sir... 5.3%.... taxis, ..
. ., my: .,.H.n...n.mw.ww,. _.
athiwmvarlfifi .3. 5. .fi . V .Mmfia

umwwm x“. x. a. 2.
—

“iii.

,1
I

a. .
9:44
. m
.. .
.mma..w.&h.n?.. . ‘ . . . . , . ‘ 13.}. fl.
.. .. 4 . . ,
£me ‘
2 «£9 a
r f L.
U 1

uJJ‘nQ- . . , . . . 1v...
. . an: . .
3a.: ..

. .
.1
E 5....
EYE...
.. .
.

.p...

:3

 

 

 

~l.
.. .\...
xii...

 

 

 

.3... in:
H.315}. Va ...:_.., t. .n .f 2». :1:

 

 

This is to certify that the

thesis entitled

Environmental Effects on the Thermal Resistance of
Salmonella, Escherichia coli 0157:H7, and Triose
Phosphate Isomerase in Ground Turkey and Beef

 

presented by

Jennifer L. Maurer

has been accepted towards fulfillment
of the requirements for

Master's degree in Food Science

 

 

£04m M. Md.

Major professor

Date 3‘! ’01

07639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIIRARY
Michigan State
University

 

 

 

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
AUG ‘0 33.2% ‘~‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

601 cJCIRCJDaieDuepes-ois

 

 

ENVIRONMENTAL EFFECTS ON THE THERMAL RESISTANCE OF
SALMONELLA, ESCHERICHIA C OLI 01571H7, AND TRIOSE PHOSPHATE
ISOMERASE IN GROUND TURKEY AND BEEF

By

Jennifer L. Maurer

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

2001

ABSTRACT
ENVIRONMENTAL EFFECTS ON THE THERMAL RESISTANCE OF
SALMONELLA, ESCHERICHIA COLI 01572H7, AND TRIOSE PHOSPHATE
ISOMERASE IN GROUND TURKEY AND BEEF
By

Jennifer L. Maurer

On January 6, 1999, the USDA-FSIS published their final ruling on performance
standards for meat products. Lethality standards for both turkey and beef products are
based on the thermal destruction of Salmonella. Processors must prove that their thermal
processes are adequate in challenge studies using meat inoculated with Salmonella.
Because of obvious safety and health concerns, it is not feasible to bring Salmonella into
a meat processing facility. Therefore, alternative approaches to verify adequate cooking
of meat in processing facilities must be identified. The objective of this study was to
determine if triose phosphate isomerase (TPI) could be used as a time-temperature
integrator (TTI) in turkey and beef products to verify the adequacy of any thermal
process.

Except for the treatments in which phosphate was added, the z-values for TPI and
S. Senfienberg were within O.5°C of each other. For turkey products, TPI showed good
potential to be a TTI. The effect of fat content in ground beef on the thermal inactivation
kinetics of TPI, a Salmonella cocktail, and E. coli 01572H7 was studied. TPI was more
temperature dependent than E. coli 01572H7 and the Salmonella cocktail. A
mathematical model will need to be developed and tested before TPI can be used as a TTI

in ground beef.

ACKNOWLEDGEMENT

I would like to thank my major professor, Dr. Denise Smith, for her support and
guidance during my years at Michigan State University. The training in the class room as
well as in the laboratory have and will continue to help my career in the food science
industry.

I would also like to thank my committee members: Dr. Booren for his help and
know-how in the meat laboratory and Dr. Ryser for his help in microbiology.

A huge thanks to my lab mates: Dr. Alicia Orta (for all her help with
microbiology and beyond), Dr. Jamie Prater, Carolyn Ross, Jin-Shan Shie, Sarah Smith,
Dr. Manee Vittayanont, and Dr. Virginia Vega-Wamer.

Finally, I would like to thank Matt and my family for all the love and support

throughout the years and my time at MSU.

This research was supported by the Michigan State University Crop and Food
Bioprocessing Center and the Cooperative State Research, Education and Extension
Service, United States Department of Agriculture under agreements number 96-3 5201 -
3343 and 99-343 82-8476. Any opinions, findings, conclusions, or recommendations
expressed in this document are those of the author and do not necessarily reflect the view

of the United States Department of Agriculture.

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. vi
CHAPTER 1: INTRODUCTION .................................................................. 1
CHAPTER 2: LITERATURE REVIEW ......................................................... 5
2.1. United States Department of Agriculture-Food Safety and Inspection Service
thermal processing guidelines .......................................................... 5

2.2. Salmonella and Escherichia coli 01572H7 ......................................... 8

2.2.1. Salmonella .................................................................. 8

2.2.2. Escherichia coli 01572H7 ........................... . .................... 9

2.3. Determination of adequate thermal processing ................................... 10

2.3.1. In situ method .............................................................. 10

2.3.2. Physical-mathematical method ............................................. 11

2.3.3. Time temperature integrators ............................................ 11

24. Factors affecting the activity of triose phosphate isomerase ..................... 13

2.5. Format of thermal death time testing ............................................... 14

2.5.1. Thermal death time tubes ................................................ 14

2.5.2. Capillary tubes ........................... . ................................. 14

2.5.3 Plastic pouches ............................................................ 15

2.6. Thermal death time studies ......................................................... 16

2.6.1. Beef ........................................................................ 16

2.6.2. Turkey ..................................................................... 19

2.7. Environmental effects on microorganisms ......................................... 19

2.7.1. The effect of fat ........................................................... 23

2.7.2. The effect ofadditives....................................... .............23

2.7.3. The effect of pH .............................. . ........................... 24

CHAPTER 3: Salmonella Senftenberg, Escherichia coli OlS7:H7, and triose phosphate
isomerase, a possible time temperature integrator, in ground turkey meat ................... 26
3.1. ABSTRACT ........................................................................... 26

3.2. INTRODUCTION .................................................................... 27

3.3. MATERIALS AND METHODS ................................................... 29

3.3.1. Meat formulations ........................................................ 29

3.3.2. Thermal inactivation experiments ....................................... 31

3.3.3. Bacterial cultures and inoculation ...................................... 32

3.3.4. Protein extraction for TPI activity ..................................... 32

3.3.5. TPI activity determination .............................................. 33

3.3.6. Bacterial counts .......................................................... 33

3.3.7. Data analysis .............................................................. 34

3.4. RESULTS AND DISCUSSION .................................................... 35

3.4.1. Proximate composition .................................................. 35

3.4.5
3.4.6.

3.4.7.

3.4.8.

. Effect of muscle type and fat content on the thermal inactivation of

TPI.. ..35

. Effect Of muscle type and fat content on the thermal inactivation of

Salmonella ............................................................... 3 9

. Effect of muscle type and fat content on the thermal 1nactivation of E

coli OIS7:H7. 4..1
Effect of salt and phosphate on the thermal inactivation Of TPI... 43
Effect of salt and phosphate on the thermal inactivation of
Salmonella” ..43
Effect of salt and phosphate on the thermal inactivation Of E. coli
OIS7:H7 ......... . ......................................................... 46
Comparison of z-values of TPI. S. Senftenberg and E. coli
OIS7:H7 ................................................................... 46

3.5. CONCLUSIONS48

CHAPTER 4: Salmonella spp., Escherichia coli 0 1 57:H7, and triose phosphate isomerase,

a possible time temperature integrator, in ground beef ........................... 50
ABSTRACT ........................................................................... 50

42 INTRODUCTION .. ..51

4 3 MATERIALS AND METHODS ................................................... 53

4.3.1. Ground beef preparation ................................................ 53

4.3.2. Thermal inactivation experiments ...................................... 53

4.3.3 Protein extraction for TPI activity ..................................... 54

4.3.4. TPI activity determination .............................................. 55

4.3.5. Bacterial cultures and inoculation ...................................... 55

4.3.6. Bacterial counts ........................................................... 56

4.3.7. Identification of Salmonella survivors ................................. 57

4.3.8. Data analysis .............................................................. 57

4.4. RESULTS AND DISCUSSION ................................................... 58

4.4.1. Proximate analysis ....................................................... 58

4.4.2. Thermal resistance of E. coli OIS7:H7 ................................ 58

4.4.3. Thermal resistance of Salmonella cocktail ............................ 61

4.4.4. Survivor identification ................................................... 62

4.4.5. Thermal resistance of TPI ............................................... 65

4.5. CONCLUSIONS ..................................................................... 67
CHAPTER 5: CONCLUSIONS ................................................................. 69
CHAPTER 6: FUTURE RESEARCH .......................................................... 71
REFERENCES ................................................................................ 72

LIST OF TABLES

Table 2.1. Permitted heat-processing temperature/time combinations for fiilly-cooked

patties .......................................................................................... 6
Table 2.2. Review of D- and z-values of thermal death time studies in beef .............. 18
Table 2.3. Review of D- and z-values of thermal death time studies in turkey ............ 20

Table 2.4. Review of D- and z-values of thermal death time studies in turkey and beef with

varying fat contents ........................................................................... 22
Table 3.1. Brine solutions of high fat turkey breast meat (HFB) ............................ 30
Table 3.2. Proximate composition and pH of turkey meat .................................... 36

Table 3.3. D-values and regression analysis for triose phosphate isomerase (TPI) in
turkey breast and thigh meat .............................................................. 37

Table 3.4. The z-values of triose phosphate isomerase (TPI), S. Senftenberg, and E. coli
OIS7:H7 in turkey breast and thigh meat ................................................ 38

Table 3.5. D-values and regression analysis for S. Senfienberg in turkey breast and thigh
meat .......................................................................................... 40

Table 3.6. D-values and regression analysis for E. coli 0157:H7 in turkey breast and
thigh meat ................................................................................... 42

Table 3.7. D-values and regression analysis for triose phosphate isomerase (TPI) in high fat
turkey breast meat (HFB) containing NaCl and phosphate (PP) ........ 44

