.0001) in whole-muscle than in ground products, however, there were no
differences among the other three product types. Also, longer exposure of
Salmonella to sublethal heating during slow cooking profiles significantly reduced

(P<0.05) subsequent inactivation.

I want to dedicate my thesis to my family, especially to my mom and dad that

have always supported me to achieve all my goals.

ACKNOWLEDGEMENTS

I really want to thank my advisor, Dr. Bradley P. Marks, for giving me the
opportunity of getting a degree as a Master in Science at Michigan State
University (MSU). I also want to thank him for his unconditional guidance and
help during my studies and for his input in all my duties.

I am grateful to Dr. Sanghyup Jeong (Dr. Sang) for all his help during my
research by creating the lethality monitor software and the needle thermocouple,
as well as his everyday aid in the lab with the hardware issues.

Also, thanks to my committee members Drs. Elliot FIyser, Alden Booren,
and Alicia Orta-Fiamirez. Dr. Booren and Dr. Orta-Ramirez, I had a great time
working and learning from you, thank you for your support and friendship
throughout my graduate studies at MSU. Also, special thanks to the Meat Lab
crew and my lab group who were always willing to help or support me.

Finally, thank you Dr. Janice Harte, it was a great experience working and

learning from you throughout my entire career at MSU.

TABLE OF CONTENTS

ABSTRACT ............................................................................................................ i
LIST OF TABLES ................................................................................................ vii
LIST OF FIGURES ............................................................................................... ix
CHAPTER 1. INTRODUCTION ............................................................................. 1
CHAPTER 2. LITERATURE REVIEW ................................................................... 4
2.1 BEEF CATTLE MEAT ................................................................................. 4
2.2 SALMONELLA ............................................................................................. 5
2.3 ILLNESS CAUSED BY SALMONELLA ....................................................... 6
2.4 BEEF PROCESSING .................................................................................. 8
2.4.1 Comminution ................................................................................... 8
2.4.2 Thermal Processing ....................................................................... 10
2.5 THERMAL RESISTANCE OF BACTERIA ................................................. 12
CHAPTER 3. EFFECT OF BEEF PRODUCT PHYSICAL STRUCTURE ON
SALMONELLA THERMAL INACTIVATION ........................................................ 17
3.1 ABSTRACT ............................................................................................... 17
3.2 INTRODUCTION ....................................................................................... 18
3.3 MATERIALS AND METHODS ................................................................... 20
3.3.1 Meat Preparation ................................................................................. 20
3.3.2 Bacterial Cultures ................................................................................ 22
3.3.3 Marinade Preparation .......................................................................... 23
3.3.4 Inoculation ........................................................................................... 23
3.3.5 Thermal Inactivation ............................................................................ 25
3.3.6 Salmonella Enumeration ..................................................................... 26
3.3.7 Data Analysis ...................................................................................... 27
3.4 RESULTS .................................................................................................. 28
3.4.1 Proximate analysis .............................................................................. 28
3.4.2 Raw Meat ............................................................................................ 28
3.4.3 Inoculation of Marinade ....................................................................... 28
3.4.4 Thermal Inactivation Results ............................................................... 28
3.5 DISCUSSION ............................................................................................ 31
3.6 CONCLUSIONS ........................................................................................ 33

CHAPTER 4. THE EFFECT OF COOKING PROFILES AND SUBLETHAL
HISTORY ON SALMONELLA THERMAL INACTIVATION IN WHOLE-MUSCLE

BEEF ................................................................................................................... 34
4.1 ABSTRACT ............................................................................................... 34
4.2 INTRODUCTION ....................................................................................... 35
4.3 MATERIALS AND METHODS ................................................................... 38

4.3.1 Meat Preparation ................................................................................. 38
4.3.2 Bacterial Cultures ................................................................................ 39
4.3.3 Marinade Preparation .......................................................................... 39

4.3.4 Inoculation ........................................................................................... 39

4.3.5 Thermal Inactivation ............................................................................ 39
4.3.6 Salmonella Enumeration ..................................................................... 44
4.3.7 Models ................................................................................................ 44
4.4 RESULTS AND DISCUSSION .................................................................. 48
4.4.1 Raw Meat ............................................................................................ 48
4.4.2 Inoculation of Marinade and Meat ....................................................... 48
4.4.3 Cooking Time at Each Cooking Profile ................................................ 48
4.4.4 Predicted Lethality Curves for Salmonella in Beef Samples for Each
Cooking Schedule ........................................................................................ 52
4.4.5 Predicted Salmonella Log Reductions Using the Stasiewicz and others
(2005 and 2006) model ................................................................................ 56
4.4.6 Salmonella Experimental Log Reductions in Beef Samples for Each
Cooking Schedule ........................................................................................ 57
4.5 CONCLUSIONS ........................................................................................ 62
CHAPTER 5. OVERALL CONCLUSIONS .......................................................... 63
CHAPTER 6. FUTURE RESEARCH ................................................................... 64
APPENDICES ..................................................................................................... 65
APPENDIX A ................................................................................................... 66
COMPARISON OF INOCULATION METHODS OF GROUND PRODUCT
SAMPLES ........................................................................................................ 66
A.1 Fine-Ground .......................................................................................... 66
A2 Coarse-ground ....................................................................................... 71
APPENDIX B ................................................................................................... 73

RAW DATA, k AND D VALUES OF SALMONELLA THERMAL INACTIVATION
ON WHOLE-MUSCLE, COARSE-GROUND, FINE-GROUND, AND PUREED

BEEF PRODUCTS .......................................................................................... 73
APPENDIX C ................................................................................................... 76
OVEN COOKING SCHEDULES TEMPERATURE PROFILES CURVES ....... 76
REFERENCES .................................................................................................... 79

vi

LIST OF TABLES

Table 3.1 First-order inactivation constant, k, and D value (mean :I: S) at 60°C,
calculated by linear regression of Salmonella survivor data for whole-muscle,

coarse-ground, fine-ground, and pureed beef. .................................................... 29

Table 4.1 The temperatures, relative humidities, and durations of each stage (1

to 5) for the six cooking schedules (control and A to E). ..................................... 42

Table 4.2 Predicted Salmonella log reductions using the Stasiewicz and others
(2005 and 2006) model for Salmonella in ground turkey based on temperature

profiles used in this study. ................................................................................... 57

Table 4.3 Salmonella experimental log reductions in beef samples at each

cooking schedule. ............................................................................................... 58

Table A.1 Salmonella counts (mean :I: standard deviation) in 6 random samples
of fine-ground pork, which were prepared by passing the pork once through the

grinder (plate with 6 mm diameter holes). ........................................................... 67

Table A2 Salmonella counts (mean :I: standard deviation) in 6 random samples
of fine-ground pork, which were prepared by passing the pork twice through the

grinder (plate with 6 mm diameter holes). ........................................................... 68

Table A.3 Salmonella counts (mean :I: standard deviation) in 6 random samples
of fine-ground pork, which were prepared by subjecting the pork loin to vacuum
(40 cmHg) marination for 20 min and then passing the pork loin once through the

grinder (plate with 6 mm diameter holes). ........................................................... 69

vii

Table A4 ANOVA: single factor, comparing Salmonella counts of Method l,

Method II, and Method III of fine-ground product inoculation. ............................. 70

Table A5 Salmonella counts (mean :I: standard deviation) in 5 random samples
of coarse-ground pork, which were prepared by passing the pork twice through

the grinder (plate with 16 mm diameter holes). ................................................... 72

Table 8.1 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of whole-muscle beef ................................................ 73

Table 8.2 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of coarse-ground beef ............................................... 73

Table 8.3 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of fine-ground beef. .................................................. 74

Table 8.4 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of pureed beef. ......................................................... 74

Table 8.5 First-order inactivation constant, k, and D values at 60°C, calculated
by linear regression of Salmonella survival data for whole-muscle, coarse-

ground, fine-ground, and pureed beef. ................................................................ 75

viii

LIST OF FIGURES
Figure 2.1 Grinder plates, with 6 mm and 16 mm diameter holes, respectively. ..9
Figure 3.1 Methods flowchart. ............................................................................ 20
Figure 3.2 Beef coring and packing into sterile brass tubes. ............................. 21

Figure 3.3 Grinding (a) and packing (b) ground beef samples into sterile brass

tubes. .................................................................................................................. 22
Figure 3.4 Blended beef puree. .......................................................................... 22

Figure 3.5 Whole-muscle beef cores inoculated by immersion in marinade ....... 25

Figure 3.6 Samples heated isotherrnally in water bath at 605°C. ...................... 26
Figure 3.7 Salmonella survival curves in whole-muscle, coarse-ground, fine-
ground, and pureed beef samples heated at 60°C. ............................................. 29
Figure 4.1 Raw beef steak sample. .................................................................... 38

Figure 4.2 Moist-air oven components arrangement (showing airflow direction).

............................................................................................................................ 41
Figure 4.3 Screen view of the Real-Time Lethality Monitor (v2). ........................ 43
Figure 4.4 Temperature curves of samples in cooking schedule A. ................... 49
Figure 4.5 Temperature curves of samples in cooking schedule 8. ................... 50
Figure 4.6 Temperature curves of samples in cooking schedule C. ................... 50
Figure 4.7 Temperature curves of samples in cooking schedule D. ................... 51
Figure 4.8 Temperature curves of samples in cooking schedule E. ................... 51

Figure 4.9 Temperature curves of samples in the control cooking schedule. ..... 52

Figure 4.10 Lethality curves for Salmonella in samples cooked using schedule A.

............................................................................................................................ 53

Figure 4.11 Lethality curves for Salmonella in samples cooked using schedule 8.

............................................................................................................................ 53

Figure 4.12 Lethality curves for Salmonella in samples cooked using schedule C.

............................................................................................................................ 54

Figure 4.13 Lethality curves for Salmonella in samples cooked using schedule D.

............................................................................................................................ 54

Figure 4.14 Lethality curves for Salmonella in samples cooked using schedule E.

............................................................................................................................ 55

Figure 4.15 Lethality curves for Salmonella in samples cooked using the control

schedule .............................................................................................................. 55

Figure 4.16 Comparison of predicted Salmonella log reductions from the

traditional, Weibull, and Stasiewicz models and experimental values. ................ 59

Figure 4.17 Comparison of the predicted versus observed relative effect of prior

cooking history on subsequent process lethality. ................................................ 61
Figure 0.1 Temperature curves of convection oven air in cooking schedule A. .76
Figure 0.2 Temperature curves of convection oven air in cooking schedule 8. .76
Figure C.3 Temperature curves of convection oven air in cooking schedule C. .77

Figure 0.4 Temperature curves of convection oven air in cooking schedule D. .77

Figure 0.5 Temperature curves of convection oven air in cooking schedule E. .78

Figure 0.6 Temperature curves of convection oven air in cooking schedule F

(control). .............................................................................................................. 78

xi

CHAPTER 1. INTRODUCTION

The meat and poultry industry accounts for 51% of United States (US)
agriculture (USDA (United States Department of Agriculture) 2006), and it is also
the top taming export ([AMI] American Meat Institute 2003). Annual sales in
2000 exceeded $100 billion ([AMI] American Meat Institute 2003). Also, the US.
is the largest beef producer worldwide (USDA (United States Department of
Agriculture) 2006; 2006), with an estimated production of 25.6 billion pounds in
2005 ([AMI] American Meat Institute 2006). With an average annual consumption
of 67 pounds per capita, beef is the most popular red meat consumed in the US.
(Davis and Lin 2005). Beef is produced by more than a 1000 facilities throughout
the continental US, which are regulated by the United States Department of
Agriculture (USDA) to ensure that the product is safe for consumption and
labeled properly ([AMI] American Meat Institute 2002).

Beef products contaminated with microbial pathogens are an economic
liability for both the consumer and the producer. It is a loss to the industry, due to
possible recalls and adverse publicity. The major cause of meat and poultry
product recalls is microbial contamination (76%) ([FSIS] United States Food
Safety and Inspection Service 2003). During 2000, the meat and poultry industry
had 78 product recalls, of which 26 were beef products ([FSIS] United States
Food Safety and Inspection Service 2003). Recalls cost a lot, because
companies have to pay for lost product, as well as laboratory analysis, and

relocation and transportation of the product.

It has been estimated that there are 1.4 million cases of salmonellosis per
year in the United States, most of which are foodbome. Foodbome diseases in
America cost billions of dollars annually (Buzby and others 1996). The economic
cost of Salmonella range from $0.9 billion to 3.7 billion when considering medical
costs, lost productivity, and most important, premature death (Frenzen and
others 1999). Salmonella is a common pathogen in meat and poultry products,
and therefore a real concern to the industry.

In 1996, a new system to inspect meat and poultry plants was developed
in the US; there are four main elements of this new system: 1) Meat and poultry
establishments must have a Hazard Analysis Critical Control Point (HACCP)
plan; 2) Federally inspected meat plants must have written sanitation standard
operating procedures; 3) To verify the process lethality for Salmonella, the Food
Safety and Inspection Service (FSIS) will test for the organism in meat and
poultry products; 4) Slaughter plants will test for generic Escherichia coli on
carcasses (Crutchfield 1999).

The USDA requires a 6.5 log reduction in Salmonella spp. for cooked
beef, ready-to-eat (RTE) roast beef, and cooked corned beef products.
Salmonella was chosen as the reference organism due to its presence in raw
poultry, beef, and pork, and also due to its importance in foodbome illness. Any
RTE meat product containing the reference organism will be considered
adulterated and needs to be recalled ([FSIS] United States Food Safety and

Inspection Service 2001 ).