Table 3.8. D-values and regression analysis for S. Senftenberg in high fat turkey breast
meat (HFB) containing NaCl and phosphate (PP) ..................................... 45

Table 3.9. D-values and regression analysis for E. coli 01 57:H7 in high fat turkey breast
meat (HFB) containing NaCl and phosphate (PP) ..................................... 47

Table 4.1. D- and z-values and regression parameters of E. coli OlS7:H7 in low (4.8% fat)
and high (19.1% fat) fat ground beef. ............................................... 59

Table 4.2. D- and z-values and regression parameters of Salmonella cocktail in fresh and
frozen inoculated ground beef ( 1 9. 1% fat). Cultures were inoculated when in the log

and stationary phases ................................................................... 60

Table 4.3. Reactions of API 20E test strips to Salmonella cocktail strains ................ 64

Table 4.4. D- and z-values and regression parameters of triose phosphate isomerase in low
(4.8% fat) and high (191% fat) fat ground beef66

vii

CHAPTER 1
INTRODUCTION

The most prevalent foodbome pathogens associated with meat and poultry
products are: Salmonella, Escherichia coli, C ampylobacter, and Staphylococcus (Bean
and Griffin, 1990). According to the Centers for Disease Control and Prevention (CDC),
there are approximately 40,000 reported cases of salmonellosis in the United States per
year (CDC, 1999b). Because many cases are unreported, the actual number may be as
much as 20 times higher. Individuals with salmonellosis develop symptoms of diarrhea,
fever, and cramps, which occur 12 to 72 hours after exposure. The illness usually lasts 4
to 7 days, and most people recover without treatment. However, in the very young,
elderly, or those who are immunocompromised, this infection can be very dangerous. Of
these cases, approximately 1,000 prove to be fatal (CDC, 1999b).

In recent years Escherichia coli 01572H7, an enterohemorrhagic strain, has
emerged as a serious threat to the safety of meat products. An estimated 10,000 to
20,000 cases of infection occurred in the United States in 1999 (CDC, 1999a). Since
1982, undercooked ground beef has been the most common source of E. coli 01 57:H7
infection (CDC, 1999a). Between November 1992 and February 1993 there were 700
cases reported of E. coli 01 57:H7 and 4 deaths in the western United States due to
ingestion of undercooked hamburgers from a fast food chain. In November 1994, 17
cases of E. coli 01 57:H7 infection in Washington State and California were traced to the
ingestion of contaminated dry fermented salami (CDC, 1995). Illness due to the

ingestion of E. coli 01 57:H7 has also been associated with raw milk (Padyhe and Doyle,

1992), apple cider (Zhao et al., 1993), turkey rolls (Carter et al., 1987), and potatoes
(Griffin and Tauxe, 1991).

The United States Department of Food Safety and Inspection Service (USDA-

F SIS) has approved irradiation of meat products to control disease-causing
microorganisms, however, thermal processing is the most common method to eradicate
pathogens. The USDA requires that meat be cooked to a specific endpoint temperature
or to specific temperatures and held for specific times. New regulations allow for meat to
be cooked by any thermal process that would ensure a 7 logm to a 5 logo reduction in
Salmonella based on the type of meat. These regulations enhance the flexibility of the
processor by allowing them to choose the thermal process to adequately cook meat.
However, there is currently no good way to test the adequacy of a thermal process “afier
the fact.”

Time-temperature integrators (TTI) have been used to verify the accuracy of
various thermal processes. A TTI is a marker, such as a protein or enzyme, that responds
to time-temperature history by undergoing an irreversible change that mimics the changes
in a target attribute, such as a pathogen, undergoing the same thermal process (Hendrickx
et al., 1995). The z-value (temperature increase needed to decrease the thermal death
time or D-value by 90%) of the target attribute can be used to select a TTI, when both the
marker and the target attribute exhibit linear inactivation kinetics. Previous studies have
shown that triose phosphate isomerase (TPI), an enzyme found in meat, has the potential
to be a TTI because the z-value is close to the z-value for Salmonella Senftenberg. In
ground beef, the z-value of TPI was 556°C, which was close to the z-values of both E.

coli 01 57:H7 and Salmonella Senftenberg which were 559°C and 625°C, respectively

(Orta-Ramirez et al., 1997). In ground turkey thigh meat the z-values for TPl. E. coli
01572H7, and S. Senftenberg were 5.8, 6.0, and 56°C, respectively (Veeramuthu et al.,

1998).

The overall goal of this study was to determine if TPI can be used as an

endogenous TTI in meat products by examining how environmental factors affect the

inactivation kinetics of TPI. The specific objectives were:

1 .

Determine the effect of muscle type, fat content, NaCl, and phosphate in ground
turkey meat on the thermal inactivation kinetics of Salmonella Senftenberg, E. coli
OIS7:H7, and TPI activity.

Determine the thermal inactivation kinetics of E. coli OIS7:H7, TPI, and an 8-strain
Salmonella cocktail in ground beef.

Determine the effect of freezing and growth phase on the thermal inactivation kinetics

of the Salmonella cocktail in ground beef.

CHAPTER 2

LITERATURE REVIEW

2.1 UNITED STATES DEPARTMENT OF AGRICULTURE-FOOD SAFETY
AND INSPECTION SERVICE (USDA-FSIS) THERMAL PROCESSING
GUIDELINES
The USDA-FSIS has issued thermal processing regulations to ensure the safety of

various cooked and partially cooked meat and poultry products since 1972 (USDA-FSIS,

1999). These regulations included processing schedules for fully — cooked ground beef

patties (Table 2.1) that ranged from holding the patties for 415 at an internal temperature

of 66. 1°C to holding for 105 at 694°C. The regulations also covered roast beef, which
included holding the internal temperature of the meat for 112 min at 544°C to 105 at

71 . 1°C.

On May 2, 1996, the USDA-FSIS proposed performance standards for certain
meat and poultry products as an alternative to “command-and-control” regulations
(USDA-FSIS, 1999). The products affected by the proposed performance standards
were: cooked beef, roast beef, and cooked corned beef; fully cooked, partially cooked,
and char-marked uncured meat patties; and certain firlly and partially cooked poultry
products. Three performance standards were addressed: lethality, stabilization, and
handling. It was proposed that cooked poultry products must meet the lethality standard
of a 7 loglo destruction in Salmonella. A 7 logm destruction means that the thermal
process should reduce the Salmonella content by a factor of 10 million. Ready-to-eat

cooked beef, roast beef, and cooked corned beef products must be processed

Table 2.1. Permitted heat-processing temperature/time combinations for firlly-cooked

ground beef patties

 

 

Minimum internal temperature Minimum holding time
at the center of each patty after temperature is reached

(”C(°F)) (S)
66.1 (151) 41
66.7 (152) 32
67.2 (153) 26
67.8 (154) 20
68.3 (155) 16
68.9 (156) 13
69.4 (157) 10

 

Federal Register (USDA-FSIS, 1999)

to achieve a 7 logw destruction in Salmonella. The F SIS proposed a 5 logm reduction for
Salmonella in fully cooked, uncured meat patties (USDA-FSIS, 1999). The stabilization
performance standard required that products be stabilized to prevent germination or
grth of spore-forming bacteria such as C lostridium botulinum, and growth of C
perfiingens should be limited to a l-log multiplication. The handling performance
standard called for safe handling, which would prevent contamination of the product afier
the lethality and stabilization standards had been met.

On January 6, 1999, the USDA-FSIS published their final ruling on performance
standards for meat products. For ready-to-eat cooked beef, roast beef, and cooked corned
beef, the lethality performance standard was a 6.5-logm reduction in Salmonella. Fully
cooked poultry products must be processed to achieve a 7-logm reduction in Salmonella
(USDA-F SIS, 1999). The USDA-FSIS has not issued a final ruling on the lethality
performance standards for uncured meat patties. The current proposal requires a 5-log10
reduction in Salmonella, however, a higher lethality standard may be issued.

Even though the microorganism of concern in many underprocessed meat
products is E. coli 01 57:H7, Salmonella is more heat resistant; therefore, by destroying
Salmonella, E. coli would also be destroyed. Listeria monocytogenes, another
microorganism of concern, has also been implicated in many food outbreaks. Although it
has been shown to be more heat resistant than Salmonella, contamination by Listeria is
indicative of recontamination of the food product after processing.

These new regulations offer more flexibility to the producer of ready-to-eat meat

products. The processor must verify, by a challenge study, that their process meets the

lethality performance standard. The USDA requires that challenge studies be conducted
using a cocktail of various Salmonella serovars in a way that accurately portrays the
processing environment. The serovars should contain several heat resistant strains of
Salmonella, as well as strains implicated in a large number of outbreaks (USDA-FSIS,

1999).

2.2 SALMONELLA AND ESCHERICHIA COLI 0157 :H7
2.2.1 Salmonella

There were an estimated 1.4 million cases of Salmonellosis in the United States in
1998 (CDC, 1999b). Two serotypes, S. Enteritidis and S. Typhimurium are responsible
for half of the salmonellosis cases. Salmonella Typhimurium is the most commonly
found foodbome serotype throughout the world (Jay, 1996), and an increasing proportion
of S. Typhimurium strains are showing resistance to multiple antibiotics (CDC, 1999b).
Salmonella Senftenberg is the most heat stable strain Of Salmonella, but rarely causes
foodbome outbreaks. Eggs, poultry, meat, and meat products are the most common
vehicles for contraction of salmonellosis by humans. The FSIS reported that the
combined prevalence in large and small plants from July 1999 to June 2000 of
Salmonella in broiler chickens was 9.9%, ground chicken was 14.4% ground beef was
5.0%, and ground turkey was 30.0% (USDA-FSIS, 2000).

Salmonellae are small, gram-negative, non—sporing rods, which are generally
unable to ferment lactose, sucrose, or salicin. Glucose and other monosaccharides are
fermented, with the production of gas (Jay, 1996). The minimum pH for growth is 4.05

with a maximum of 9.0. The optimum growth pH is between 6.6 and 8.2. (Jay, 1996).

Salmonellae are unable to tolerate a high concentration of NaCl. A brine solution of 9%

is bactericidal.

2.2.2 Escherichia coli 0157:H7

Since the first outbreak in 1982 from contaminated hamburgers, E. coli 0157:H7
has emerged as a pathogen of concern (Riley et al., 1983). In 1998 there were an
estimated 10,000 to 20,000 cases of infection from E. coli 01 57:H7 in the United States
(CDC, 1999a). Most strains of E. coli are found in the intestinal tract of humans and
animals, however, the 01 57:H7 strain can be particularly harmful. About 2-7% of the
infections lead to hemolytic uremic syndrome, and 3-5% of these cases are fatal (CDC,
1999a). Undercooked contaminated ground beef is the most common cause of E. coli
01 57:H7 infection, however, there have been outbreaks associated with raw milk
(Padyhe and Doyle, 1992), apple cider (Zhao et al., 1993), turkey rolls (Carter et al.,
1987), and potatoes (Griffin and Tauxe, 1991).

E. coli OIS7:H7 has been shown to be more sensitive to heat than salmonellae,
but can survive, with little change in number, for 9 months at -20°C in ground beef
patties (Doyle and Schoeni, 1984). The best growth temperature for E. coli OIS7:H7 is
between 30 and 42°C, with 37°C being optimal (Doyle and Schoeni, 1984).

With the outbreak of E. coli 01 57:H7 in apple cider, the acid tolerance of E. coli
01 57:H7 has been studied. E. coli 01 57:H7 survived 56 days in tryptic soy broth that
was acidified to pH _>_ 4.0 using various acids for pH control (Conner and Kotrola, 1995).

The optimum pH for growth of E. coli OIS7:H7 is around pH 7.0.

2.3 DETERMINATION OF ADEQUATE THERMAL PROCESSING

The USDA currently uses three tests to determine the endpoint temperature (EPT)
of cooked meat products. They are: the bovine catalase test (USDA, 1989), the protein
“coagulation test” (USDA, 1986a), and the acid phosphatase activity method (USDA,
1986b). These EPT tests are specific for a given product of the same size and shape.
The bovine catalase test and the protein “coagulation test” are subjective tests that are
dependent on the temperature of the product and do not consider time. The acid
phosphatase activity method does take into account the time/temperature history of the
product, however, results from this method do not correlate with actual EPT (Townsend
and Blankenship, 1989). Because both temperature and time are factors in determining
adequate processing, and the method should correlate with actual EPT of meat products,
these tests are not ideal.

Three different approaches have been developed to evaluate the adequacy of
thermal processing: 1) in situ methods, 2) physical-mathematical approaches, and 3)
Time Temperature Integrators (TTI).

2.3.1 In situ method

In the in situ method, a thermal quality attribute is measured before and after
thermal processing. This test is often used to measure vitamin loss, residual enzyme
activity, and microbial load. The main advantage of the in situ method is that the
adequacy of the thermal process on the variable of interest is accurately known. The
disadvantages of this method are that in some cases (e. g. a lethality of 109), the sample
size would be too large to monitor. It is also laborious, expensive, and time consuming

(Hendrickx et al., 1995).

10

2.3.2 Physical-mathematical method

The most widely accepted alternative to the in situ method is the physical-
mathematical approach. The theory behind this approach is that after the kinetic
parameters of the attribute of interest are identified, the temperature history can be used
to determine the thermal processing adequacy. This temperature history can be obtained
by either actual physical measurement or from constructive computation (Hendrickx et
al., 1995). The major disadvantage of this method is that it depends heavily on the
physical input parameters (thermophysical properties of food, flow characteristics, and
motion and temperature inside the package) required. These inputs can be impossible to
measure accurately. The use of thermocouples in many processes is not possible, and in
some instances, the thermocouple may disturb the kinetics of the attribute being
measured. The use of TTIs is the most promising alternative method for process

evaluation.

2.3.3 Time-temperature integrators

A time-temperature integrator (TTI) undergoes an irreversible, time/temperature
dependent change which can be used to verify that thermal processing is adequate. A
TTI should: 1) quantify the impact of a thermal process on a target attribute, 2) not
disturb. the heat transfer of the product, 3) be easy to recover from the food after
processing, and 4) be quickly and easily prepared for monitoring (Hendrickx et al., 1995).
A TTI should rrrirrric the change in a target attribute, such as a pathogen, undergoing the

same thermal process. TTIs are either biological, such as microorganisms or enzymes; or

11

chemical or physical, in which case they are dependent upon purely chemical or physical
changes in response.

To be a good TTI, an endogenous enzyme must become insoluble or inactivated
at temperatures commonly used for processing meat products, and be easily recoverable.
Wang et al. (1996) found that lactate dehydrogenase (LDH), triose phosphate isomerase
(TPI), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), phosphoglycerate mutase
(PGAM), and acid phosphatase (AP), might be useful in assays to verify processing
temperatures of cooked ground beef patties. For an endogenous TTI to be used, its 2-
value should equal the z-value of the pathogen used to verify adequacy of thermal
processing. Orta-Ramirez et al. (1997) calculated that TPI in ground beef had a z-value
of 556°C, which was close to that of E. coli 01 57:H7 and S. Senfienberg which were
5.59 and 625°C, respectively. Hsu (1997) tested several endogenous bovine enzymes,
peroxidase, TPI, AP, and LDH, under optimal and sub-optimal time-temperature
processing conditions to determine if any could be used as TTIs for precooked roast beef.
Both studies showed that TPI could be used as a TTI to determine thermal processing
adequacy using medium (583°C) and high (622°C) temperature schedules for beef roasts
(Hsu et al., 2000). The authors concluded that TPI might be useful to verify adequate
thermal processing of beef.

Sair et al. (1999) tested beef patties and found that TPI activity decreased from
20.26 to 6.28 U/kg meat in ground beef when cooked to the EPT of 71 . 1°C. The USDA
recommends that consumers cook beef patties to an internal temperature of 71.1°C.

Therefore, a TPI activity of 6.28 U/kg or below would indicate adequate cooking.

12

Several endogenous enzymes (LDH, GAPDH, creatine kinase, TPI, AP, turkey
serum albumin, and immunoglobulin G) were tested to determine if any could be used as
a TTI in turkey products. Among the proteins analyzed, the temperature dependence of
TPI in both low- and high-fat ground turkey was most similar to that of E. coli 01 57:H7
and S. Senfienberg. The z-value of TPI averaged 5.8 and 54°C in low— and high-fat
ground turkey (Veeramuthu et al., 1998). TPI was the only endogenous enzyme tested
which had a z-value close to that of Salmonella, indicating this enzyme might be used as

a TTI in turkey products.

2.4 FACTORS AFFECTING THE ACTIVITY OF TRIOSE PHOSPHATE
ISOMERASE

The effects of muscle type, animal maturity, fat concentration, and frozen storage
of beef on TPI activity were studied. Sair et al. (1999) found that muscle type did affect
TPI activity, with the trapezius muscle having the highest activity (2213 U/kg meat). The
vastus intermedius muscle had the lowest activity (595.0 U/kg meat), and the
semimembranosus (1984.6 U/kg meat) and longissimus dorsi (1754.4 U/kg meat)
muscles were intermediate. Because ground beef is a combination of these muscles, the
TPI activity was assumed to be an average of these activities. TPI activities were the
same in carcasses of different maturity (Sair et al., 1999). Frozen storage (-13° C for 10
months) of the raw steer muscles had no effect on TPI activity (Sair et al., 1999).
However, ground beef containing 9.3 and 20% fat had higher TPI activities than ground

beef containing 30% fat. When ground beef containing 0.5% added tripolydiphosphate i

13

(TPDP) and 1.0% NaCl was cooked to 628°C. 0.5% TPDP increased TPI activity (135.5

U/kg meat) when compared to the control (31.9 U/kg meat) (Sair, 1997).

2.5 FORMAT OF THERMAL DEATH TIME TESTING
2.5.1 Thermal death time tubes

Traditionally, thermal death time studies have been carried out by packing meat
into 10 x 60 mm glass thermal death time (TDT) tubes. A 1 g (Kotrola et al., 1997) or 2
g (Veeramuthu et al., 1998) sample was packed into the tube and then heat sealed. A
thermocouple was inserted into the center of a control tube to record the time required for
the center of the meat to reach the target temperature. The disadvantage of this system is
the time required for the internal temperature of the meat to reach the target temperature
(come-up time) is long (60 s). The outside of the sample reaches or exceeds the target
temperature before the internal region. The microorganisms or enzymes on the outside of
the sample are inactivated long before the target temperature is reached at high
temperatures. The advantage to this method is that a real meat system is used, not a broth

or slurry.

2.5.2 Capillary tubes

Orta-Ramirez (1999) conducted TDT studies with a fluorescent protein using
capillary tubes (Accu-fill 9O Micropet, Cat. No. 4624, Becton-Dickinson and Co.,
Parsippany, NJ). The protein solution (200 pl) was drawn into the tube by capillary
action and then heat sealed at one end and Teflon tape at the other. The tubes were

placed into a wire rack and immersed in a water bath. The come-up time was reported to

14

be 102tls. Schuman and Sheldon (1997) also used capillary tubes loaded with raw egg
yolk or whites inoculated with Salmonella spp. and L. morrocvtogenes to validate the
USDA pasteurization protocols for inactivation of L. monocytogenes in liquid egg
products. The advantage of this system is that the come—up times are limited to seconds.
However, the disadvantage to this system is that ground meat does not fit into the
capillary tubes. A meat slurry, which may not represent a true meat system, must be

used.