Salmonellosis outbreaks associated with beef continue to happen, even
after the implementation of HACCP, good manufacturing practices, and
implementation of the new meat and poultry regulations in 1996. Improving beef
production practices and tools might help prevent salmonellosis.

Previous studies by Orta-Ramirez and others (2005) concluded that there
is a significant difference in Salmonella thermal resistance in whole-muscle and
ground beef, which led to the hypothesis that the degree of grinding might have
an effect on thermal inactivation of Salmonella and therefore might need to be
considered in thermal processing of beef products.

Consequently, with the goal of achieving food safety, the main objectives

of this study were:

1. To determine the effect of beef product structure on Salmonella thermal
inactivation.
2. To quantify the effect of sublethal thermal history on thermal inactivation of

Salmonella in whole-muscle beef subjected to various cooking schedules.
After a general literature review (Chapter 2), Chapters 3 and 4 of this
thesis are presented as stand-alone papers addressing objectives 1 and 2,

respectively.

CHAPTER 2. LITERATURE REVIEW

This chapter contains background information arranged as follows: first,
the basic characteristics of beef and its relationship to Salmonella are given.
Second, Salmonella and its characteristics as a pathogen are discussed in the
context of its importance in the beef industry. Third, meat processing procedures
and technologies are covered, with a focus on grinding and thermal processing of
beef. Thermal processing is the primary focus of this chapter, due to its
importance in microbial inactivation and food safety in the beef industry, which is

the main focus of this research.

2.1 BEEF CATTLE MEAT

Meat is animal tissue that is used as food; it includes adipose tissue,
epithelial tissue, connective tissue, nervous tissue, and its main component,
muscle. The largest category of meat, based on its consumption, is red meat, of
which the most common are: beef, pork, lamb or mutton, and veal (Aberle and
others 2001). Beef comes from cattle, usually slaughtered at 1 to 3.5 years of
age (for Prime, Choice, Select, and Standard beef grades). All grades of lean
retail raw beef contain water (68.4 - 73.0%), protein (19.0 - 22.8%), total lipids
(3.3 - 10.2%), and ash (0.9 — 1.5%) (Anderson and Hoke 1990).

The composition of beef muscle makes it a good environment for microbial
growth, due to its water content and water activity (0.99), and available nutrients

used by microorganisms (Doyle and others 1997). Bacteria that are commonly

present on beef include: Clostridium perfringens, Salmonella, Listeria
monocytogenes, and Escherichia coli (Aberle and others 2001).

Salmonella outbreaks associated with beef are still occurring. This
pathogen can be introduced to the carcass surface by cross contamination
during hide removal and evisceration, or by contact with surfaces and workers’
hands (Doyle and others 1997). For example, a 1994 USDA survey
demonstrated that on average 1% of beef carcasses in the US. were
contaminated with Salmonella ([FSIS] United States Food Safety and Inspection
Service and others 1994). Sofos and others (1999) examined 3,780 samples of
US beef carcasses and found that Salmonella was present in 2.1% to 15.5% of
the samples (including cows and bulls). In 2001, the prevalence of Salmonella in
raw ground beef was 2.8% ([FSIS] United States Food Safety and Inspection
Service 1998-2001). Therefore, understanding Salmonella epidemiology and
identifying pathogen reduction methods is necessary to improve existing safety

strategies for the beef industry.

2.2 SALMONELLA

Salmonellae are facultative anaerobic Gram-negative, rod-shaped
bacteria (0.7- 1.5 x 2.0- 5.0 pm) belonging to the family Enterobacteriaceae (Bell
and Kyriakides 2002). The World Health Organization states that there are two
species of Salmonella, enterica and bongori, each with multiple serovars

(Montville and Matthews 2005). The most common sources of Salmonella are the

gastrointestinal tract of humans or domestic and wild animals, including birds and
rodents (Bell and Kyriakides 2002).

Most Salmonella organisms are motile via peritrichous flagella, but some
are non-flagellated (Montville and Matthews 2005). This organism is
chemoorganotrophic, which means that it can utilize different organic substrates
to metabolize nutrients by fermentation or respiratory pathways. Its optimal
growth temperature is 37°C, but some behave as psychrotrophic bacteria, which
can grow between 2 and 4°C, while others grow at temperatures as high as 54°C
(some serovar Typhimurium) (Montville and Matthews 2005). In general,
sublethal temperature is considered to be around 50°C (Doyle and others 1997;
Bell and Kyriakides 2002). Salmonella heat resistance is affected by prior thermal
history, water activity (aw), substrate environment and composition, and pH (Bell
and Kyriakides 2002; Montville and Matthews 2005). The organism grows
between 4.5 and 9.5 pH, with an optimum of 6.5 to 7.5. Salmonella does not
grow in foods with water activity lower than 0.93 or those containing 3 to 4% salt

(NaCl) (Montville and Matthews 2005).

2.3 ILLNESS CAUSED BY SALMONELLA

Salmonella causes illnesses such as enteric (typhoid) fever,
salmonellosis, systemic infections (aseptic reactive arthritis, Reiter’s syndrome,
and ankylosing spondylitis) (Doyle and others 1997; Montville and Matthews
2005), and may even result in complications (endocarditis, meningitis, and

pneumonia) (Buzby and others 1996). Most illnesses provoked by Salmonella

are associated with consumption of food, mainly of animal origin (Troller 1986;
Tauxe 1991). Salmonella Typhimurium, S. Enteritidis, and S. Newport are the
most common Salmonella serovars implicated in US. clinical cases (Buzby and
others 1996), S. Typhimurium being the most common in beef products (Merck
1992).

Typhoid fever is a bacterial infection that is associated with the typhoid
and paratyphoid strains of Salmonella. Only a few cells of these organisms might
be necessary to cause infection; incubation usually takes between 7 to 28 days.
Symptoms include diarrhea, fever, abdominal pain, headache and prostration;
most of the time it is treated with drugs (Doyle and others 1997; Montville and
Matthews 2005).

Salmonellosis is an acute gastroenteritis, usually appearing 8 to 72 h after
ingesting Salmonella. The infectious dose can be as low as one cell, but usually
102 to 103 are required (Buzby and others 1996); its severity depends on the
serotype, quantity, and immune status of the host. lmmunocompromised
individuals, infants, and the elderly are most vulnerable. Symptoms caused by
salmonellosis can include the following: abdominal discomfort, diarrhea,
dehydration, fever, headache, nausea, stomachache, and vomiting (Buzby and
others 1996).

Salmonella infection depends on availability or competence to find
attachment sites on the intestinal wall, and on the ability of the organism to attach
and invade the intestinal cells (Montville and Matthews 2005). When the

pathogen finds an attachment site and contacts the epithelial cells, it produces

proteinaceous appendages on its surface, facilitating attachment. After
attachment to the host cells, signaling between the host cell and the
microorganism results in invasion of the intestinal cells, provoking diarrhea
(Doyle and others 1997; Montville and Matthews 2005).

Diarrheagenic enterotoxin and therrnolabile cytotoxin are two important
Salmonella virulence factors (Montville and Matthews 2005). The former is
released in the cytoplasm of the infected host cell, causing diarrhea. The latter
promotes the migration of Salmonella into other tissues (Doyle and others 1997;
Montville and Matthews 2005).

The economic and health damage attributed to Salmonella can be avoided
by ensuring the safety of our food supply. One important way to address this
problem is by conceiving a safer beef industry, which can be achieved by proper

processing of its products, using the best technology and tools.

2.4 BEEF PROCESSING

A main objective of meat processing is preservation of the product by
inhibiting pathogens and spoilage microorganisms (Pearson and Tauber 1984).
There are multiple ways to process meat to achieve different food products; two
well known processes are comminution and thermal processing.

2.4.1 Comminution

Comminuted meat products originated from the idea of making a more

desirable and attractive product than its components, including less palatable

cuts of meat, meat trimmings, and fat from other cuts (Ranken 2000).

Comminution is the process of reducing the particle size of raw meat materials;
its objective is to obtain a uniform composition and particle size of the product,
and to increase the meat tenderness (Aberle and others 2001).

A disadvantage of comminuting meat is that the microorganisms that were
present on its surface are then distributed in the increased surface area of the
comminuted product. Also, the beef particle size reduction process ruptures the
cell tissues and releases water, fluids, and nutrients, which can potentially
enhance microbial growth (Doyle and others 1997).

Common machinery used in the meat industry for communition includes
meat grinders, bowl choppers, emulsion mills, knifes, dicers/ strip cutters, and
flakers (Ranken 2000; Aberle and others 2001). Grinding is the meat process of
forming uniform cylinders of fat and lean. The grinder components are: 1) pusher,
2) filler tray, 3) barrel, 4) screw feed, 5) blade, 6) plate (Fig. 2.1), and 7) nut. The
purpose of the screw feed is to convey the meat through the barrel and to press it
through the wholes of the plate. The blade cuts the pressed meat and aids in
filling the plate holes. The thickness and diameter of the plate holes determines

the size of the ground product (Pearson and Tauber 1984).

 

 

,:,:Z:.:. "1 1
9': P :‘Q N '3 j
1”“ 7
6 mm 16 mm
PLATE PLATE

 

Figure 2.1 Grinder plates, with 6 mm and 16 mm diameter holes, respectively.

 

2.4.2 Thermal Processing

The following section includes background information on thermal
processing (i.e., cooking). Thermal processing is a physical process that involves
heat transfer, which is described later in more detail. Also, this section contains
information on pasteurization and commercial sterilization, which are the two
types of thermal processing (different due to the temperatures used). Next, an
overview of the technologies used for these two types of beef processing are
covered. And finally, the importance of thermal processing in the beef industry is
addressed.

Thermal processing is the process in which beef is exposed to high
temperatures. Cooking occurs due to the physical phenomenon of heat transfer.
Heat transfer can occur as conduction, convection, and radiation, or as a
combination of modes. Conduction is molecular level heat transfer within a mass;
convection is heat transfer between a fluid and a mass; and radiation is the heat
exchange through space by electromagnetic radiation (Singh and Heldman
1998). The heat transfer properties of meat vary with the composition and
structure of the product.

The main reasons why meat is cooked are to: 1) inhibit spoilage and
pathogenic organisms; 2) inactivate endogenous enzymes that cause
deterioration; 3) achieve an appealing color, flavor, and tenderness; 4) create a
variety of different products; and 5) create convenient products for customers

(Pearson and Tauber 1984; Ranken 2000; Aberle and others 2001).

10

Thermal processing in the beef industry is generally achieved using either
dry heat or moist heat. There are two types of heating in thermal processing of
meats: moderate (pasteurization) and extreme heating (commercial sterilization).
Pasteurization is used for the majority of ready-to-eat meat products. The product
endpoint temperature in this procedure is between 58 to 75°C, in order to
inactivate most vegetative microorganisms, but not spores.

Heating and/or cooking of beef has been accomplished by the use of
different pasteurization methods like smoking, hot air drying, and steam or gas
heating (in an oven or smokehouse). Hot air drying is usually used to dehydrate
the product; this procedure is typically used in beef jerky processing (Aberle and
others 2001).

Meat products can be smoked and cooked, or only smoked or only
cooked. Natural smoking is the process of exposing the product to wood smoke,
usually done in a smoke house (Aberle and others 2001). Three types of
smokehouses are recognized: 1) natural air circulation, 2) air conditioned or
forced air, and 3) continuous. Cooking is usually combined with the smoking step
by the use of steam or gas heat (Pearson and Tauber 1984). Usually, cooking is
conducted at high humidity environments, to enhance heat transfer.

In contrast to pasteurization/cooking, the outcome of a commercial
sterilization process is a shelf stable product, which, after being subjected to
temperatures typically around 121°C, contains nominally no spores or vegetative
cells (i.e,. probability < 1012); however, texture and palatability are compromised

(Aberle and others 2001). A well known sterilization process used in the industry

11

is retorting. It is used when the beef is canned and cooked to sterility; the
products are subjected to high pressure and temperatures of 121°C or higher.
Cooking in the beef industry is a significant processing step, due to its
effect on product safety. It is important to consider the factors that affect thermal
processing, such as the product and its composition, temperature, and
temperature differential (AT), time exposure to heat, and environmental
conditions (i.e., humidity). In addition, when creating a beef cooking schedule, it
is important to consider all the factors that will affect product quality, but
especially safety, because there are aspects that affect pathogen thermal

resistance as well.

2.5 THERMAL RESISTANCE OF BACTERIA

The heat resistance of bacteria is most often reported as a D value and 2
value. The D value is the time required to reduce the initial count of viable cells or
spores by 90% at a given temperature (Bell and others 2005). The 2 value is the
necessary increase in the temperature needed to reduce the D value of a
microorganism 10-fold (Bell and others 2005). These values are unique for each
organism and substrate (Aberle and others 2001; Montville and Matthews 2005).
Thermal resistance of bacteria can be influenced by various intrinsic and extrinsic
factors.

The reported thermal resistance of a bacterial pathogen depends on its
previous thermal history (Aberle and others 2001). It also depends on growth

temperature, stage of growth, strain, heat shock, and detection or enumeration

12

method or media (Tomlins and Ordal 1976; Doyle and Mazzotta 2000).
Quintavalla and others (2001) found that the Dseoc values for 88 different strains
of Salmonella in Tryptone Soy Broth (TSB) were not identical; the average 053°C
value was 1.34 min, with a range of 0.79 to 2.67 min. Also, bacteria are more
readily inactivated by heat during the exponential as opposed to stationary
growth phase (Humphrey and others 1995).