2.5.3 Plastic pouches

Another way to conduct TDT studies is to pack the meat into small plastic
pouches and then compress it into a thin layer. Juneja et al. (1997) packed 15 x 22.9 cm
sterile Whirl-pak sampling pouches with 3 g of ground beef or chicken inoculated with E.
coli 01572H7 and compressed the meat to a thickness of 1-2 mm. The size of the pouch
and sample can vary. Schoeni et al. (1991) packed 10 g of meat inoculated with L.
monocytogenes into 205 mm x 85 mm boilable pouches (polyvinal outer layer
polyethylene inner layer; Dazey Corporation, Industrial Airport, Kansas) and used a
rolling pin to flatten the bag to a thickness of 1.4 mm. The advantage of the pouches is
that same amount of meat can be placed into the pouch as the traditional TDT tubes, but
with a shorter come-up time. Meat can be used in the pouches opposed to a meat slurry

in capillary tubes.

15

2.6 THERMAL DEATH TIME STUDIES
2.6.1 Beef

Goodfellow and Brown (1978) were commissioned by the USDA to determine a
time-temperature protocol that would ensure the safety of “rare” beef roasts. They used a
cocktail consisting of six strains of Salmonella in ground beef to determine D-values for a
7-log decrease in beef roasts. The strains consisted of Salmonella Typhimurium strain
TM] (used because it was previously reported by Ng et al. (1969)), Salmonella Newport,
Salmonella Agona, Salmonella Bovis-Morbificans and Salmonella Muenchen. Ground
beef was inoculated with the cocktail to achieve 1.7 x 107 CFU/g of meat. At 572°C the
D-value was 38-42 min and the z-value was 561°C (Table 2.2). The USDA developed
the time—temperature schedules for cooking roast beef using this data.

The USDA recognized the importance of E. coli 0157:H7 in meat and poultry
products, and commissioned Line et al. (1991) to determine the thermal lethality of E.
coli 0157:H7 in ground beef and beef roasts. For the study, two fat levels of ground beef
were used. The lean (2.0% fat) and fatty (30.5% fat) ground beef were inoculated to
contain approximately 107 CFU E. coli OIS7:H7/g and heated at 51.6, 57.2, and 627°C.
Survival of the organisms were determined by two methods: enumeration on plate count
agar (PCA) containing 1% sodium pyruvate and by the 2-h indole test. The D-value for
E. coli in 2% fat beef at 572°C (4.10 min) was the same as that reported by Goodfellow
and Brown (1978) for the 6-strain Salmonella cocktail. Both were enumerated on PCA.
At 572°C and 627°C, Salmonella had higher D-values than E. coli 01 57:H7. However,
at 516°C, E. coli0157zH7 (Line et al., 1991) had a higher D-value (115.5 min) than

Salmonella (54.3 min) (Goodfellow and Brown, 1978). Even though different

16

microorganisms were being compared, Line et al. (1991) determined the higher D-values
of E. coli 01 57:H7 to be due to differences in the recovery media and plating procedures.

Listeria monocytogenes is another microorganism of concern in the meat industry.
The USDA-FSIS commissioned a study by Fain et al. (1991) to determine the thermal
lethality of Listeria in beef and poultry products. Lean (2.0% fat) and fatty (30.5% fat)
ground beef were inoculated to contain 107 CPU Listeria /g and heated at 51.6, 57.2, and
627°C. The D-values for Listeria were higher than those found for Salmonella sp.
(Goodfellow and Brown, 1978) and E. coli OlS7:H7 at 516°C (Line et al. 1991). At
572°C, the D-values for Listeria were lower than those for Salmonella and E. coli
OIS7:H7. At 627°C, Listeria and Salmonella had similar D-values. The z-value for
Salmonella spp. (Goodfellow and Brown, 1978) was higher than E. coli and Listeria
(Fain et al., 1991).

Orta-Ramirez et al. (1997) determined the D- and z-values for E. coli OIS7:H7
and Salmonella Senfienberg in ground beef (3.8% fat) (Table 2.2). At 58°C, the D-value
for E. coli OIS7:H7 was higher than that found by Line et al. (1991). The z-value for E.
coli 01 57:H7 obtained by Orta-Ramirez (1997) was also higher than the one obtained by
Line et al. (1991). Differences in D- and z-values could be due to differences in fat
content, pH of the meat, recovery methods and microbial strains, as well as the use of
cultures in different grth phases (Orta-Rarnirez et al., 1997). Salmonella Senftenberg
was more heat resistant than E. coli OIS7:H7 which was consistent with the results found

by Line et al. (1991 ). The D-values of S. Senftenberg in ground beef were much higher

17

 

 

 

 

 

«no we
mod me
:8: t .2 mm
._e a Neeeeaeco the 8.? m meeeeeaeem ..3.
S .0 we
9.0 8
:8: 3e mm
.3 a Neééeto oew ..3 mm 2320 :8 .m
8e 6%
com NR
:8 c ._e a can own an. _ w 03 3836955 4
one 68
2 .e N:
:8: ..e a 2.: t; S: o: 2:20 :8 .m
3.3 2e
3.? New .3386
$8: 5.55
e5 26.3800 En Ne- _ e e. a 33.555. sage

 

8:20.31

Auev 02a?”

AEEV o:_a>-D

Gov 2383th

EmEmwBEE—z

 

moon 5 862m 08: 53c .982: we mo=_e>-N can ..Q «o >633. .~.~ 035.

18

than the D-values obtained using a 6-strain Salmonella cocktail (Goodfellow and Brown.

1978). However, S. Senflenberg is very heat stable (N g et al., 1969).

2.6.2. Turkey

Veeramuthu et al. (1998) determined the D- and z-values for E. coli OIS7:H7 and
S. Senftenberg in ground turkey thigh meat (Table 2.3). The D-value for E. coli 01 57:H7
at 60°C was 5.47 min. This is 6.07-fold higher than the D-value for E. coli 01 57:H7 in
turkey found by Kotrola and Conner (1997), 9.94-fold higher than the D—value found by
Ahmed et a1. (1995), and 2.9-fold higher than the D-value found by Juneja et al. (1999)
(Table 2.3). The z-value for E. coli 01572H7 found by Veeramuthu et al. (1998) was also
higher than the z-values found by Ahmed et al. (1995) and Kotrola and Conner (1997) in
turkey meat. These differences could be due to pH, fat content, and method of
enumeration (Veeramuthu et al., 1998). The D-values of S. Senftenberg were higher than
E. coli 01 57:H7 (Veeramuthu et al., 1998) (Table 2.4). However, the z-value for S.
Senftenberg was lower than the z-value for E. coli 01 57:H7 (Veeramuthu et al., 1998).
In turkey thigh meat, it appears that S. Senftenberg is less temperature dependent than E.

coli01572H7.

2.7 ENVIRONMENTAL EFFECTS ON MICROORGANISMS

Many factors including water, fat, salts, carbohydrates, pH, proteins and additives

may affect the heat resistance of microorganisms.

l9

 

 

 

 

 

 

8. we
5. no
a: 8
on: 2m
88: a a... 282 one 3.: mm $320 :8 m:
a; we
8.: 8
$8 : ..e a 33.88; ea 2: _N R meeeeeceem .:
of we
2% 8
38: ..e a Eases; 3 3.8 mm EH35 :8 ...:
ed 8
mm a
«.2 mm
:8: .228 :8 226x :1. as. am EH35 :8 ...:
who 8
Re 2
:8: ._e 8 85,2 4:. 3.2 0m $320 :8 .m
oococomoz GOV o:_m>-N AEEV 0373-0 Gov Ssfibqfioh EmSwEEEE

 

.AB—c2 E 3:53 0:5 Snot .952: .«o mos—gs: use .9 mo 3030M .n.n 033.

20

 

 

 

 

 

 

 

 

who 8
wow 3
www. ow: 8 ca :1: were
who we
ewe mm
4:. 3.8 cm on $3 awe:
:3 8
8.2 R
wwe Sww ow ow: $0388
23 8
on: mm
3:. Sew 8 ca .8383
new 8
$8: 8.: 2
..e a 85.? w: ewww ow ca .8335 2:20 :8 m:
:3 ewe
ww New
_we 3: 9: Ge: £2388
owe 2e
:8: 2. NR
..e a 2.: 2.... www e: an $0.382 Ewflo :8 .m
8:053. Gev o=_w>-~ AEEV 02.5-9 838095... Cs.“ e\ev :82 EwSmEE£E

 

.wEoEoo é wart“; FEB :85 can zoo—c3 E 863% 085 foot .9505 Mo wo:_m>-N use to .«o 3033— .v.N «Bah.

21

 

 

 

ad oo
YN mm
0.: mm
ewe www am an 8:18.55
0.0 cc
wd mm
:8 : 3.. mm
._w a 22.3. t; wwe 9. ca .89 acts ES 5 :8 m:
8:20.23. CL o:_m>-~ AEEV 02min— EBSQQEQH Cum e\ev $02 EwEamBEBE

 

.Aeeeeev ew 28H

22

2.7.1 The effect of fat

Line et al. (1991) studied the effect of fat content on thermal inactivation of E.
coli OIS7:H7 in beef (Table 2.4). At 516°C, the D-value of E. coli 01 57:H7 in high fat
beef (30.5% fat) was 1.47-fold higher than in low fat beef (2.0% fat).

This trend was confirmed by Ahmed et al. (1995), who observed higher D-values
in E. coli 01 57:H7 as fat content of ground beef and turkey increased. Ahmed et al.
(1995) stated that this was likely due to the decreased water content of the meat, which
alters heat transfer. Kotrola et al. (1997) did not see an increase in D-values of E. coli
01572H7 with increased fat content. Juneja and Eblen (2000) also observed lower D-
values with increased fat content in an 8-strain Salmonella Typhimurium DT 104 cocktail
inoculated into ground beef. However, the thermal inactivation kinetics of the DT 104
cocktail show deviated from first order kinetics, with an initial lag period before death
occurred. This could account for the lower D-values.

Ahmed et al. (1995) and Kotrola et al. (1997) observed a decrease in temperature
dependence (z-value) with increased fat content in turkey with E. coli 01 57:H7.
However, Line et al. (1991) did not report the same trend for E. coli01572H7 in ground

beef

2.7.2 The effect of additives

Sodium chloride, sodium lactate, and phosphate are common additives in
processed meat products. Kotrola and Conner (1997) studied the effects of these
additives in turkey meat on the heat inactivation of E. coli 01 57:H7. Ground turkey

breast meat contained either 3% or 11% fat, and either 8% NaCl or 4% Na lactate. One

23

formulation contained 8% NaCl, 4% Na lactate. and 0.5% polyphosphate. The samples
were heated at 52, 55, 57, and 60°C. Both NaCl and Na lactate increased the D-values for
E. coli OIS7:H7 in turkey containing 3 and 11% fat at all temperatures; however NaCl
had a greater effect. The highest D-values were found in the formulation which
contained all three additives. The authors explained that additives reduced the water
activity, and that the survival of E. coli 0157:H7 increased as water activity decreased.
The effect of NaCl and phosphate buffer on the thermal resistance of Clostridium
botulinum spores in a turkey slurry was studied by Juneja et al. (1995). Salt (1, 2, or 3%
(wt/vol.)) was added to a turkey slurry inoculated with 107 spores, packed into 17 x 60
mm vials and heated at 75, 80, 85, or 90°C. The D-values for C. botulinum increased in

the 1% salt-added slurry, but decreased in the 2 and 3% salt-added slurries.

2.7.3 The effect of pH

Abdul-Raouf et al. (1993) studied the effect of pH on the grth and survival of
E. coli 01 57:H7. A ground roast beef slurry (pH 6.03) was adjusted to pH 5.0 (:1: 0.02)
using solutions of either 10% acetic, citric, or lactic acid. The slurry (125 mL) was
placed into flasks and heated in a water bath to either 52, 54, or 56°C. When the desired
temperature was reached, the inoculum was added. Samples were withdrawn at specific
time intervals, serially diluted in 0.1% peptone water and plated on either modified
Sorbitol MacConkey agar (MSMA) or tryptic soy agar. The researchers found that the
organisms became less heat stable in the acidified slurries and stability was also
dependent on the type of acid used. Acetic acid was the most effective followed by lactic

acid. Citric acid was least effective. The researchers also discovered that MSMA was a

24

poor medium for the recovery of heat injured cells. Heat injured or stressed cells have
decreased ability to form colonies in a selective medium such as MSMA, as compared to

a non-selective medium such as TSA.

25

CHAPTER 3
THE THERMAL RESISTANCE OF SALMONELLA SENFTENBERG,
ESCHERICHIA COLI OIS7:H7, AND TRIOSE PHOSPHATE ISOMERASE IN

GROUND TURKEY MEAT.

3.1. ABSTRACT
The effects of muscle type, fat content, salt, and phosphate on thermal inactivation
kinetics of S. Senftenberg and E. coli 01 57:H7, and the enzyme triose phosphate
isomerase (TPI) were studied in turkey breast and thigh meat. Thermal resistance
(D-values) was highest in TPI followed by S. Senftenberg and E. coli 01 57:H7. The
z-values for E. coli 0157zH7 were lower than those for TPI and S. Senfienberg. Salt
increased the z-values of both TPI and S. Senfienberg, but did not have an effect on
E. coli 01571H7. Except for the phosphate treatments, the z-values for TPI and S.
Senfienberg were within 0.5°C of each other. Hence, TPI might be useful as a time
temperature integrator (TTI) in phosphate containing products if a mathematical
model can be developed to relate the lethality of Salmonella to inactivation of TPI.
TPI has potential to be a time temperature integrator to verify processing adequacy
in turkey meat, because TPI and S. Senftenberg have similar temperature

dependencies (z-value).

26

3.2. INTRODUCTION

There are approximately 1.4 million cases of salmonellosis in the United States
per year (CDC, 1999b). The illness usually lasts 4-7 days and most people recover
without treatment. However, in cases involving those who are very young, elderly, or
immunocomprorrrised, this infection can be life threatening. Of these cases,
approximately 1,000 prove to be fatal (CDC, 1999b).

In recent years, Escherichia coli 01 57:H7, an enterohemorrhagic strain, has
emerged as a threat to food safety. Undercooked ground beef has been most commonly
associated with this illness, however additional cases have been traced to many other
products including dry fermented salami (CDC, 1995), raw milk (Padyhe and Doyle,
1992), apple cider (Zhao et al., 1993 ), turkey rolls (Carter et al., 1987), and potatoes
(Griffin and Tauxe, 1991).

USDA regulations previously required that cured and uncured poultry products be
cooked to an internal temperature of 683°C and 71 . 1°C, respectively. However, on
January 6, 1999, the USDA-F SIS published performance standards for poultry products,
which shified the focus from "command and control" regulations listing specific
processing times and temperatures. The new lethality standards allow the processor to
use any thermal processing technique, provided they can document that the process will
decrease Salmonella populations by 7-loglo (USDA-FSIS, 1999). These new regulations
offer flexibility by permitting the use of more than one processing step to ensure safety

while increasing product quality.

27

The USDA stated that challenge studies, using Salmonella-inoculated poultry
products, must be conducted by the processor to prove the adequacy of the thermal
processing method. Due to obvious food safety regulations these challenge studies can
not be conducted in commercial processing facilities. Hence, alternative approaches are
needed to verify the adequacy of the thermal process in the processing plant.

An endogenous time-temperature integrator (TTI) is a marker, typically a protein or
enzyme, that undergoes an irreversible, time/temperature dependent change, which can
be used to verify adequate thermal processing. The TTI should have the same response
as the target attribute when temperature is the only rate-determining factor. The z-value
(temperature increase needed to decrease the thermal death time or D-value by 90%) can
be used to select a TTI if the marker and the target attribute both exhibit linear
inactivation kinetics.

Several endogenous proteins were screened by Wang et al. (1996) to determine if
any could be used as TTI's. Triose phosphate isomerase (TPI) emerged as the most
feasible. The z-value for TPI in ground turkey thigh meat was 58°C as compared to 6.0
and 56°C for E. coli 0157:H7 and Salmonella Senftenberg, respectively (Veeramuthu et
al,1998)

Before TPI can be used as a TTI in poultry products, the effect of environmental
factors on TPI, Salmonella and E. coli 01572H7 must be determined. The objective of
this study was to assess the impact of muscle type, fat content, salt, and phosphate
concentration on the thermal inactivation parameters of S. Senftenberg, E. coli 01 57:H7,
and TPI in turkey breast and thigh meat to determine the feasibility of using TPI as an

endogenous TTI to verify processing adequacy of turkey products.

28

3.3. MATERIALS AND METHODS

3.3.1. Meat formulations

Skirrless turkey breast meat, skin-on turkey breast meat, skinless thigh muscle and
thigh trimmings were obtained from Bil Mar Foods (Zeeland, M1) on the day of
slaughter. The meat (45.5 kg) was placed into plastic bags, packed with dry ice and
transported to Michigan State University, where it was kept below 3°C for 24 h. The
meat was then chopped in a bowl chopper (Model VCM4OE, Hobart Mfg. Co., Troy,
OH) for 15 s on low speed and 1 min on high speed. The ground low fat turkey breast
meat (LFB), high fat turkey breast meat (HFB), and low fat turkey thigh meat (LFT) were
prepared similarly. HFT was prepared by mixing skinless thigh muscle and thigh
trimmings in a Hobart mixer (Model A200, Hobart Mfg. Co., Troy, OH) for 5 min at 4°C.
The temperature of the meat remained below 155°C during chopping. Fat content of the
four meat types was determined by ether extraction (AOAC Method 991.36, 1996).

Brine solutions (Table 3.1) were made by dissolving NaCl or phosphate (PP, 50%
Food Grade Liquid Potassium Diphosphate, Butcher & Packer Supply Co., Detroit, MI)
in water. Brine solutions were used to prepare HFB containing 1% NaCl, 2% NaCl, 1%
NaCl and 0.25% PP, and 1% NaCl and 0.5% PP. Controls were prepared by adding
water (3%) to LFB, LFT, and HF T. The solutions were added dropwise to 2400 g meat
to achieve a 3% increase in weight while mixing in a Hobart mixer (Model A200, Hobart
Mfg. Co., Troy, OH) on low speed. After adding the brine or water, the mixture was
mixed at high speed for 5 min. The meat blends were divided into 120 g portions,

vacuum packaged in polyethylene-laminated nylon pouches, and frozen at

29

Table 3.1. Brine solutions of high fat turkey breast meat (HFB)

 

 

Brine Solutions NaCl (g) PP (mL) Water (mL)
HFB 0 0 72.0
HFB, 1% NaCl 24.72 0 72.0
HFB, 2% NaCl 49.44 0 72.0
HFB, 1% NaCl, 0.25% PP“ 24.72 12.0 66.0
HFB, 1% NaCl, 0.5% PP 24.72 24.0 60.0

 

° PP=potassium diphosphate

30

-10°C. The frozen meat was irradiated at IOkGy to eliminate indigenous microflora.
Moisture, fat and protein contents of the meat were determined by AOAC (1996)
Methods 991.36, 981.1, and 950.46B respectively. The pH was determined as

described by Sair et al. (1999).