Additionally, sublethal heating can significantly increase Salmonella
thermal resistance. Heat shock occurs when the organism is heated above its
maximum growth temperature (Mackey and Derrick 1986; Bunning and others
1990). Wesche and others (2005) reported Deooc values of a heat shocked, eight-
strain Salmonella cocktail (heat shocked for 30 min at 54°C) in ground turkey.
Thermal resistance of the heat-shocked Salmonella was about 56.1% and
105.9% greater than unshocked Salmonella when plated on Trypticase soy agar
(nonselective media) containing 0.6% yeast extract and xylose lysine
desoxycholate agar (selective media), respectively.

Food composition also affects thermal resistance of bacteria (Murphy and
others 2000). Salmonella thermal resistance is affected by the food’s intrinsic
factors such as total solids content (Blackburn and others 1997; Doyle and
Mazzotta 2000), pH (Blackburn and others 1997; Doyle and Mazzotta 2000),
water activity (Doyle and Mazzotta 2000; Carlson and others 2005), fat content
(Smith and others 2001), some food additives (Doyle and Mazzotta 2000), and
redox potential (Lihono and others 2001). Also, survival of Salmonella in meat

depends on the meat species and method of enumeration (Quintavalla and

13

others 2001). Quintavalla and others (2001) showed that eight strains of
Salmonella exhibited greater thermal resistance in pork compared to culture
media.

Previous research has shown that the D values for bacteria in meat and its
products increase with fat content (Hansen and Riemann 1963; Stumbo 1973;
Ahmed and others 1995; Juneja and others 2001; Smith and others 2001). Smith
and others (2001) reported that the 050°C values for S. Typhimurium DT104 in
ground beef increased more than two-fold when the fat content increased from
4.8 to 19%. This same pattern also was observed for two E. coli strains in
morcilla (a blood sausage) which had different fat levels. When the fat in morcilla
increased 20%, the 054°C value for E. coli G335 increased from 4.9 to 5.4 min.
Also, when the fat in morcilla was increased in the same proportion (20%) the
054°C value for E. coli O157:H7 increased from 5.5 to 6.4 min (Oteiza and others
2003)

Other studies showed that the thermal resistance of Salmonella is
enhanced at a lower water activity environment (Goepfert and others 1970;
O'Donovan- Vaughan and Upton 1999). Carlson (2005) actually verified that
thermal resistance of Salmonella increased 64% as aw decreased from 0.99 to
0.95 in ground turkey samples.

Also, bacterial resistance to thermal processing is affected by the meat’s
pH. One study demonstrated that E. coli O157:H7 was inactivated faster as the
pH of the acidified beef slurry decreased from 5.98 to 4.70 (Abdul-Raouf and

others 1993).

14

Another factor that has an effect on thermal resistance of Salmonella,
which has been reported only by Orta—Ramirez and others (2005), is the beef
physical structure. They compared whole-muscle beef and ground beef at 55, 60,
and 625°C. The only difference between the whole-muscle samples and ground
samples was the physical structure. The results revealed that an eight-strain
Salmonella cocktail had an enhanced thermal resistance, more than doubled, in
whole-muscle compared to ground beef, at all three temperatures (Orta-Ramirez
and others 2005). That study accounted only for the difference between a whole-
muscle and ground product, without addressing the range of potential product
structures (or explaining the underlying cause of the observed differences).

The cooking procedures in the meat industry are usually identified as
critical control points in HACCP programs (Aberle and others 2001). Due to the
importance of complying with the regulations, the cooking schedule must be
validated to prove that it achieves a 6.5 log reduction of Salmonella in beef
products ([FSIS] United States Food Safety and Inspection Service 2001).
Usually, the cooking schedule is validated by the use of thermal inactivation
models, due to the inability to test pathogen-inoculated products in commercial
facilities. A widely used modeling tool, the American Meat Institute Process
Lethality Spreadsheet (AMI-PLS), is not very precise, because it does not include
all the variables that affect the thermal inactivation of Salmonella, such as
moisture content, aw, fat content, and pH of the product, as well as sublethal
heating, and physical structure of the meat. Therefore, new models need to be

created, that include these other important factors. This will enable the beef

15

industry to generate more reliable validations of thermal processes, to both

ensure product safety and comply with the USDA regulations.

16

CHAPTER 3. EFFECT OF BEEF PRODUCT PHYSICAL STRUCTURE
ON SALMONELLA THERMAL INACTIVATION

3.1 ABSTRACT

Numerous studies have assessed thermal inactivation of Salmonella in
beef. However, the impact of muscle structure has been considered only
recently, with several studies reporting enhanced thermal resistance in whole-
muscle as compared to ground meat. The functional relationship between meat
product structure and Salmonella thermal resistance has not been reported, and
it is not known whether thermal resistance decreases with the degree of grinding.
Therefore, the objective of this study was to determine the relationship between
thermal resistance of Salmonella and degree of grinding (whole-muscle, coarse-
ground, fine-ground, and beef puree). Each of the four product types was
irradiated to sterility and inoculated with a marinade containing an eight-serovar
Salmonella cocktail (108 CFU/mL). Samples (5.00 :I: 0.05 g) were packed into
sterile brass tubes (12.7 mm diameter), sealed, and held in a water bath at 60°C
for 8-11 durations (at,30 5 intervals). lntemal sample temperature was monitored
with a thermocouple. Samples were then serially diluted and plated on
PetrifilmTM aerobic count plates to enumerate surviving salmonellae. All samples
had the same composition, thermal history, and initial Salmonella counts;
therefore, differences in thermal resistance were due entirely to the degree of
grinding (i.e., product structure). Overall, thermal resistance of Salmonella was.
higher (P<0.0001) in whole-muscle than in the other three products, but there

were no differences among the other three products.

17

3.2 INTRODUCTION

Whole-muscle meat is less likely than ground beef to contain bacterial
pathogens, with more meat-related outbreaks traced back to the latter (Doyle and
Mazzotta 2000). Although the interior of intact, whole—muscle products has long
been assumed to be sterile (Elmossalami and Wassef 1971), contamination and
survival of pathogens have been reported (Elmossalami and Wassef 1971; Gill
and Penney 1982; Jay 2000). The few studies that have been performed on
migration of Salmonella into whole-muscle products have found that the
pathogen does migrate and survive inside the muscle (Elmossalami and Wassef
1971 ; Warsow 2003; Velasquez 2006).

After the inner muscle is contaminated, thermal inactivation rate for
Salmonella varies based on strain, product, and environmental factors (Doyle
and Mazzotta 2000). Salmonella thermal inactivation is affected by total solids
content (Blackburn and others 1997; Doyle and Mazzotta 2000), pH (Blackburn
and others 1997; Doyle and Mazzotta 2000), water activity (Doyle and Mazzotta
2000; Carlson and others 2005), fat content (Smith and others 2001), some food
additives (Doyle and Mazzotta 2000), and redox potential (Lihono and others
2001). Also, survival of Salmonella in meats depends on meat species (Ghazala
and others 1995; Veeramuthu and others 1998). The pathogen location or
attachment site also plays a role; when attached to meat, Salmonella exhibits
greater thermal resistance than when unattached or suspended in liquid media
(Thomson and others 1979; Humphrey and others 1995; Doyle and Mazzotta

2000).

18

For all fully-cooked, ready-to-eat meat products, Salmonella lethality is the
basis for the USDA-FSIS performance standards, which require a 6.5-log
reduction in Salmonella spp. for cooked beef products ([FSIS] United States
Food Safety and Inspection Service 2001). However, commonly used tools to
calculate process lethality do not account for all factors that affect pathogen
survival (Guo and Marks 2005). One factor that has not been included in the
existing models for calculating process lethality is the physical structure of the
meat product.

Previous studies have demonstrated that Salmonella has a higher thermal
resistance in whole-muscle beef and pork than in equivalent ground products
(Orta-Ramirez and others 2005; Velasquez 2006). However, those studies tested
only whole-muscle and fine-ground products, and did not explain the cause of the
observed effects or model the relationship between product structure and
Salmonella inactivation. Therefore, the objective of this study was to evaluate the
relationship between thermal resistance of Salmonella and degree of grinding

(whole-muscle, coarse-ground, fine-ground, and beef puree).

19

3.3 MATERIALS AND METHODS

The overall experimental design for this study consisted of preparing and
inoculating four different types of product (whole-muscle, coarse-ground, fine-
ground, and pureed beef) from the same original lot of beef (Fig. 3.1). The

details of each step are described in the following sections.

WHOLE-MUSCLE BEEF

 

 

CO ES ‘
GRINDING (16 mm PLATE) GRINDING (6 mm PLATE) GRINDING (5 mm PLATE) X5

MARINATIONI mocuumou (20 MIN)

INOCULATION INOCULATION INOCULATION
WHOLE-MUSCLE
SAMPLES GRINDING (16 mm PLATE) GRINDING (6 mm PLATE) BLENDED (1 MIN)
COARSE-GROUND FINE-GROUND PUREED
SAMPLES SAMPLES SAMPLES

Figure 3.1 Methods flowchart.

3.3.1 Meat Preparation

Beef chuck, shoulder clods (1140, according to (North American Meat
Processors Association 1997)) were obtained from Packerland-Plainwell, Inc.
(Plainwell, Mich.) 48 h after slaughter. Cubic pieces containing the triceps brachii
long head were removed from the shoulder clods, vacuum-packaged at the
Michigan State University (MSU) meat laboratory, and frozen and held at
-29°C. All samples were transported on dry ice to CFC Logistics in Quakertown,
Pa., where they were irradiated (~10 kGy) to eliminate background flora. Sterility
of the beef was tested by diluting the samples (1:5) in peptone water (Difco,

Becton, Dickinson and Company, Sparks, Md., USA.) and plating on PetrifilmTM

20

aerobic count plates (3M, St. Paul, Minn., USA). All samples were prepared
immediately before starting the experiment.

Moisture and fat contents were determined in quadruplicate from four
different beef shoulder clods using AOAC methods 950.468 and 991.36,
respectively (Association of Official Analytical Chemists 1996).
3.3.1.1 Whole-muscle

Whole-muscle samples were obtained from the sterile beef muscle by
using a sterile coring device (1.27 cm diameter, G.R. Electrical Mfg. Co.,
Manhattan, Kans., USA.) to obtain whole-muscle cylinders (5.00 :I: 0.05 g) 6 to 8

cm in length (Fig 3.2).

 

 

Figure 3.2 Beef coring and packing into sterile brass tubes.

3.3.1.2 Coarse— round and fine- round beef

 

Irradiated beef was ground in a sterile grinder (Model 5126, Toledo
Chopper, Toledo, Ohio, USA), by passing the beef sample twice through a
plate with 16 or 6 mm holes to obtain coarse-ground and fine-ground samples,

respectively (Fig. 3.1 and Fig 3.3).

21

 

(b)
Figure 3.3 Grinding (a) and packing (b) ground beef samples into sterile brass

tubes.

3.3.1.3 Beef puree

Irradiated beef was comminuted in a sterile grinder, passing five times
through a plate with 6 mm diameter holes (as described in the previous section),
and then blended at puree speed for 1 min in an Osterizer blender (Model 6641,

Oster ®, Mexico) (Fig 3.4).

 

Figure 3.4 Blended beef puree.

3.3.2 Bacterial Cultures
As in previous Salmonella thermal inactivation studies by Orta-Ramirez

and others (2005), 8 Salmonella serovars of moderate to high thermal resistance

22

(Juneja and others 2001) were obtained from Dr. V.K. Juneja (USDA-ARS,
Eastern Regional Research Center, Wyndmoor, Pa.). S. Montevideo FSIS 051
(beef isolate); S. Thompson FSIS 120 (chicken isolate); S. Enteriditis H3527
(chicken isolates phage type 13A); S. Enteriditis H3502 (chicken isolates phage
type 4); S. Hadar MF60404 (turkey isolate); S. Copenhagen 8457 (pork isolate),
S. Typhimurium DT104 H3380 (human isolate); and S. Heidelberg F50388G1
(human isolate).

Each strain was stored at -80°C in tryptic soy broth (TSB) (Difco, Becton
Dickinson, Sparks, Md., U.S.A.) containing 20% glycerol. In preparation for use,
the eight serovars were each subjected to two 24 h transfers in T88 at 37°C.
3.3.3 Marinade Preparation

A generic marinade formula was prepared to contain 96.0% filtered and
deionized water, 3.2% NaCl (J.T. Baker, Philipsburg, N.J., USA), and 0.8%
potassium phosphate solution (Butcher and Packer Supply 00., Detroit, Mich.,
USA). After combining and totally dispersing the components, the marinade
was poured into glass bottles (520 mL each), autoclaved to sterility (121°C for 20
min), and stored at room temperature.

The moisture content of each marinated product (whole-muscle, coarse-
ground, fine-ground, and beef puree) was also determined in duplicate after
marination using AOAC method 950.468.

3.3.4 Inoculation
Immediately before inoculating the beef samples, the Salmonella cocktail

was prepared by combining equal volumes (9 mL) of each serovar in a centrifuge

23

bottle. After centrifugation (6000 x g, 20 min at 4°C), the pellet was resuspended
in 520 mL of the marinade to yield ~10° CFU/mL.

The concentration of Salmonella in the marinade was verified by serial
dilution in 0.1% peptone water, which was plated in duplicate on PetrifilmTM
aerobic count plates.

Several methods for inoculating the four different products were evaluated
to determine the effects on initial counts and subsequent thermal resistance (see
Appendix A). The methods used in this study are described below for each of the

four products.

24

3.3.4.1 Whole-muscle
Whole-muscle cylinders were immersed in the inoculated marinade for 20
min, which resulted in a marinade uptake of approximately 0.15 g marinade/g

beef (Fig. 3.5).

 

Figure 3.5 Whole-muscle beef cores inoculated by immersion in marinade.

3.3.4.2 Coarse-ground and fine-ground beef

Ground samples were inoculated dropwise in the same proportion as the
uptake of the whole-muscle samples (0.15 g marinade/g beef) and then passed
through the grinder a second time.
3.3.4.3 Beef puree

The samples to become puree were inoculated dropwise (0.15 g
marinade/g beef) after grinding and before the blending step.
3.3.5 Thermal Inactivation

Whole-muscle, ground, and pureed beef samples were aseptically packed
into sterile brass tubes (12.7 mm diameter, 10 cm long), which were sealed with

sterile rubber stoppers wrapped with Teflon® tape.

25

Samples were then heated isothermally in an agitated, 605°C water bath
(NESLAB Instruments Inc., Newington, NH, USA.) with the internal
temperature monitored using a thermocouple probe (Type T, 1.0 mm, Omega
Engineering, Stamford, Conn., U.S.A.) inserted in the center of the sample (Fig
3.6).

The first tube was removed from the water bath when the lag time was
reached. The lag time was the time when the sample reached the target
temperature (60°C). After the first sample was removed, a brass tube was
removed every 30 s and placed in an ice bath. The whole-muscle sample
treatments continued for 5 min, and the coarse-ground, fine-ground, and pureed
sample treatments for 4.5, 3.5, and 5 min, respectively. Whole-muscle, coarse-
ground, and fine-ground samples were each tested in duplicate, and the beef

puree samples were tested in triplicate.

 

Figure 3.6 Samples heated isothermally in water bath at 605°C.

3.3.6 Salmonella Enumeration
The cooled samples were diluted (1:10) in 0.1% peptone water,

homogenized for 180 s in a masticator (Model 0410, IUL Instruments USA, Inc.

26

Cincinnati, Ohio, USA), serially diluted in peptone water, and plated on
PetrifilmTM aerobic count plates (48 h, 37°C). All work was accomplished under
aseptic conditions and using sterile instruments.
3.3.7 Data Analysis

Salmonella survivor curves at 60°C were determined for all samples by
plotting the logarithm of the survival ratio vs. time. The k values were calculated
as the slope from linear regression of Ln (N/No) vs. t for each replicate test series
for a given product, where N/No was the survival ratio, and t was the time (min).
The D values for each sample were then calculated as 2.303/k. The k values for

each product type were compared using Tukey—Kramer’s test (or=0.05).

27

3.4 RESULTS

3.4.1 Proximate analysis

The raw beef contained 73.0 1 0.93% moisture and 4.5 :I: 0.07% fat. The
whole-muscle, coarse-ground, fine-ground, and pureed samples had 75.93 :I:
0.99, 71.28 :t 0.98, 73.85 a: 0.99, 75.01 :r: 0.01% moisture content, respectively.
3.4.2 Raw Meat

The uninoculated, irradiated beef samples contained no detectable
background microflora as determined from plating 1:5 dilutions on PetrifilmTM
aerobic count plates.
3.4.3 Inoculation of Marinade

The marinade was inoculated to obtain ~108 Salmonella CFU/mL. Before
each experiment, the inoculated marinade was plated to determine the
Salmonella concentrations, which were 8.7 :I: 0.2 log CFU/mL.
3.4.4 Thermal Inactivation Results

The heating lag times for whole-muscle, coarse-ground, fine-ground, and
pureed beef ranged from 2.70 to 2.90, 1.77 to 2.33, 1.90 to 2.53, and 2.48 to
2.58 min, respectively. There were no statistical differences (oc=0.05) between
the lag times of the different products.

After the lag time, Salmonella counts for the to samples of whole-muscle,
coarse-ground, fine-ground, and pureed beef were 6.9 :I: 0.2, 7.6 :1: 0.2, 7.3 :1: 0.0,
and 7.4 :t 0.3 Log CFU/g, respectively. There were no differences (or=0.05)

between the initial counts of the inoculated samples.

28

Thermal resistance of Salmonella was higher (P<0.0001) in whole-muscle

than in coarse-ground, fine-ground, and pureed samples (Fig. 3.7, Table 3.1).

    
 

; Whale—.rriisae— ‘
, l X Coarse-ground

l
‘. D Fine-ground

-3.0 w,
H O Pureed ‘ ‘
_4.0 ji-H-Whole nuscle, Regression :
‘ 1—-Coarse/Fine/Puree,
Regression
O 1 2 3 4 5 6

Time (min)

Figure 3.7 Salmonella survival curves in whole-muscle, coarse-ground, fine-

ground, and pureed beef samples heated at 60°C.

Table 3.1 First-order inactivation constant, k, and D value (mean :1: S) at 60°C,
calculated by linear regression of Salmonella survivor data for whole-muscle,

coarse-ground, fine-ground, and pureed beef.

 

 

 

 

 

|8eef Structure kaooc (mind) Deane value (min) R‘Lr
Whole—muscle 0.87 :I: 0.20 2.73 :I: 0.64 0.73
Coarse-ground 1.99 :I: 0.10 1.16 :t 0.06 0.92
Fine-ground 2.06 x 0.29 1.12 :I: 0.20 0.90
Puree 1.87 :t 0.11 1.26 :I: 0.07 0.91

 

 

 

 

 

 

29

Based on a Tukey-Kramer multiple means test (or=0.05), the rate of
Salmonella inactivation in the whole-muscle samples was lower (P<0.05) than in
the other three sample types, but there was no difference (P>0.05) among the

rates in the three ground/pureed sample types.

30

3.5 DISCUSSION

All samples came from the same original lot of beef, and had the same
thermal history and initial Salmonella counts; therefore, differences in thermal
resistance were due to the degree of grinding (i.e., product structure). Salmonella
had a higher heat resistance in whole-muscle compared to coarse-ground, fine-
ground, and pureed beef. Even though the purpose of this study was not to
understand the mechanism by which Salmonella exhibits more heat resistance in
whole-muscle compared to its comminuted products, some assumptions can be
drawn that might lead to further research to understand this phenomenon.

Orta-Ramirez and others (2005) found the same trend when comparing
whole-muscle beef and ground beef, stating that the physical arrangement of the
food components within the food matrix might cause a difference in bacterial
thermal resistance. According to Doyle and Mazzotta (2000), the location of
Salmonella in the food may also have an effect on the organism’s thermal
resistance.

Previous research has shown that the amount of fat in meat affects the
heat resistance of bacteria, which increases with fat content (Hansen and
Riemann 1963; Stumbo 1973; Ahmed and others 1995; Juneja and others 2001;
Smith and others 2001). It is possible that bacteria attach to segregated fat tissue
in the whole-muscle, giving a protective effect to those microorganisms. The fat
protective effect might be lost in the ground and pureed products, due to the

homogenous distribution of fat (Orta—Ramirez and others 2005).

31

The water status may also affect thermal resistance of Salmonella.
Perhaps the chemical potential of water in whole-muscle is lower than in the
ground product. Even if the moisture content is the same, it might be possible
that some limited osmotic potential across muscle cell membranes might
increase Salmonella thermal resistance. It has been shown that the thermal
resistance of Salmonella is significantly greater when the aw of meat is reduced
(Carlson and others 2005).

Others have suggested that Salmonella has a higher thermal resistance
when attached to meat than when suspended in broth (Thomson and others
1979; Humphrey and others 1995; Murphy and others 1999). In muscle, bacteria
may have a better opportunity to attach than in ground product (Orta-Ramirez
and others 2005). Therefore, Salmonella in the ground and pureed beef products
may be suspended in the liquid component of the food, making it more
susceptible to thermal inactivation.

The results of this study suggest that thermal process validations should
also consider the structure of beef, which appears to affect the inactivation of

Salmonella.

32

3.6 CONCLUSIONS

Previous studies have shown the importance of product composition on
thermal resistance of Salmonella, but the distribution of the components has not
been considered. In this research, the physical structure of beef products
influenced Salmonella thermal resistance. However, no significant difference in
thermal resistance was seen between coarse-ground, fine-ground, and pureed
samples. Therefore, it might be important for Salmonella thermal inactivation
models to consider whether a product is whole-muscle or ground, but not the

degree of grinding.

33

CHAPTER 4. THE EFFECT OF COOKING PROFILES AND
SUBLETHAL HISTORY ON SALMONELLA THERMAL INACTIVATION IN
WHOLE-MUSCLE BEEF

4.1 ABSTRACT

Sublethal heating is known to have an effect on Salmonella thermal
resistance. However, it has not yet been considered when validating thermal
process compliance with regulations, which require a 6.5 log reduction for
Salmonella in cooked and ready-to-eat beef products.

Therefore, the objective of this study was to determine the effect of
cooking profiles (i.e., thermal history, including sublethal heating) on Salmonella
thermal inactivation in whole-muscle beef. Small, irradiation-sterilized beef steaks
were inoculated with an 8-strain Salmonella cocktail (~108 CFU/g) and subjected
to 6 different cooking schedules until 6.5 log reductions in Salmonella were
predicted by traditional log-linear inactivation kinetics applied to real-time lntemal
temperature data. The cooked samples were immediately cooled, serially diluted
in peptone water and plated on PetrifilmTM aerobic count plates to enumerate
survivors. The experimental results were compared to predictions from both the
traditional model and a recently-developed model accounting for the effect of
sublethal thermal history on subsequent inactivation rate.

Sublethal heating, occurring during slow heating profiles, significantly

decreased (P<0.05) Salmonella inactivation.

34

4.2 INTRODUCTION

Ready-to-eat (RTE) and cooked beef products need to comply with USDA
regulations, which require a 6.5 log reduction of Salmonella spp. ([FSIS] United
States Food Safety and Inspection Service 2001). While used as the reference
organism, inactivation of Salmonella also implies the reduction of other
pathogens that might be present in the food. However, when verifying thermal
processes, the effect of sublethal heating is not considered, which might lead (in
certain cases) to a miscalculation of process lethality and a potentially unsafe
product.

Cooking is an effective way to reduce microorganisms in food. However,
when bacteria are heated to temperatures slightly above their growth range, they
undergo a sublethal heat shock, which can increase their subsequent thermal
resistance (Mackey and Derrick 1986; Bunning and others 1990). Beef products
are usually heated slowly to avoid fluid loss, which may lead to heat-stressed
microbial cells, because the product can spend appreciable time in the sublethal
temperature range. In addition, other procedures can also result in heat-stressed
bacteria in food, including interrupted cooking cycles, carcass exposure to hot
acid sprays, or holding product under heat lamps or in warming trays (Mackey
and Derrick 1987; Bunning and others 1990; Murano and Pierson 1993; Juneja
and Novak 2003; Youself and Courtney 2003).

Studies have shown that when Salmonella is exposed to sublethal
heating, which includes temperatures of 40 to 50°C for 15 to 30 min, the

organism develops an enhanced thermal resistance. This increased thermal

35

resistance is acquired rapidly due to the synthesis of heat shock proteins
(Mackey and Derrick 1986; Mackey and Derrick 1990; Humphrey and others
1993). Also, heat stressed Salmonella increase their concentration of saturated
membrane phospholipids, due to the variations in the fatty acid composition of
the membrane (Doyle and others 1997). Consequently, the cell membrane
looses fluidity, which results in a pathogen with an increased resistance to heat
injury (Humphrey and others 1993).

Previous research shows that Salmonella that has been subjected to
sublethal heat can acquire some protection against subsequent heating at
normal lethal temperatures (Mackey and Derrick 1986). Wesche and others
(2005) showed that the Deooc values of a heat shocked (54°C for 30 min) eight-
strain Salmonella cocktail in ground turkey increased ~56% and 106%, compared
to unshocked Salmonella when recovered on trypticase soy agar and xylose
lysine deoxycholate media, respectively. In a study with ground beef, the time at
60°C needed for 4 log reductions in heat shocked E. coli O157:H7 increased by
~50% when compared to unshocked cells (Juneja and others 1998). The
previously observed effect of sublethal heating on thermal resistance could
potentially affect the safety and regulatory compliance of slow-cooked beef
products; however, this has not been previously tested, modeled, or reported for
beef

In a previous study, a model to predict the effect of prior sublethal thermal
history on thermal inactivation of Salmonella in ground turkey was developed

(Stasiewicz and others 2005; 2006). This model was developed using data

36

obtained from samples (0.1 g) sealed in a container and heated in a
therrnocycler, which is not a real industrial scenario, where larger amounts of
meat are usually moist cooked in a smoke house or oven. Also, the programmed
heat treatments used in that study were not simulating an industrial cooking
schedule. Due to the sample size, cooking method and schedule, and the meat
species used by Stasiewicz and others (2005 and 2006), predictions of the effect
of sublethal heating on Salmonella thermal inactivation need further investigation
for beef processed under industrial conditions.