3.3.2. Thermal inactivation experiments

Ground turkey was thawed under cold running for 5 rrrin water prior to each
thermal inactivation study. Enzyme inactivation experiments were performed using
uninoculated meat that had been irradiated. Bacterial inactivation experiments were
performed using irradiated meat that was inoculated with either S. Senftenberg or E.
coli 01 57:H7 as described below. The meat was placed into a 60 ml syringe (Becton
Dickinson and Co., Franklin Lakes, NJ) and 1g meat was extruded into a 5 x 25.5 cm
polyethylene-laminated nylon pouch (Butcher & Packer Supply Co., Detroit, MI). The
meat was then rolled between two guides to a thickness of 1mm using a large glass test
tube. The bag was heat sealed using a soldering iron, refiigerated and used within 4h.
A type-T (copper-constantan) thermocouple (Part # TJ48-SCPSS-O32G-3.25-OST-M,
Omega Engineering, Inc., Stamford, CT) was inserted into a bag containing sterile meat
and sealed on both sides. The thermocouple was connected to a Danook Data
acquisition system equipped with a DBK 19 Thermocouple Card (Omega Engineering,
Inc., Stamford, CT). The bags were placed into a Polystat circulator bath (Model 1268-
52, Cole-Farmer Instrument Company, Chicago, IL) set at 55, 58, 61, 64, or 67°C. The
bags were removed at specific time intervals and immediately placed into an ice-water

bath. Preliminary experiments determined what time intervals were selected to ensure a

31

3-log decrease for the bacterial inactivation studies, and a 1-log decrease in enzyme

activity for the enzyme inactivation studies.

3.3.3. Bacterial cultures and inoculation

Salmonella Senftenberg (ATCC 43894) and Escherichia coli OIS7:H7 (ATCC
43 894) were stored frozen in 20% glycerol at -80°C. Two 24 h transfers were made into
TSB incubated at 37° C prior to use in thermal inactivation studies. Static cultures were
grown 24 h at 37°C in tryptic soy broth (TSB) (Difco Laboratories, Detroit, MI). After
the first 24 h transfer, the microorganisms were plated to determine the amount to be
inoculated into meat. New cultures were grown every 2 weeks.

Meat (120 g) was inoculated to contain 1.0 x 107 CFU/g meat. The meat was
rrrixed manually using sterile gloves in a sterile bowl to ensure even distribution of
microbes. Dye was used in preliminary trials to determine that mixing was adequate.

One gram of meat was aseptically transferred to nylon pouches as previously described.

3.3.4. Protein extraction for TPI activity

Cooked turkey was transferred to scintillation vials (Research Products
International Corp., Mount Prospect, IL), combined with 3 volumes (w/v) ofO. 15 M
NaCl, 0.01 M sodium phosphate buffer, pH 8.0 (PBS) and mixed for 30 s using a Fisher
Vortex Genie2TM (Scientific Industries, Inc., Bohemia, NY). The extracts were stirred,
without foaming using magnetic stir bars, for 15 min at 4° C and immediately
centrifuged at 6000xg for 10 min at 4°C. The supernatant was filtered through

Whatman no. 1 paper and held on ice until assayed for TPI activity within 24 h.

32

3.3.5. TPI activity determination

TPI activity was determined as described by Sair (1997). The reaction buffer
contained 1.0 mL 0.2 M triethanolamine buffer, pH 8.0 (TEA), 0.2 mL, 15 mM
glyceraldehyde-3-phosphate, 10 ul B-nicotinamide adenine dinucleotide (10 mg/ml), 10
pl glycerol-3 -phosphate dehydrogenase (1.5 mg/ml), and 10 111 protein extract. The
change in absorbance was measured at 340 nm for 1 min at 25°C. The protein extract
was diluted with water as needed so that change in absorbance per minute was less than

0.2 (Beisenherz, 1955). TPI activity was calculated using the following equation:

Activity (U/l) = 1000 x A x V
T x E x d x (O/D)

 

A = Absorbance

V = Assay volume (ml)

T = time (min)

E = extinction coefficient (6.310 L x mmol-l x cm)
D = distance of light path (1 cm)

0 = sample volume (0.01 ml)

D = dilution of sample

3.3.6. Bacterial counts

Within 4 h of heat treatment, the cooked turkey (1 g) was transferred into sterile
18 oz Whirlpak bagsTM and manually homogenized with 9 mL of 0. 1% sterile peptone

water. Appropriate dilutions were made with 0.1% sterile peptone water before S.

33

Senftenberg and E. coli 0157:H7 were plated in duplicate on Petrifilm Aerobic Count

Plates (3M, St. Paul, MN). The samples were counted after 24 h of incubation at 37°C.

3.3.7. Data Analysis

D- and z-values were calculated assuming first-order kinetics (Pflug, 1997). All
experiments were performed in triplicate. D-values (time required to reduce the
microbial count or enzyme activity by 90%) were calculated using linear regression
analysis of time vs. log TPI activity or log CFU/g with Microsoft Excel (version 97).

At least 4 data points with a correlation coefficient >085 were used. The z-values were
calculated using linear regression analysis of temperature vs. log D-value with
Microsofi Excel (version 97).

Statistical analysis of D- and z-values was conducted using one way analysis of
variance (SAS Institute Inc., 1995). The Tukey-Kramer HSD test was used to compare

D- and z-values with the mean square error at the 5% level of probability.

34

3.4 RESULTS AND DISCUSSION
3.4.1. Proximate composition

LFB and HFB contained 0.6 and 4.4% fat, respectively (Table 3.2). LF T and
HFT contained 5.3 and 7.4% fat, respectively. Breast meat had a lower pH than thigh
meat. The addition of salt to HFB did not affect pH, however, the addition of phosphate
increased the pH to that of thigh meat.

3.4.2. Effect of muscle type and fat content on thermal inactivation of TPI

The D-values of TPI were greater in turkey thigh meat than in turkey breast
meat at each temperature (Table 3.3). The D-values of TPI in LFT (pH 6.2) were
greater than those of HFB (pH 5.7) even though the fat contents of the two meat
formulations were similar (5.3% vs. 4.4%) The pH of the meat or the muscle isoforrn of
TPI may have influenced the stability of the enzyme.

Veeramuthu et al. (1998) observed higher D-values in 9.8% fat ground turkey
thigh meat than in 4.3% fat thigh meat. At 58°C, the D-value of TPI in 4.3% fat turkey
thigh meat was 38.46 min compared to 92.64 min in 9.8% fat turkey thigh meat. The
D-values found in our study were also higher in both the I-IFB and HF T compared to the
LFB and LFT, respectively. The D-values of TPI in turkey thigh meat were also similar
to the D-values of TPI in turkey thigh meat found by Veeramuthu et al. (1998).

The z-value of TPI in LFT was higher than in LFB, HFB, or I-IFT (Table 3.4).
Fat content did not have an effect on the z-values of TPI in breast meat, however, LFT
had a higher z-value than HFT. Veeramuthu et al. (1998) found that TPI had a z-value
of 5.8 min in 4.3% fat ground turkey thigh meat, and 5.4 min in 9.8% fat ground turkey

thigh

35

22388 n .5 885 5 ewz n mm: 885 E 3e. 1 m5 ewe: E 82 u E: awe: E 3e. u k: ..

.Amucv :88 2: CO 8.5 3353. 2: H .82: 05 we 83898 8: wo:_a> e

 

 

Some 8.03.2 9.83.2 :wuwww .E .83 .52 .x: .95
8.92 e co. 73.2 3.3: .8 ween...“ .E 3mg .62 ex: .3:
woeeww 5.3.9: 8.3.: 2.3: 52 am .95
_oweww $38.: 3392 2.30.4 .02 .x; .95
5.33 5.04.2 3.03.: 2.048 E:
Seeww $.33. w _ .oeeww 8.30.0 m:
5.33 33w. : 03.: we :3: E:
8.33 8.312 5.353 8.33 E
E. E: as; 582.. 3} ..er 23%: 3? é a:

u:e:a.=:..8.m :82

 

$on :83: Co In can coEwoquo 895on .Nd 035.

36

2: 3 33230 3 0:735 2; .

.Amucv :38 of mo Stu e898; 2: “_n :85
AmodAav “costs 8: 8a 5:2 0:8» 2: 3 330:8 832388 08mm 2: £53» mos—«35 ..

:85 3 .32 u Em 585 E 32 n P: Ema: a .32 n E: 55 E 32 u r: .

 

 

 

 

 

eod umodflmo; mowow 336. he
wad :wodfied Sema wenmd- we
23 mOdeNQQ owwmw wmnod- _e
cod pmmdflodw wemmw wowed- mm mm:
wad amodfiofio 2w. .w ommw. T he
cod candida omwoa w_eN.o- we
mod $035.: 12w wooed- _e
mod n: _ . 33.3 e. mmw 3.86. we mu:
3.0 cmodfiweg each. m wooed- be
mad <2.ofimw.e emena :26. we
36 <o_.mfioo.mm eomwd mmwod- _e
23 30.3303 wwwfim 85.0. mm .5:
mod <2.03%.. wmmna nmmmd- we
36 <endeumms owmna ohm—d- we
3.0 1632.3 meend mowed- _e
eod mmesfiowe Emma wflod- mm H”:
.35 Q
"~— o=_a>-n 2.3.8:..-» 2.2m «bananas—oh ...—33.558 :32

 

:38 swat ecu 335 >83: 5 AEC 88080£ Sesamosq 82: 8m $335 cofimfime e5 mo:_a>-0 .n.m 933—.

37

.Amucv :88 2: go St» 23.33 2: H :88
2: 3 33298 fl o:_a>-n ugh .GoquV «caught—V bazaoficma 8: 8a 5:28 a 55:3 Etofloqa 2:8 05 53» mos—#3 ._

08.38% u an 585 3 :wE u mum 3305 E 32 n max— EwEu é FEE u HE £35 3 32 n .54 .

 

38

 

 

(Sam? ES 63% <2 in? A: $3 .62 3: “PE
<33»: 333mg <2 in? E $8: .652 x; E:
<_ _ .933 can??? 08.933 62 :N .95
<83»: 0.33.3me a _ 38¢ 62 $195
5.: .32 .v £3.3on ES .933 mm:
7.3.3:? 2232 a 33% PS
1233 £33m 33:6 E:
as??? $9386 x8336 E
E ”R _o =8 a mgcocfim w EH assuming :32
«~05

amid can :35 >82: 5 2+? _0 :8 Q 28 ,mbncotcom .m AEC omSoEomm 829.23 82.: mo mos—SEN of. in 933,—.

meat. The z-values of TPI in ground turkey thigh meat found in our study are very

similar to those found by Veeramuthu et al. (1998).