Therefore, the objective of this study was to determine the effect of
sublethal heating on Salmonella in whole-muscle beef subjected to simulated

oven cooking schedules spanning a range of heating times.

37

4.3 MATERIALS AND METHODS

To test the effect of sublethal heating on Salmonella in whole-muscle beef,
small inoculated beef steaks (~3.4 cm diameter, 1 cm thick) were cooked in a
moist air convection oven at different cooking schedules, and survivors were
enumerated on PetrifilmTM aerobic count plates. The five different cooking
schedules, according to conventional process lethality verification tools, should
have achieved 6.5 log reductions of Salmonella in the product.
4.3.1 Meat Preparation

The same irradiated beef chuck, shoulder clods (1140, according to
(North American Meat Processors Association 1997)) were used as in the
previous study (Section 3.3.1).

The sterile beef muscle was sampled using a sterile coring device (3.4 cm
diameter) to obtain whole-muscle cylinders. The cylinders were then cut, using a
sterile scalpel, into small steaks, ~1 cm thick (Fig. 4.1), so that the muscle fibers

were roughly parallel with the short dimension.

 

Figure 4.1 Raw beef steak sample.

38

4.3.2 Bacterial Cultures

The same Salmonella cultures were used and prepared as described in
Chapter 3 (Section 3.3.2) preparation methods.

4.3.3 Marinade Preparation

The same generic marinade formula was prepared, as described in
Chapter 3, Section 3.3.3.

4.3.4 Inoculation

Immediately before inoculating the beef samples, the Salmonella cocktail
was prepared by combining equal volumes (9 mL) of each serovar in a centrifuge
bottle. The bottle was centrifuged (6,000 x g, 20 min at 4°C), and the supernatant
was removed. The pellet was resuspended in 520 mL of the marinade. As a
result, the marinade contained ~10° CFU/mL. Two samples were then
simultaneously immersed in the inoculated marinade for 20 min. One sample
was used to determine the initial counts of Salmonella, and the remaining sample
was used for thermal inactivation.

The Salmonella concentration in the marinade was verified by serial
dilution in 0.1% peptone water, which was plated in duplicate on PetrifilmTM
aerobic count plates.

4.3.5 Thermal Inactivation

Samples were thermally processed in a custom, computer-controlled,

moist air convection oven. The oven control parameters were dry and wet bulb

temperatures (°C). The oven and control software were both designed and

39

constructed by Dr. Sanghyup Jeong at Michigan State University (East Lansing,
Mich., USA).

The oven chamber (76.2 cm x 50.8 cm x 45.7 cm) is made of stainless
steel and is insulated with fiberglass. The key elements of the oven are the
steam generator, the conditioning chamber, the electrical heating elements, the
centrifugal fan, and the sample chamber (Fig. 4.2) (Tripuraneni 2006). The steam
generator (750 W) has an immersion heater to create steam from distilled water.
The steam is injected into the conditioning chamber by a solenoid and a pressure
regulating valve (344,738 N/m2). The conditioning chamber has 4 strip heaters
(350 W each) to heat and condition the air that is circulated (~1.3 m/s) by the
centrifugal fan (6 W) to the sample chamber. The sample chamber (10 cm x 10
cm x 10 cm) is where the samples (supported on a wire mesh screen) are placed
and cooked; the temperature in the sample chamber is monitored by a
thermocouple (type T).

The air relative humidity (RH) is monitored by a polymer sensor (Vaisala
DRYCAP®, Vaisala, Helsinki, Finland) and a dew point transmitter (Vaisala
DMP246, Vaisala, Helsinki, Finland) (Tripuraneni 2006). All the heaters and valve
controls are controlled by computer software. The oven dry bulb temperature can
range from 25 to 200°C (:1 °C) and RH from 0 to 90% (11%) and 90 to 100%

(12%) (Carlson 2002).

40

 

 

 

 

CONDITIONING CHAMBER

 

 

 

 

 

 

 

 

SAMPLE CHAMBER STEAM
GENERATOR

 

 

 

 

 

 

Figure 4.2 Moist-air oven components arrangement (showing airflow direction).

Six different cooking schedules (control and A to E) were used (Table 4.1)
to roughly approximate the internal temperature history that would occur across a
range of commercial cooking processes. Each cooking schedule had the same
dry and wet bulb temperatures in each of five stages, but the product hold times
at each temperature were different, except for the control schedule. In the control

cooking schedule, samples were subjected to isothermal conditions at 60°C.

41

Table 4.1 The temperatures, relative humidities, and durations of each stage (1

to 5) for the six cooking schedules (control and A to E).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a. 2' Cooking Schedule
E
O o o
c I" H A *" A I A B C D E
33 8) a 9 g 9 G Control
45 . .
g 3 Time (mm)
1 32.0 28.1 80.0 n/a 4.0 9.6 15.2 20.7 26.3
2 35.0 33.1 90.0 n/a 4.5 10.8 17.1 23.3 29.6
3 45.0 42.9 90.0 n/a 6.5 15.6 24.6 33.7 42.8
4 55.0 52.8 90.0 n/a 6.5 13.2: 17.4: 17.1: 16.6:
1.7* 1.5* 0.5* 1 .1*
5 60.0 57.7 90.0 10.41- 9.1: n/a n/a n/a n/a
0.5 1.3*

 

 

 

 

 

 

 

*The end time was when 6.5 Salmonella log reductions were predicted by

traditional log-linear inactivation kinetics.

The internal temperature of the samples and the oven temperature were
monitored using thermocouples (Type T, 1.0 mm, Omega Engineering, Stamford,
Conn., U.S.A.). Both thermocouples were connected to an OMEGA data logger
(Danook 120, OMEGA Technologies Company, Stamford, Conn., USA),
which transmitted the information to a computer where the data were collected.
The software that was used to collect the data was a Real-Time Lethality Monitor
(v2) (created by Dr. Sanghyup Jeong at Michigan State University, East Lansing,
Mich., USA), which used the sample internal temperature to calculate the

predicted Salmonella log reductions that were achieved in the sample. The D

42

value (3), 2 value (3), reference temperature (°C), target lethality, and sampling

interval (3) were inputs to the Real-Time Lethality Monitor (Fig.4.1)

 

 

REAL-TIME LETHALITY MONITOR Position cursor at a cell

where you want to record

CHEESE. Letflafltjfi T8139! _|:§th.a_m'l’ sampled data before you

launch monitor.
6.50
l

I
u.._..._.-..__..,_ _ _—.—.-. __...._ i .— -.._._ .1..— m.

 

 

90900916894099. ,. -

 

 

_...9h1__f.._._cr~_2_ ;._:Eh.3__-_.__-_cn.4___ '
2930 l 29.00 43.00 E 13.90 -
Remark > girl-ti. i999 _' 39.. 19

Microbial Inactivation Parameters

0’ -__ __§_2_-_3:S

Tr_ __ 80:0

2 _. 9.99750
TargetL _ _“6.5:log[NlNo]

 

Data .Accjtris‘rtirjn Parameters:
Desired sampling interval ____WH__ 1‘ s

   

 

"‘99" if
Number of Samples 3599992: BIOSYSTEMS

LEU‘IOIII! IDOEII CI‘ISI'II'IEI 4; v r95; L399}? .\

 

 

 

 

   

 

 

Figure 4.3 Screen view of the Real-Time Lethality Monitor (v2).

When the lethality monitor showed that the sample had reached a
predicted 6.5 log reductions of Salmonella, the samples were immediately
removed from the oven and immersed in 36 g of cold, sterile 0.1% peptone water
(4°C) for cooling. Samples cooled down to 40°C in 21.8 x 13.3 s. Triplicate

samples were tested in each cooking schedule.

43

4.3.6 Salmonella Enumeration

The cooled samples were diluted 1:10 in sterile 0.1% peptone water,
taking into account that they have already been diluted in 36 g of peptone water.
Each diluted sample was homogenized for 180 s in a masticator (Model 0410,
IUL Instruments USA, Inc. Cincinnati, Ohio, USA), serially diluted in peptone
water, and plated on PetrifilmTM aerobic count plates (48 h, 37°C). All work was
accomplished under aseptic conditions using sterile instruments.

4.3.7 Models

The Deooc (82.3 s) and 2 values (585°C) that were set in the Real-Time
Lethality Monitor (v2) software were calculated from data reported in a previous
study by Orta—Ramirez and others (2005). That previous study tested the thermal
resistance of Salmonella (same 8-serovar cocktail) in whole-muscle beef from
which the values were calculated via linear regression of the survivor curves
(n=12)

After subjecting the samples to different cooking schedules, the
temperature profiles that were used to calculate Salmonella log reductions in
whole-muscle beef using the traditional model (Real-Time Lethality Monitor v2)
were then used to calculate predicted Salmonella lethality using a WeibulV
Arrhenius model and the path-dependent, Stasiewicz model (Stasiewicz and
others, 2005 and 2006).

Both of these models were developed with data from the same Salmonella
cocktail as this study, but with ground turkey, not whole-muscle beef. However,

these models are the only known direct comparison of a model accounting for the

44

effect of sublethal history on subsequent thermal resistance. Therefore, these
two models were applied to the beef results to predict a relative, expected impact

of sublethal history on process lethality.

45

4.3.7.1 Traditional model

The traditional model was a log-linear model. The D value (Equation 4.1)
was calculated for each time interval (t) from the reference D value (Def), 2 value
(2), reference temperature (Tref), and internal temperature (T(t)) of the sample at
each time interval. Then the Salmonella log reductions (Log N/No) (Equation 4.2)
were calculated from the previously calculated D values, at each time step, and
the cumulative log reductions were calculated as the sum of log reductions at

each time step.

Tm, T(t)
D = Dref x10 2 Equation 4.1
.0 _r_1-_ _d_t .
9 No - D Equation 4.2

4.3.7.2 Weflzgll/ Arrhenius mod_el

The traditional log-linear model should not be used when the thermal
inactivation of the organism is non-linear; then the Weibull model is commonly
used (Stasiewicz and others 2006). This model calculates the lethality of the non-
linear survival curve, described as the power-log function (Equation 4.3) where b

and n are temperature-dependent parameters (Equation 4.4).

N
IOQ‘N‘O‘ = ‘btn Equation 4.3

b(T) = brefe—B‘LTItl—T; I

 

Equation 4.4

46

From isothermal inactivation of the same Salmonella cocktail in ground
turkey, Stasiewicz (2005 and 2006) reported n: 0.54, brag: 1.12, [31: 31275, and
Tref= 58°C.
4.3.7.3 Path-depengent (Stasiewicz) model

The Stasiewicz model is a modified Weibull/Arrhenius model, which
considers the effect of Salmonella sublethal history in ground turkey samples.
The key modification of the model is that b is a function of both the current state

and prior sublethal history. Sublethal history was quantified as 1' (Equation 4.5).

_ T=HSupper _ .
t— IrzHSlower [T(t) HSIowerIdt Equation 4.5

Then the model for b becomes:

 

Ln(b(t)) = Ln(bref )_ E1(T(t) ‘ T1 f ]‘ P2T Equation 4.6
re

Where HS.ower = the lower temperature in the sublethal region, HSupper =
the upper temperature in the sublethal region, and T(t) = current temperature at
every given time.

This modified model was then used to calculate potential lethality for all of

the cooking schedules.

47

4.4 RESULTS AND DISCUSSION

4.4.1 Raw Meat

The raw beef contained 73.04 1 0.93% moisture and 4.47 1 0.07% fat.
The uninoculated, irradiated beef samples contained no detectable background
microflora after plating 1:5 dilution.

4.4.2 Inoculation of Marinade and Meat

The marinade was inoculated to obtain ~108 Salmonella CFU/mL, with an
average Salmonella population of 8.78 1 0.43 log CFU/mL as determined by
direct plating.

Initial Salmonella counts in untreated samples of the whole-muscle beef
were 7.86 1 0.20 log CFU/g, with no differences ($0.05) observed between
samples used in the six different cooking schedules.

4.4.3 Cooking Time at Each Cooking Profile

Even though each cooking schedule (A to E) had 5 cooking stages, only
cooking schedule A required 5 stages to achieve the predicted 6.5 log reduction
of Salmonella. Because the predicted 6.5 log reduction was reached in stage 4 of
cooking schedules 8, C, D, and E, the samples were removed from the oven
before reaching stage 5.

Total cooking times for schedules A, B, C, D, E, and the control were
30.60 1 1.25 min, 49.20 1 1.65 min, 74.27 1 1.51 min, 94.83 1 0.45 min, 115.30 1
1.11, and 10.40 1 0.53 min, respectively (Figures 4.4 to 4.9).

The shape of the temperature curves from cooking schedules A, B, C, D,

and E in Figures 4.4 to 4.8, respectively, all have the same shape; the difference

48

is the time held at each cooking stage. For the control cooking schedule, the
temperature curves were different from the other cooking schedules, because the

samples were cooked under a constant oven condition.

70; 1
601

so 4

Temperature (°C)

 

0 1 1 1 11 .1 .1 .1 1 1 1 1 .11 1 _.1 1 ,1 1 1 11 1 1 1 1, 1 1 1 1.
0 500 1000 1 500 2000
Time (s)

Figure 4.4 Temperature curves of samples in cooking schedule A.

49

Temperature (°C)
«b or on \I
o o o o

 

1 1 11 11 .1 1. 1 1 1. .1 .1 1 1. 1 1 1..

0 500 1000 1500 2000 2500 3000 3500
Time (s)

Figure 4.5 Temperature curves of samples in cooking schedule 8.