3.4.3. Effect of muscle type and fat content on thermal inactivation of Salmonella

The D-values for Salmonella were the same (P<0.05) in LFT and HFB (Table
3.5) at 58°C, but at the higher temperatures, HFB had higher D-values than LFT.
Salmonella was more thermally stable in breast meat than thigh meat. The D-values of
HFT were higher than LFT at all the temperatures, and the D-values of I-H‘B were
higher than LFB at all temperatures except for 58°C. Increased fat content increased
the D-values of Salmonella in turkey meat. This trend was observed with E. coli
0157:H7 in turkey meat (Ahmed et al., 1995). Veeramuthu et al. (1998) reported that
Salmonella had a D-value of 13.24 min at 60°C in 4.3% fat thigh meat in 10 x 75mm
glass tubes enumerated on Petrifilm Colifonn Plates. The higher D-value could be due
to a longer come-up time (60 s) for the glass test tubes as compared to the pouches (8 5)
used in this study. With a longer come-up time the initial microbial count is decreased,
reducing the slope of the regression line, which results in a lower D-value.

Even though the D-values are different, Veeramuthu et al. (1998) reported a 2-
value of 56°C for S. Senfienberg in 4.3% fat thigh meat. This value is similar to the 2-

value (525°C) found in the present study.

39

Amuse :88 2: mo Ste 3353 2: H :88 2:
3 33298 i o:_m>-Q of. .GodAaV 280%“. 8: 2m 8:2 08% 2: .3 330:8 828383 083 05 £515 mos—gd ._

385 E FEE H mm: 585 3 32 n m": :35 5 .3: n in Ewe“ E 32 u #3 ..

 

 

 

 

 

mod <_o.onmow.o eowme woowd- ee
mod «No.32.— oame mg»..- we
cod moodae— .w eSme mowmd- _e
wad ammohofiw mmwme 32.? mm mum
mod mmodafl .o mmmwe woe~ .m- ee
No.0 08.336 mneme 35.7 we
owe oweduflma Nine wowmd- _e
3.0 midfiefiw wm _e.e mm: .0. mm mm:
5o mmodagd mmmee mmmmw- ee
mod moodamfio nowme 3527 we
cod <om.oawm.w wmmwe 386.. _e
8.0 <ow.ohmm.: e_me.e eewod- mm .5:
owe meodaomd onmme oeeow- ee
mod odo— .onmwed hemme e_em. T we
36 oxfonw—M mmnwe 33.5.. 3
mad mammahge wm _e.e e2 _d- mm .H..:
$5.5 Ge
"a 2:579 2.3.8:..-» 2.2m 9.3.22.th ..=oE.ao...—.

 

See swat new 585 5%.: E whencoacom ..e .8 max—£8 :238w2 ecu mos—g D .m.m 03:.

40

 

3.4.4. Effect of muscle type and fat content on thermal inactivation of E. coli
0157 :H7

Fat content and muscle type did not affect the D-values for E. coli 0 1 57:H7 in
turkey meat. Even though the D-value of E. coli 0157:H7 in LFT was higher than both
LFB and HFB (Table 3.6) at 55°C, the D-value of HFT was not difl°erent (p<0.05) from
LFB. At the higher temperatures, no difference was seen between the different muscle
types and fat contents. Veeramuthu et a1. (1998) reported a D-value of 79.54 min at
55°C in 4.3% fat thigh meat. The higher D-values could be due to different heating
methods as discussed previously, or enumeration methods. Kotrola et al. (1997) studied
the thermal inactivation kinetics of E. coli 0157:H7 in 3% fat ground turkey breast
meat and obtained a D-value of 12.5 min at 55°C. This value is similar to the value
obtained in the present study.

The z-values (Table 3.4) of the four meat formulations were not different.
Kotrola et a1. (1997) reported a z-value of 4.47°C for E. coli 0157:H7 in turkey meat,
and Ahmed et al. (1995) reported a z-value of 474°C for E. coli 01572H7 in 3% fat
ground turkey breast meat. These values are similar to the ones obtained in this study.

Veeramuthu et al. (1998) reported a z-value of 6.0°C for E. coli 0157zH7 in
turkey thigh meat, which is higher than the z-values obtained in this study. Differences
could be due to due to different heating methods, as well as different recovery methods.
Veeramuthu et al. (1998) used Petrifilm® Coliforrn Count Plates, a selective medium,
which would result in recovery of fewer heat injured cells. A higher number of heat
injured cells will form colonies on a non-selective media, such as Petrifilm® Aerobic

Count plates used in the present study, which would be more temperature dependent.

41

Table 3.6. D-values and regression analysis for E. coli 0157zH7 in turkey breast and

 

 

 

 

 

thigh meat
Meat formulation‘l Temperature Slope y-intercept D-valueb R2
(”C) (min)

LFT 55 -00410 7.1350 24.393131‘ 0.98
58 -03243 7.1338 3.08d:0.28B 0.98
61 -1.8700 7.1347 0.53iO.O7B 0.95
64 -42257 6.2393 024150.01A 0.91

HFT 55 -0.0384 7.0525 26.0412. 16A 0.88
58 -0.3354 7.3069 2.983:O.13B 0.97
61 -2.0368 7.0616 0.493007B 0.92
64 -7.2650 6.6790 0. 1410.028 0.89

LFB 55 -00445 6.8424 22.47:}:1 .25A 0.91
58 -04079 6.6400 2.45i0.16C 0.91
61 -2.8667 6.7978 0.353011B 0.90
64 -5.5088 6.0398 0.18i0.07A'B 0.88

HFB 55 -00517 6.8010 193431.19B 0.97
58 -0.2366 6.7638 4.23:0.10A 0.92
61 -1 .0696 5.9209 0.933002A 0.90
64 -8.6690 6.4450 0.12i002B 0.94

 

‘ LFT = low fat thigh; HFT = high fat thigh; LFB = low fat breast; HFB = high fat

breast

b D-values within the same temperature followed by the same letter are not different
(p>0.05). The D-value is expressed as the mean i the standard error of the mean (n=3).

42

3.4.5. Effect of salt and phosphate on thermal inactivation of TPI

Addition of salt decreased the D-values of TPI in HFB (Table 3 .7). Sair (1997)
observed decreased activity of TPI in ground beef with 1.0% NaCl when heated at both
62.8 and 694°C. Addition of phosphate to HFB containing 1% NaCl resulted in an
increase in the D-values of TPI. This effect may be related to the pH of the meat
formulations. The pH of HFB containing phosphates was about 0.3 to 0.4 unit greater
than the meat alone. Addition of 0.5% phosphate with 1.0% NaCl increased activity of
TPI above that of the control. Phosphate increased the meat pH, which could account
for this effect. The z-value of TPI in HFB was the same (P<0.05) as that in HFB
containing 1% NaCl, and 1% NaCl with PP. The z-value of TPI in HFB containing 2%

NaCl was higher than the other formulations.

3.4.6. Effect of salt and phosphate on the thermal inactivation of Salmonella

The D-values of Salmonella increased at most temperatures as the salt content of
HFB increased from 0 to 2% (Table 3.8). Thermal inactivation of S. Enteritidis in
tryptic soy broth (TSB) was more rapid in the presence of 0.5% NaCl, than in the
presence of 3.5% NaCl (Blackburn et al., 1997). Blackburn et al. (1997) observed that
decreased water activity increased the heat resistance of bacteria.

The addition of phosphate increased the D-values of Salmonella in HFB
containing 1% NaCl. The optimum pH for grth of Salmonella is 6.6 to 8.2 (Jay,
1996). The addition of phosphate increased the pH of the meat closer to the optimum

for growth, which might also increase heat resistance of the microorganism.

43

$.05 :88

2: «0 echo Emcee; one a :86 2:

we eemmeaxo fl 02.3-0 2:. AmodAav Engage Ho: 8a 632 2:8 05 .3 326:8 2322—88 088 2: £53» mos—gd .

 

 

 

 

 

 

3o ..eoeamwe e33... 82.”. e

8e 68.32; 38w 683- we

ewe amnewoew 82w woome- _e

ewe ..3. _ameam eeemw Nwwee- em .5 :3 .62 w... .3:

ewe ...eewowe 2%..“ Boom- e

8e ..o. ewe. ._ mwmmw «New? we

8e maeaoew eewNw N22- 3

3e 63.9.2.2 wee ow Reed- we .5 $3 632 <1 .mm:

woe aseaeme e2 3 N23- we

Noe gained 89.,” :3. _- we

8o 28.33: was: wmwwe- _e

woe eenofiwe ”Sow 62 .e- we 52 gm .mm:
:3 65.3w? omewa Sam- we

woe 68.330 8:? $2 . _- we

93 ezeflww «Saw egg. 6

woo eweewmwé oeomw oeeoe- mm 662 x. .mm:

eee $8.948; mewow 3.3.? we

woe 23.8.35 5er 88.? we

8.0 womofleé owwmw mee- _e

8e ..mmeieew weme womee- mm me:
"~— 23: .9356 2.8.3.5...» 2.2m AOL 9.3.22.th 23.53;.

 

EA: 823% e5. 6.2 @6358

8%.: 36:. 335 >813 5 :wE E :95 omeeoEofl Sesame—E 82.: Be marge cofimfiwfl use $29-0 .bd 03.2.