NOD-hU'IODV
OOOOO

Temperature (°C)
o

 

_.L
00

1 1 1 1 ,111 1 1

0 500 1000 1500 2000 2500 3000 3500 4000 4500
Time (s)

Figure 4.6 Temperature curves of samples in cooking schedule C.

50

7O

Terrperature (°C)

 

0 1000 2000 3000 4000 5000 6000 7000
Time (s)

Figure 4.7 Temperature curves of samples in cooking schedule D.

QVCD
OOOO

 

Temperature (°C)
(.0 vk 01
O O

N
O

10

O ..l . - - _. _ .- 1 1 1. 1 1. . _ 1 _. .1 - _. 1 1 .
1 000 2000 3000 4000 5000 6000 7000 8000
Time (s)

Figure 4.8 Temperature curves of samples in cooking schedule E.

51

 

 

MOD
0

Temperature (°C)
0

 

—L
00

0 1 00 200 300 400 500 600 700
Time (s)

Figure 4.9 Temperature curves of samples in the control cooking schedule.

4.4.4 Predicted Lethality Curves for Salmonella in Beef Samples for Each
Cooking Schedule

The predicted process lethality for each replicate in each treatment, based
on traditional log-linear inactivation modeling (Figures 4.10 to 4.14) did not begin
accumulating appreciably until the end of each cooking schedule. However, the
lethality of the control cooking schedule started increasing early, because of the
rapid increase in sample temperature. Longer durations of each cooking stage

increased the time to reach a predicted 6.5-log reduction in Salmonella.

52

Nwhmmfl

Salmonella Log Reductions

 

0 . 1 __ 1111111 _ ___ -___ ____---9

1 000 1 500 2000

Figure 4.10 Lethality curves for Salmonella in samples cooked using schedule A.

Salmonella log reductions
m w a. 01 o: \r

 

1 500 2000 2500 3000

Figure 4.11 Lethality curves for Salmonella in samples cooked using schedule 8.

53

Salmonella log reductions

Salmonella log reductions

 

 

3000 4000 5000 6000
Time (s)

Figure 4.13 Lethality curves for Salmonella in samples cooked using schedule D.

54

Salmonella log reductions
N on J:- 01 0) xi

1

 

 

 

 

A1#:::;::::~——

0.1—111111-- -- --11-1 -- -11- .1-
4000 4500 5000 5500 6000 6500 7000

Time (s)

Figure 4.14 Lethality curves for Salmonella in samples cooked using schedule E.

Salmonella log reductions
re 00 A 01 o> xi

0111.11 1.111111 1
200 300 400 500 600 700

Time (s)

Figure 4.15 Lethality curves for Salmonella in samples cooked using the control

schedule.

55

4.4.5 Predicted Salmonella Log Reductions Using the Stasiewicz and
others (2005 and 2006) model

According to the Stasiewicz model, all cooking schedules should have
achieved between 6.15 and 7.09 log reductions in Salmonella when the effect of
sublethal heating is not considered (Table 4.2). On average, the Weibull model
(based on data from ground turkey) predicted 6.56 log reductions in Salmonella,
which was very close to the predicted 6.5 log reductions predicted by the
traditional model (based on data from whole-muscle beef). When the effect of
sublethal history was considered, the predicted Salmonella log reductions of
cooking schedules A, B, C, D, E, and control were ~13%, 22%, 56%, 57%, 70%,

and 3% lower, respectively, than when not considering sublethal heating.

56

Table 4.2 Predicted Salmonella log reductions using the Stasiewicz and others

(2005 and 2006) model for Salmonella in ground turkey based on temperature

profiles used in this study.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NC* I C** I NC*I C** I NC*j C“ NC" 0“
Cooking Replicate
Schedule 1 2 3 Avera e 18
A 7.13 6.08 7.05 6.11 7.09 6.28 7.091004 6.161011
8 6.73 5.46 6.66 5.48 6.28 4.47 6.561024 5.141058
C 6.15 2.92 6.19 2.80 6.11 2.47 6.151004 2.731024
D 6.26 3.62 6.29 1.95 6.31 2.55 6.291003 2.711085
E 6.41 1.71 6.50 2.23 6.31 1.83 6.411010 1.921027
Control 6.84 6.62 6.86 6.66 6.98 6.78 6.891007 6.691008
All samples 6.561035
NC" -Predicted Salmonella log reductions calculated without considering

sublethal heating, using a Weibull/Arrhenius model.
0“ -Predicted Salmonella log reductions calculated considering sublethal

heating using the Stasiewicz and others (2005 and 2006) model.

4.4.6 Salmonella Experimental Log Reductions in Beef Samples for Each
Cooking Schedule

All samples had the same composition, initial Salmonella counts, and
were rapidly cooled when the predicted Salmonella log reductions were 6.5;
therefore, all of the differences in Salmonella inhibition are due to the cooking
schedules (Table 4.3). There was no difference in the experimental Salmonella
lethalities among cooking schedules A, B, and the control. In addition, there were
no significant differences (0:005) between the outcomes from schedules 8, C,

D, E, and the control.

57

 

Table 4.3 Salmonella experimental log reductions in beef samples at each

cooking schedule.

 

 

 

 

 

 

 

 

 

 

 

Logreductions

Cookin Re Iicate

schedulge 1 p2 3 Average 3
A 6.49 5.87 >7.89 6.75 **" 1.03
8 3.88 4.12 4.83 4.28 “a” 0.49
c 3.98 4.52 1.66 3.39 M 1.52
o 4.31 1.82 2.00 2.71 **° 1.39
E 3.43 2.92 2.18 2.84 "a 0.63

Control 5.43 4.90 5.44 5.26 "a" 0.31

 

 

 

 

 

 

“Results not connected with the same letter are significantly different (0:0.05),

based on a Tukey-Kramer mean comparison.

Based on a Student’s t test, the Salmonella log reductions in beef samples
that were cooked according to schedules A and control were not different
(P>0.05) from the predicted 6.5 log reductions of the traditional model. Therefore,
these cooking schedules comply with USDA regulations and closely match the
lethalities predicted by the traditional model. Also, according to a Student’s t test
(0:0.05), samples cooked in schedules 8, C, D, and E had lower log reductions
in Salmonella, compared to predicted log reductions from the traditional model.
The longer the samples were exposed to sublethal heating, the more resistant
the organisms became to lethal temperatures (Fig. 4.16). Enhanced thermal
resistance of the pathogen can be seen in samples cooked using schedules 8
and C, which achieved only ~3 to 4 log reduction in Salmonella. Also, cooking

schedules D and E, yielded reductions of only ~27 to 2.8 log.

58

10.00 I Traditional model
E Weibull model
[I] Stasiewicz model

   
  

      

        
         
     
   
  

   
   

           
         

         
  

  
    
        
     
         
   
     
     

   
    

  
  
 

        

  
 
 
       
       
       
       
     
    

      

   
   
       
       
    

            

- 19] Experimental results
.
i
(I) I t
c 1| b1
_ 1‘
,9 7.00 *1 = *k k *1 *g ' b
4.- = a. 1' *g =
0 = ,0... I =
3 = ,9... = =
U 6 -00 E ’0’ E E =
a: _ ,0... _ _ =
\— — _ —- —
_ ,0... _ _ _
5’ E ’0’ E = E - .
2 5 00 = ’0‘ = = = ’9‘
' = 1:0: r = = = 5‘:
to = 59,- I = = = 30,-
§ _ . . 1 _ — — . . .
= 0,4 u = = = 00
0 4 00 — a. v. — — - ’o"
c - = v. m = *h = = 5"
O = r o N = = = '0‘
_ .0... .0; — — — 94'.
E = ,0... .6; = v.9; = ,. ,. = 9,0,
\ — O . . ‘ — . . , — f f — . .
(U 3 -00 = ”0‘6 VI = 90’. = = 30‘
- . . . . . _ . . - . . ‘ — . . .
<0 = 9.! N = W = v. 0.0. = 9.!
= ,9 o, o r = m = > o 0 o = .0 0.
_ o O o _ o o _ o c o 1 _ o
_ ,6... .0; _ w _ v.6. 0.0, _ .0...
2 .00 = ,0... .04 = a; = 56, o... = 9.5
= ’01 c‘o‘ = ’o’o‘ = ’0’. ’9‘ = 999
= 1‘9.- .‘o‘ = 'o’o' = 5.0 ’0’. = 90‘
= 9.! N = w = 5°. r... = 2.9
_ O O 4 - r O 1 - D O O O — O O
1 00 — 1 ., — o o — o a o r — - or
I - . . — . . . — p . . . — . .
= - 6- = .0 o. = o c 9 1 = ~ .
_ N. _ o _ >0 M _
— 3.. — 9... - b... .O.‘ — ' ‘
o 00 . = .0... .= 30; = r o =

 

A B C D E Control
Cooking schedule

Figure 4.16 Comparison of predicted Salmonella log reductions from the

traditional, Weibull, and Stasiewicz models and experimental values.

* Bars (Figure 4.16) within an individual cooking schedule not connected by the
same letter are significantly different, based on a Tukey-Kramer means test

(0:0.05).

Only cooking schedule A yielded similar results when compared to the
traditional model, Weibull model, Stasiewicz model, and experimental results.
The traditional model and the Weibull model were not statistically different for any

of the cooking schedules. For cooking schedules A, B, C, and D, the

59

experimental Salmonella log reductions were not statistically different from the
predicted log reductions from the Stasiewicz model, which considers sublethal
history. For schedule E, the slowest cooking schedule, the Stasiewicz model
slightly under-predicted lethality, and for the control schedule, the fastest cooking
schedule, the Stasiewicz model slightly over-predicted lethality.

In general, the Stasiewicz model reasonably predicted the impact of prior,
sublethal heating on subsequent process lethality (Fig. 4.17). However,
differences between the Stasiewicz model and the experimental results are
probably due to the fact that the model was developed with data from ground
turkey under different conditions and cooking modes. However, the model
demonstrates how sublethal heating can lead. to enhanced survival of
Salmonella, and thus might threaten the safety of slow-cooked meat products, if
not considered. A beef-based model should be developed to more accurately

calculate lethality when such products are subjected to sublethal thermal history.

60

h 01

O)

Observed predicted effect on log NINi,
Traditional minus Expenmental

 

 

 

0 1 2 3 4 5
Predicted protective effect on log NlNo,
Weibull minus Stasiewicz

Figure 4.17 Comparison of the predicted versus observed relative effect of prior

cooking history on subsequent process lethality.

61

4.5 CONCLUSIONS

Sublethal heating has a significant effect on Salmonella thermal
inactivation in whole beef samples. When the samples were exposed longer to
sublethal heating, the pathogen became more heat resistant to lethal
temperatures. Therefore, significantly more salmonellae survived in samples that,
according to traditional calculations, should have achieved a 6.5-log reduction.
Sublethal heating might be a risk factor, which has not been considered in the
beef industry, but which should be considered in thermal lethality models being

used for commercial process verification.

62

CHAPTER 5. OVERALL CONCLUSIONS

Salmonella is more thermally resistant in whole-muscle beef compared to
its comminuted counterpart. However, the degree of grinding has no effect on
thermal inactivation of Salmonella.

The thermal resistance of Salmonella is higher when the organism has
been subjected to previous sublethal heating during cooking. Longer exposure to
sublethal temperatures increases resistance to subsequent lethal temperatures.
Cooking schedules that subject beef to sublethal temperatures might increase
the organisms’ subsequent thermal resistance.

When verifying beef process lethality, the beef industry should consider
that the thermal resistance of Salmonella varies due to the structure of the meat
and previous sublethal thermal history. Therefore, the current lethality prediction
tools might be underpredicting the Salmonella inactivation in cooked beef
products, which can in turn lead to a potentially unsafe product that does not

comply with the current USDA regulations for RTE and cooked beef products.

63

CHAPTER 6. FUTURE RESEARCH

The purpose of this research was to determine the effect of beef physical
structure and sublethal heating on Salmonella thermal inactivation. Because both
factors were shown to impact product safety, future research should consider the
creation of a predictive model that considers these two variables.

Also, because this study was only designed to determine whether beef
structure affected Salmonella thermal inactivation, the mechanism by which this
pathogen is more heat resistant in whole-muscle compared to its ground
counterpart was not determined. Future research should determine the
mechanism by which muscle structure or arrangement protects Salmonella
against thermal inactivation. Such information will help the beef industry address
this issue and may lead to further research to minimize the enhanced thermal
resistance of this pathogen.

The work on the cooking of beef at sublethal temperatures should also be
applied to ground beef patties to determine the difference and the effect on
Salmonella thermal inactivation when compared to whole-muscle products. This
project should also be extended to the pilot plant level to verify these results, and
assess the potential for this problem to occur in current industrial beef cooking

schedules.

64

APPENDICES

65

APPENDIX A

COMPARISON OF INOCULATION METHODS OF GROUND PRODUCT
SAMPLES

To determine the appropriate inoculation method (seeking homogenous
distribution of the Salmonella and minimized processing injury) of the ground
beef samples, for the study on “The Effect of Beef Product Physical Structure on
Salmonella Thermal Inactivation” (Chapter 3), several procedures were tested,
as:

1) Method I: Chopped pork was inoculated before passing the product the

first time through the grinder.

2) Method II: Ground pork was inoculated before passing the product a

second time through the grinder.

3) Method Ill: The pork loin was marinated and vacuumed before

grinding.
The methods and results are described in this Appendix.

The Salmonella cocktail and marinade used in these tests were prepared
in the same manner as described in Chapter 3 (Sections 3.3.2 and 3.3.3).

A.1 Fine-Ground

Irradiated (~10 kGy ) pork loins (Iongissimus dorsi muscle) were used to
test how homogenous the Salmonella inoculum was distributed throughout the
fine-ground pork samples. Three different procedures were tested to determine

the best method of inoculation.

66

Sterility of the pork was tested by diluting (1:5) the samples in peptone
water (DifcoTM, Becton, Dickinson and Company, Sparks, Md., USA.) and
plating on PetrifilmTM aerobic count plates (3M, St. Paul, Minn., USA). All
samples were prepared immediately before starting the experiment.

Method l

The pork loin was first sliced into 3 cm thick pieces and then inoculated
dropwise in the same proportion as the uptake of the whole-muscle samples
(0.15 g marinade/g meat) before the product was passed once through the
grinder (fitted in a plate with 6 mm diameter holes).

After inoculation, 6 random samples (5 1 0.05 g) were taken to enumerate
Salmonella, using the same procedure as described in Chapter 3 (Section 3.3.6).

The results are shown in Table A.1.

Table A.1 Salmonella counts (mean 1 standard deviation) in 6 random samples
of fine-ground pork, which were prepared by passing the pork once through the

grinder (plate with 6 mm diameter holes).

 

 

 

 

 

 

 

 

 

Sample Duplicate plates Average Log Average S Log
CFUIL CFUIg CFU/g CFUIg L39 CFUIg CFUIg

1 1 .48E+07 1.44E+07 1 .46E+07 7.16
2 1.50E+07 1 .51E+07 1.51E+07 7.18
3 2.60E+07 2.50E+07 2.55E+07 7.41 7 50 0 44
4 7.20E+07 6.20E+07 6.70E+07 7.83 ' '
5 1 .64E+07 1 .44E+07 1 .54E+07 7.19
6 1 .74E+08 1 .60E+08 1 .67E+08 8.22

 

 

 

 

 

 

 

67

 

Method II

The pork was ground through a plate with 6 mm diameter holes. Then, the

product was inoculated dropwise in the same proportion as the uptake of the

whole-muscle samples (0.15 g marinade/g meat) before the product was passed

a second time through the grinder using the same plate.

After inoculation, 6 random samples (5 1 0.05 g) were taken to enumerate

the existing Salmonella. The enumeration of Salmonella was performed in the

same manner as described in Chapter 3 (Section 3.3.6). Results are reported in

Table A2.

Table A2 Salmonella counts (mean 1 standard deviation) in 6 random samples

of fine-ground pork, which were prepared by passing the pork twice through the

grinder (plate with 6 mm diameter holes).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Duplicate plates Average Log Average S Log
CFU/l CFUIg CFUIg CFU/g LogCFU/g CFUIg__
1 6.20E+07 6.50E+07 6.35E+07 7.80
2 5.80E+07 6.30E+07 6.05E+07 7.78
3 3.50E+07 3.10E+07 3.30E+07 7.52 771 013
4 3.40E+07 4.10E+07 3.75E+07 7.57 ' '
5 6.10E+07 6.80E+07 6.45E+07 7.81
6 6.60E+07 4.90E+07 5.75E+07 7.76
MM

A pork loin (298.49 9) was packed in a sterile bag containing 44.77 g of

marinade. The bag containing the product and its marinade were subjected to

vacuum (40 cmHg) for 20 min. Then, the pork was ground once through a plate

with holes of 6 mm diameter.

68

 

After inoculation, 6 random samples (5 1 0.05 g) were taken to enumerate

the existing Salmonella. The enumeration of Salmonella was performed in the

same manner as described in Chapter 3 (Section 3.3.6). Results are reported in

Table A.3.

Table A.3 Salmonella counts (mean 1 standard deviation) in 6 random samples

of fine-ground pork, which were prepared by subjecting the pork loin to vacuum

(40 cmHg) marination for 20 min and then passing the pork loin once through the

grinder (plate with 6 mm diameter holes).

 

 

 

 

 

 

 

 

Sample Duplicate plates Average Log Average S Log
CFUIg CFUIg CFUIg CFU/g Log CFU/g CFUIg
1 4.10E+07 5.20E+07 4.65E+07 7.67
2 2.18E+09 2.25E+09 2.22E+09 9.35
3 1.13E+07 1.13E+07 1.13E+07 7.05 789 081
4 1.23E+08 1.20E+08 1.22E+08 8.08 ' '
5 8.30E+07 8.80E+07 8.55E+07 7.93
6 1.91E+07 1.93E+07 1.92E+07 7.28

 

 

 

 

 

 

 

69

 

Table A.4 ANOVA: single factor, comparing Salmonella counts of Method l,

Method II, and Method III of fine-ground product inoculation.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SUMMARY
Average

Grows Count Sum Log CPU/g Variance
Methodl
Salmonella 6 44.99435 7.499059 0.189311
counts
Method II
Salmonella 6 46.24000 7.706667 0.016227
counts
Method III
Salmonella 6 47.36575 7.894291 0.655134
counts
ANOVA

Source of

Variation SS df MS F P value F crit
Between Groups 0.469026 2 0.234513 0.81743 0.4603 3.68232
Within Groups 4.303359 15 0.286891
Total 4.772384 1 7

 

 

 

 

The three different methods do not show a statistical difference in the

counts of Salmonella on the fine-ground pork product. Method III, the vacuum

method, has the highest standard deviation and involves an extra step to the

procedure. Because Method lll requires more time and would add a higher error

to the study, this method was discarded.

There is very little difference between the Salmonella counts in the

products of Methods I and II. Also, when looking at the standard deviations of

the samples when using these two methods, the distribution of Salmonella

seemed to be homogenous. Therefore, for the purpose of the study, “The Effect

of Beef Product Physical Structure on Salmonella Thermal Inactivation”, Method

70

 

II was used, because this method results in product with a more homogenous
structure as described by the standard deviation.
A.2 Coarse-ground

To determine whether Method ll of the fine-ground product was
applicable to the coarse-ground product, the homogeneity of the distribution of
Salmonella was tested by taking 5 random samples (5 1 0.05 9) after inoculation
and grinding in the same manner, but through a plate with 16 mm diameter
holes. The enumeration of Salmonella was performed in the same manner as

described in Chapter 3 (Section 3.3.6). Results are reported in Table A5.

71

Table A5 Salmonella counts (mean 1 standard deviation) in 5 random samples

of coarse-ground pork, which were prepared by passing the pork twice through

the grinder (plate with 16 mm diameter holes).

 

 

 

 

 

 

 

 

Sample Duplicate plates Average Log Average S Log
CFU/g CFU/g CFU/g CFU/g Log CFU/g CFUIL
1 1.69E+07 1 .50E+07 1 .60E+07 7.20
2 5.80E+07 6.20E+07 6.00E+07 7.78
3 7.80E+07 6.20E+07 7.00E+07 7.85 7.70 0.28
4 7.00E+07 8.20E+07 7.60E+07 7.88
5 6.30E+07 5.70E+07 6.00E+07 7.78

 

 

 

 

 

 

 

 

The distribution of Salmonella seems to be homogenous throughout the

coarse-ground product.

72

APPENDIX B

RAW DATA, k AND D VALUES OF SALMONELLA THERMAL INACTIVATION
ON WHOLE-MUSCLE, COARSE-GROUND, FINE-GROUND, AND

PUREED BEEF PRODUCTS

Table 8.1 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of whole-muscle beef.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . Ln N/No
Time (min) Test 1 Test 2
0.0 0.00 0.00
0.5 -0.07 -0.50
1.0 -‘I .14 -0.28
1.5 -1.33 -1.34
2.0 -3.63 -1.83
2.5 -1.38 -3.04
3.0 -3.38 -2.65
3.5 -1.97 -2.05
4.0 -4.77 -2.25
4.5 -5.59 -2.76
5.0 -4.33 -4.51
Negative Slope (k value, min") 1.01 0.72

 

 

 

Table 8.2 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of coarse-ground beef.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . Ln N/No
Time (min) Test 1 Test 2
0.0 0.00 0.00
0.5 0.33 -1.17
1.0 -0.74 ~0.79
1.5 -'I .13 -1.21
2.0 -1.72 -2.52
2.5 -2.78 -2.58
3.0 -4.05 -5.20
3.5 -5.72 -6.98
4.0 -7.17 -7.66
4.5 -7.83 -9.15
_Nggative Slope (R value, min")) 1.91 2.07

 

 

 

73

Table 8.3 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of fine-ground beef.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . Ln N/No

Time (mm) Test 1 Test 2
0.0 0.00 0.00

0.5 -0.72 -0.86

1.0 -0.99 -1.42

1.5 -1.29 -1.89

2.0 -1.69 -2.65

2.5 -3.90 -4.74

3.0 -5.46 -6.92

3.5 -7.25 -6.94
Negflve Slope (k value, min") 1.99 2.13

 

 

Table 8.4 Data for Ln (N/No, Salmonella survival ratio) vs. time at 60°C, and the

k values for each replicate of pureed beef.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . Ln N/No

T'me (mm) Test 1 Test 2 Test 3
0.0 0.00 0.00 0.00

0.5 -0.66 -0.88 -0.25

1.0 -1.48 -1.55 -0.60

1.5 -2.08 -1.86 -1.56

2.0 -3.49 -3.07 -2.62

2.5 -4.68 -4.70 -4.01

3.0 *ND -4.58 -4.02

3.5 -8.33 -5.78 -4.32

4.0 -8.95 -7.78 -5.09

4.5 -9.60 -9.10 *ND

5.0 -8.71 -8.92 -8.06
Negative Slope (k value, min") 2.12 1.92 1.56

 

 

*ND: sample was not under detection limit

74

Table 8.5 First-order inactivation constant, k, and D values at 60°C, calculated
by linear regression of Salmonella survival data for whole-muscle, coarse-

ground, fine-ground, and pureed beef.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

D Average
Beef product . .1 Average S S
k(min ) . -1 . -1 value Dvalue .
structure k(min ) (min ) (min) (min) (min)
1.01 2.28
Whole-muscle 0.72 0.87 0.20 3.19 2.73 0.64
Coarse- 1 .91 1 .20 ,
ground 2.07 1.99 0.11 1.11 1.16 0.07
. 1.99 1.16
Fine-ground 2. 1 3 2.06 0.10 1.08 1.12 0.06
2.12 1.09
Pureed 1.92 1.87 0.29 1.20 1.26 0.20
1.56 1.48

 

 

 

 

 

 

 

 

 

75

APPENDIX C

OVEN COOKING SCHEDULES TEMPERATURE PROFILES CURVES

80‘

 

6
2.1
2
D
‘8'
8
5
1— l

20 1

10 l

0 500 1000 1500 2000
Time (s)

Figure 0.1 Temperature curves of convection oven air in cooking schedule A.

70-

 

Temperature (°C)

0‘ e -

0 500 1000 1500 2000 2500 3000 3500
Time (s)

Figure 02 Temperature curves of convection oven air in cooking schedule 8.

76

 

Temperature (°C)

0 1000 2000 3000 4000 5000
Time (s)

Figure 03 Temperature curves of convection oven air in cooking schedule C.

NOD-501
0000

Temperature (°C)

—L
O

0 1 000 2000 3000 4000 5000 6000
Time (s)

C

Figure C.4 Temperature curves of convection oven air in cooking schedule D.

77

CO

 

NOD-#01
0

Temperature (°C)
o

_L
O

0 1111 1 1.1- 1 1 .1 1 1.1 1 1-1.11. 1 1 1.1 1 1 1l
1 000 2000 3000 4000 5000 6000 7000 8000
Time (s)

Figure 0.5 Temperature curves of convection oven air in cooking schedule E.

O
I

111111

0

 

 

Temperature (°C)
-t N OD # 01
O O O O
l

0 1 00 200 300 400 500 600 700
Time (s)

Figure C.6 Temperature curves of convection oven air in cooking schedule F

(control).

78

REFERENCES

[AMI] American Meat Institute. 2002. Pathogen Control in Ground Beef.
Arlington: American Meat Institute. Available from:
http://www.meatami.com/Content/NaviqationMenu/PressCenter/FactSheet
s lnfoKits/FactSheetPathoqenControlinGroundBeef.pdf. Accessed Dec 7,
2006.

[AMI] American Meat Institute. 2003. Overview of US. Meat and Poultry
Production and Consumption. Arlington: American Meat Institute.
Available from:
http://wwwmeatami.com/Content/NaviqationMenu/PressCenter/FactSheet
s lnfoKits/FactSheetMeatProductionandConsumption.pdf. Accesed July
31, 2006.

[AMI] American Meat Institute. 2006. Background Statistics: US. Beef and Cattle
Industry. Available from:
http://wwwers.usda.qov/news/BSECoveraqe.htm. Accessed July 30,
2006.

[FSIS] United States Food Safety and Inspection Service. 1998-2001. Progress
Report on Salmonella Testing of Raw Meat and Poultry Products, 1998-
2001. United States Department of Agriculture. Available from:
http://www.fsis.usda.cpv/OPHS/haccp/salm4vear.pdf. Accessed July 31 ,
2006.

[FSIS] United States Food Safety and Inspection Service. 2001. Performance
Standards for the Production of Processed Meat and Poultry Products;
Proposed Rule. In: Register F, editor: Available from:
http://www.fsis.usdagov/oppde/rdad/frpubs/97-013p.htm. Accessed July
31, 2006. p 12589-636.

[FSIS] United States Food Safety and Inspection Service. 2003. Meat, Poultry,
and Egg Products Inspection

2000 Report of the Secretary of Agriculture to the US. Congress. Available from:
http://www.isis.usda.qov/oa/pubs/rtc2000/rtc2000chap3.htm. Accessed
December 8, 2006.

[FSIS] United States Food Safety and Inspection Service, Science and
Technology, Microbiology Division. 1994. Nationwide Beef Microbiological
Baseline Data Collection Program: cows and bulls: October 1992-1993.
Washington, USA: USDA. Available from:

79

http://www.fsis.usda.qov/OPHS/baseline/steer1 .pdf,
http://www.fsis.usda.qov/OPHS/baseline/steer2.pdf,
http://www.fsis.usda.qov/OPHS/baseline/steer3.Ddf. Accessed July 31,
2006.

Abdul-Raouf UM, Beuchat LR, Ammar MS. 1993. Survival and growth of
Escherichia coliO157zH7 in ground, roasted beef as affected by pH,
acidulants, and temperature. Applied and Environmental Microbiology
59(8):2364-8.

Aberle ED, Forrest JC, Gerrard DE, Mills EW, Hendrick H8, Judge MD, Merkel
RA. 2001. Principles of Meat Science. Fourth Edition ed. Dubuque, Iowa:
KendalVHunt Publishing Company. 1-6, 126-7, 49-51, 55-77, 90-95 p.

Ahmed NM, Conner DE, Huffman DL. 1995. Heat-resistance of Escherichia coli
O157:H7 in meat and poultry as affected by product composition. Journal
of Food Science 60(3):606-10.

Anderson BA, Hoke IM. 1990. Composition of Foods: Beef Products: United
States Department of Agriculture. 9-15 p.

Association of Official Analytical Chemists A. 1996. Official methods of analysis.
16th ed. Arlington, Va: AOAC.

Bell C, Kyriakides A. 2002. Salmonella: A practical approach to the organism and
its control in foods. Oxford, UK: Blackwell Science. 1-20, 116 p.

Bell C, Neaves P, Williams AP. 2005. Food Microbiology and Laboratory
Practice. 1st ed. Oxford, UK: Blackwell Publishing. 299, 306 p.

Blackburn Cd-W, Curtis LM, Humpheson L, Billon C, McClure PJ. 1997.
Development of thermal inactivation models for Salmonella enteritidis and
Escherichia coli0157zH7 with temperature, pH and NaCl as controlling
factors. lntemational Journal of Food Microbiology 38(1):31-44.

Bunning VK, Crawford RG, Tierney JT, Peeler JT. 1990. Therrnotolerance of
Listeria monocytogenes and Salmonella lyphimurium after Sublethal Heat
Shock. Applied and Environmental Microbiology 56(10):3216-9.

80

Buzby JC, Roberts T, Lin C-TJ, MacDonald JM. 1996. Bacterial Foodbome
Disease: Medical Costs and Productivity Losses. Food and Consumer
Economics Division, Economic Research Service, US. 1-81 p.

Carlson TR. 2002. Effect of Water Activity and Humidity on the Thermal
Inactivation of Salmonella During Heating of Meat [MSc thesis]. East
Lansing: Michigan State University. 36-7 p.

Carlson TR, Marks BP, Booren AM, Ryser ET, Orta-Ramirez A. 2005. Effect of
water activity on thermal inactivation of Salmonella in ground turkey.
Journal of Food Science 70(7):M363-M6.

Crutchfield SR. 1999. New Federal Policies and Programs for Food Safety. Food
Review. p 2-5.

Davis CG, Lin B-H. 2005. Factors Affecting U.S. Beef Consumption. USDA.
Report nr LDP-M-135-02.

Doyle ME, Mazzotta AS. 2000. Review of studies on the thermal resistance of
salmonellae. Journal of Food Protection 63(6):779-95.

Doyle MP, Beuchat LR, Montville TJ. 1997. Food Microbiology: Fundamentals
and Frontiers. Washington DC: ASM Press. 83-96, 129-52 p.

Elmossalami E, Wassef N. 1971. Penetration of some microorganisms in meat.
Zentralblatt fiir Veterinarmedizin Reihe B Journal of veterinary medicine
Series B 18(5):329-36.

Frenzen PD, Riggs L, Buzby JC, Breuer T, Roberts T, Voetsch D, Reddy S, the
FoodNet Working Group. 1999. Salmonella Cost Estimate Update Using
FoodNet Data. Food Review 22(2):10-5.

Ghazala S, Coxworthy D, Alkanani T. 1995. Thermal kinetics of Streptococcus
faecium in nutrient broth/sous vide products under pasteurization
conditions. Journal of Food Processing and Preservation 19(4):243—57.

Gill CO, Penney N. 1982. Bacterial Penetration Of Muscle Tissue. Journal of
Food Science 47(2):690-1.

81

Goepfert JM, lskander IK, Amundson CH. 1970. Relation of the heat resistance
of Salmonella to the water activity of the environment. [Microorganisms].
Applied Microbiology 19(3):429-33.

Guo 8, Marks BP. 2005. Generalized Modeling for Thermal Inactivation of
Salmonella in Meat and Enhancement of the AMI Process Lethality
Spreadsheet. Baltimore, Maryland USA: 51$t lntemational Congress of
Meat Science and Technology. 1-4 p.

Hansen NH, Riemann H. 1963. Factors affecting the heat resistance of
nonsporing organisms. Journal of Applied Bacteriology 26:314-33.

Humphrey T, Richardson NP, Statton KM, Rowburry RJ. 1993. Effects of
temperature shift on acid and heat tolerance in Salmonella enten'ditis
phage type 4. Applied and Environmental Microbiology 59:3120-2.

Humphrey T, Slater E, McAlpine K, Rowbury R, Gilbert R. 1995. Salmonella
enteritidis phage type 4 isolates more tolerant of heat, acid, or hydrogen
peroxide also survive longer on surfaces. Applied and Environmental
Microbiology 61 (8):3161-4.

Jay J. 2000. Modern Food Microbiology. New York: Chapman and Hall. 279 p.

Juneja VK, Eblen BS, Marks HM. 2001. Modeling non-linear survival curves to
calculate thermal inactivation of Salmonella in poultry of different fat
levels. lntemational Journal of Food Microbiology 70(1/2):37-51.

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. Journal of Food Science 66(1):146-52.

Juneja VK, Klein PG, B.S.Marmer. 1998. Heat shock and thennotolerance of
Escherichia coli O157:H7 in a model beef gravy system and ground beef.
Journal of Applied Microbiology 84:677-84.

Juneja VK, Novak JS. 2003. Adaptation of foodbome pathogens to stress from
exposure to physical intervention strategies. In: Youself AE, Juneja VK,
editors. Microbial stress adaptation and food safety. Boca Raton, FL: CRC
Press. p 31-53.

82

Lihono MA, Mendonca AF, Dickson JS, Dixon PM. 2001. Influence of sodium
pyrophosphate on thermal inactivation of Listeria monocytogenes in pork
slurry and ground pork. Food Microbiology 18(3):269-76.

Mackey B, Derrick C. 1986. Elevation of the heat resistance of Salmonella
typhimurium by sublethal heat shock. Journal of Applied Bacteriology
61 (5):389-93.

Mackey BM, Derrick C. 1990. Heat shock protein synthesis and therrnotolerance
in Salmonella typhimurium. Journal of Applied Bacteriology 69:373-83.

Mackey 8M, Derrick CM. 1987. The effect of prior heat shock on the
therrnoresistance of Salmonella thompson in foods. Letters in Applied
Microbiology 5:115-8.

Merck. 1992. The Merck Manual and General Medicine. 16th ed. Rahway, N.J.:
Merck & Co., Inc.

Montville TJ, Matthews KR. 2005. Food Microbiology An Introduction.
Washington DC: ASM Press. 85-99 p.

Murano EA, Pierson MD. 1993. Effect of heat shock and incubation atmosphere
on injury and recovery of Escherichia coli O157:H7. Journal of Food
Protection 562568-72.

Murphy RY, Marks 8P, Johnson ER, Johnson MG. 1999. Inactivation of
Salmonella and Listeria in ground chicken breast meat during thermal
processing. Journal of Food Protection 62(9):980-5.

Murphy RY, Marks 8P, Johnson ER, Johnson MG. 2000. Thermal inactivation of
Salmonella spp. and Listeria innocua in the chicken breast patties
processed in a pilot-scaIe-air-convection oven. Journal of Food Science
65(4):706-1 0.

North American Meat Processors Association. 1997. The Meat Buyers Guide. 1st
ed. Reston, VA: North American Meat Processors Association. 13 p.

O'Donovan- Vaughan C, Upton M. 1999. Food Safety: The combined effect of
reduced water activity (aw) and heat on the survival of Salmonella

83

Typhimurium. Tuijtelaars A, Samson R, Rombouts F, Noterrnans S,
editors. Wageningen, Netherlands: Foundation Food Micro '99. 85-7 p.

Orta-Ramirez A, Marks B-P, Warsow C-R, Booren A-M, Ryser E-T. 2005.
Enhanced thermal resistance of Salmonella in whole muscle compared to
ground beef. Journal of Food Science 70(7):M359-M62.

Oteiza JM, Giannuzzi L, Califano AN. 2003. Thermal inactivation of Escherichia
coli O157:H7 and Escherichia coli isolated from morcilla as affected by
composition of the product. Food Research lntemational 36(7):703-12.

Pearson AM, Tauber FW. 1984. Processed Meats. Second Edition ed. Westport,
Connecticut: AVl Publishing Company, Inc. 1-4, 192-5, 69-84, 87-99 p.

Quintavalla S, Larini S, Mutti P, Barbuti S. 2001. Evaluation of the thermal
resistance of different Salmonella serotypes in pork meat containing
curing additives. lntemational Journal of Food Microbiology 67(1/2):107-
14.

Ranken MD. 2000. Handbook of Meat Product Technology. Malden, MA:
Blackwell Science. 105-15, 19-24 p.

Singh RP, Heldman DR. 1998. Introduccion a la Ingenieria de los Alimentos.
Garcia FJ, Bacaicoa PG, translator. 2nd ed. Zaragoza, Spain: Editorial
Acribia, S.A. 157-62 p.

Smith SE, Maurer JL, Dita-Ramirez A, Ryser ET, Smith DM. 2001. Thermal
inactivation of Salmonella spp., Salmonella typhimurium DT104, and
Escherichia coli O157:H7 in ground beef. Journal of Food Science
66(8):1 164-8.

Stasiewicz MJ, Marks BP, Orta-Ramirez A. Modeling the effect of prior thermal
history on the thermal inactivation rate of Salmonella in turkey; 2005; New
Orleans, Louisiana.

Stasiewicz MJ, Marks BP, Orta-Ramirez A. 2006. Unpublished data. Modeling
the Effect of Prior Sublethal Thermal History on the Thermal Inactivation
Rate of Salmonella in Ground Turkey. Michigan State University. East
Lansing, Mich.

84

Stumbo CR. 1973. Thermobacteriology in food processing. New York: Academic
Press. 79-104 p.

Tauxe RV. 1991. Salmonella: A postmodern pathogen. Journal of Food
Protection 54(7):563-8.

Thomson JE, Bailey JS, Cox NA. 1979. Phosphate and heat treatments to
control Salmonella and reduce spoilage and rancidity on broiler carcasses.
Poultry Science 58(1):139-43.

Tomlins R, Ordal Z. 1976. Thermal injury and inactivation in vegetative bacteria.
Skinner F, Hugo W, editors. New York: Academic Press.

Tripuraneni MC. 2006. Fat Transport Properties and Mechanisms During
Cooking of Ground Meat [MSc thesis]. East Lansing: Michigan State
University. 36-7 p.

Troller JA. 1986. Water relations of foodbome bacterial pathogens -- an updated
review. Journal of Food Protection 49(8):656-70

USDA (United States Department of Agriculture). 2006. Animal Production and
Marketing Issues. Retreived July 31, 2006.
http://www.ers.usda.qov/briefinq/animalproducts/.

USDA (United States Department of Agriculture). 2006. Cattle.
httpzl/wwwers.usda.qov/briefinq/cattle/. Last accessed July 31, 2006.

Veeramuthu GJ, Price JF, Davis CE, Booren AM, Smith DM. 1998. Thermal
inactivation of Escherichia coli O157:H7, Salmonella senftenberg, and
enzymes with potential as time-temperature indicators in ground turkey
thigh meat. Journal of Food Protection 61 (2):171-5.

Velasquez A. 2006. Thermal Resistance and Migration of Salmonella spp. into
Marinated Pork Products [MSc thesis]. East Lansing, MI: Michigan State
University. 19-38 p.

Warsow CR. 2003. Penetration of Salmonella spp. into Whole Muscle Turkey
Breasts During Vacuum Tumbling Marination [MSc thesis]. East Lansing,
MI: Michigan State University. 41-95 p.

85

Youself AE, Courtney PD. 2003. Basics of stress adaption and implications in
new-generation foods. In: Youself AE, Juneja VK, editors. Microbial stress
adaption and food safety. Boca Raton, FL: CRC Press. p 1-30.

86

 

Illl I IIIII