44

Table 3.8. D-values and regression analysis for S. Senftenberg in high fat turkey breast

(HFB) meat (4.4% fat) containing NaCl and phosphate (PP)

 

 

 

 

 

 

Meat formulations Temperature Slope y-intercept D-value‘ R2
(”C) (min)

HFB 58 -0.1235 6.5853 8.10:1:O.13B 0.94
61 -02403 6.3106 4.16i0.09B 0.90
64 -1.8818 6.3910 1.13i0.020 0.95
66 -2.4908 6.2406 0403:001B 0.95

HFB, 1% NaCl 58 -0.0866 6.5049 11.55i1.00A 0.91
61 -03005 6.4121 3.33i0.32C 0.91
64 -1.6676 6.4424 0.60:1:0.19D 0.93
66 -3.3879 6.3746 03010.03C 0.94

HFB, 2% NaCl 58 -00797 6.6914 12.551253A 0.88
61 -O.2187 6.5732 4.57i0.73‘”"B 0.94
64 -0.6399 6.4341 1.56:1:006A 0.93
66 -1.8872 6.5272 0.53:0.09A 0.93

HFB, 1% NaCl, 0.25% PP 58 —0.0793 6.6285 1261:1231A 0.91
61 -O.2181 6.3910 4.591017A 0.95
64 -O.6658 6.2475 1.50i0.08A'B 0.91
66 -2.1728 6.4584 0.46fl:0.11A’B 0.93

HFB, 1% NaCl, 0.5% PP 58 -0.0781 6.6417 12.8i1.49A 0.87
61 -O.2182 6.4063 45810.14A 0.92
64 -O.7189 6.1527 1.393003B 0.91
66 -1.8801 6.1232 0.53:1:002A 0.90

 

’ D-values within the same temperature followed by the same letter are not difierent

(p>0.05). The D-value is expressed as the mean i the standard error of the mean (n=3).

45

The z-value of Salmonella in turkey averaged 5.3°C and was not influenced by

the addition of 1% NaCl , 2% NaCl, or phosphate (Table 3.4).

3.4.7. Effect of salt and phosphate on thermal inactivation of E. coli 0157:H7

Salt decreased the thermal stability of E. coli 0157:H7 in HFB. (Table 3.9)
Kotrola and Conner (1997) observed an increase in D-values for E. coli 01 57:H7 with
the addition of salt (8%) to 3% fat turkey meat. The different effect of NaCl on thermal
stability as compared to the present study may be due to the high amount added.
Phosphate had no effect on the thermal stability of E. coli 01 57:H7 in HFB. Kotrola and
Conner (1997) also added 0.5% polyphosphate as well as 4% sodium lactate to turkey
meat which increased the D-value at 57°C to 13.0 min. In our study, addition of 0.5%
PP to HFB containing 1% NaCl decreased the D-value to 11.8 min at 58°C. Kotrola and
Conner (1997) added 4% sodium lactate, which could cause the increase in D—values
instead of the polyphosphate.

The z-values of E. coli 0157:H7 were not affected by addition of salt or
phosphate (Table 3.4). Kotrola and Conner (1997) observed an increase in z-values with
the addition of NaCl. The higher z-values may be due to the higher amounts of NaCl

added (3% and 11%) than in the present study (1% and 2%).

3.4.8 Comparison of z-values of TPI, S. Senftenberg and E. coli 0157 :H7
Except for treatments containing phosphate, the z-values of TPI and S.
Senftenberg were within O.5°C of each other (Table 3.4). Veeramuthu et al. (1998) found

the z-values of TPI and S. Senftenberg to be within 0.2°C of each other in ground turkey

46

Table 3.9. D-values and regression analysis for E. coli 01 57:H7 in high fat turkey

breast (HFB) meat (4.4% fat) containing NaCl and phosphate (PP)

 

 

 

 

 

 

Meat formulations Temperature Slope y-intercept D-value‘I R2
(’0 (min)

HFB 55 -00517 6.8010 193421;] .19A 0.97
58 -0.2366 6.7638 4.23:1:0.10A 0.92
61 -1.0696 5.9209 0.938002A 0.90
64 -8.6690 6.4450 0.12:1:O.OZB 0.94

HFB, 1% NaCl 58 -0.0822 6.5163 12.17d:0.86C 0.91
61 -6.1010 6.3966 1.64zt0. 19B 0.92
64 -3.3638 6.8159 0.30i0.02B 0.98
66 -73253 6.7026 0.14ztO.OlB 0.98

HFB, 2% NaCl 58 -0.0662 6.7483 15.1 1:1:O.813 0.97
61 -O.6322 6.9703 1.582t0.188 0.91
64 -2.3576 6.9308 0.42:1:012B 0.98
66 -52909 7.0725 0.198002A 0.96

HFB, 1% NaCl, 0.25% PP 58 -0.0880 6.6912 11.36d:0.81c 0.92
61 -13431 6.9151 0.74i020" 0.88
64 .48926 6.8880 0.208004C 0.89
66 -8.0495 6.4971 0.12i0.01B 0.86

HFB, 1% NaCl, 0.5% PP 58 -0.0850 6.7122 1 1.76:1:097r 0.95
61 -0.7765 6.3845 1.29i0.09C 0.90
64 -33919 6.6731 0.298003B 0.95
66 -6.2621 6.4730 0.16i0.02A 0.93

 

 

a D-values within the same temperature followed by the same letter are not different

(p>0.05). The D-value is expressed as the mean :1: the standard error of the mean (n=3).

47

thigh meat. However, the z-values of E. coli 01572H7 found in the present study were
not close to the z—values of TPI (Table 3.4), whereas Veeramuthu et al. (1998) found 2-
values of E. coli 0157:H7 similar to the z-values of TPI. Veeramuthu et al. (1998) used

a different strain of E. coli 01 57:H7, which could account for the difference in z-values.

3.5. CONCLUSIONS

TPI was more heat resistant than Salmonella or E. coli OIS7:H7, suggesting the
enzyme will continue to function as an indicator, even after Salmonella and E. coli
0157:H7 have been inactivated during cooking. The stability of TPI was influenced by
NaCl and phosphate more than both Salmonella and E. coli 01 57:H7. TPI was more
thermally stable in thigh than breast meat, whereas muscle type did not affect the
thermal stability of Salmonella or E. coli 01 57:H7. This could be due to meat pH,
which would explain why the addition of phosphate increased the thermal resistance of
TPI. Phosphate raised the pH of the meat closer to the optimum pH of TPI. The
different muscle isoforrns of TPI could also have an effect on the thermal resistance of
TPI.

The z-values for both TPI and Salmonella were within O.5°C of each other when
compared in the same formulation, except when phosphate was present. Taken
together, these results suggest that TPI may be useful as a TTI to verify the adequacy of
a thermal process based on inactivation of Salmonella in turkey breast and thigh meat
containing salt, but not phosphate. It is possible that TPI can be used in products
containing potassium diphosphate by developing a mathematical model to relate the

lethality of Salmonella to the inactivation of TPI in the presence of a particular

48

phosphate. Both TPI and Salmonella were more heat resistant than E. coli OIS7:H7.
which suggests that a thermal process designed to eliminate Salmonella would also
eliminate E. coli 01 57:H7. Since Salmonella cannot be taken into a processing plant,
the endogenous muscle enzyme, TPI, appears to have potential for use as a TTI to
verify that the new lethality performance standards for turkey products are met during

processing.

49

CHAPTER 4
THE THERMAL RESISTANCE OF SALMONELIA SPP., ESCHERICHIA COLI

OIS7:H7, AND TRIOSE PHOSPHATE ISOMERASE IN GROUND BEEF

4.1 ABSTRACT
The effect of fat content on the thermal inactivation kinetics of E. coli OlS7:H7 and
triose phosphate isomerase (TPI) in ground beef and the effect of grth phase and
freezing on a Salmonella cocktail was studied. E. coli 01 57:H7 was more heat
resistant at 55°C (D=22.47 min) than the 8-strain Salmonella cocktail (D=16.34
min), however at 58, 61, and 63°C the heat resistance of these pathogens was
similar. Both TPI and E. coli 01572H7 were more heat resistant in high fat (19%)
than in low fat (5%) ground beef. The Salmonella cocktail was more thermally
stable in the stationary phase than in the log phase after inoculation into ground
beef. Thermal stability (D-values) of the Salmonella cocktail decreased after the
inoculated ground beef was frozen. TPI (z=6.54°C) was less temperature dependent
than E. coli 0157:H7 (z=3.74°C) and the Salmonella cocktail (z=4.03°C). When
designing challenge studies to verify adequate thermal processing of meat products,
a Salmonella cocktail should include serovars with high heat resistance. The
cocktail used in this study was less heat resistant than E. coli 01572H7 at 55°C,
which could result in a <5D reduction in E. coli 01572H7, and possible

consumption of unsafe beef.

50

4.2 INTRODUCTION

On January 6, 1999, the USDA-F SIS proposed regulations on the processing of meat
products, which switched the command and control regulations to performance standards.
The old regulations still meet the lethality performance standards proposed. For cooked
beef patties, the proposed lethality standard calls for a 5-log10 decrease in Salmonella spp.
(USDA-F SIS 1999). The beef patty lethality standard has not been finalized because of
concerns that this standard may not be adequate to ensure safety. The lethality
performance standard for roast beef requires a combination of thermal and non-thermal
processes that result in a 6.5 log“) decrease in Salmonella spp. (USDA-FSIS, 1999).

These performance standards allow the processor more flexibility to utilize various
thermal processes to produce a higher quality product while also eradicating pathogens.
However, challenge studies must be used to confirm that the thermal processes are
adequate. For challenge studies, the USDA-FSIS recommends using a Salmonella
cocktail or mixture of Salmonella serovars of high heat resistance which have been
implicated in food-borne outbreaks (USDA-FSIS 1999). Juneja et al. (1999) determined
D- and z-values for many strains of Salmonella in chicken broth and meat products to
determine which strains would be the most appropriate to use in challenge studies.

Thermal resistance of microorganisms is affected by intrinsic factors in meat such as
muscle type, pH, and fat content. The grth phase of the microorganism can also affect
heat resistance. Cells that had reached stationary phase (24h incubation in tryptic soy
broth) were shown to have maximum heat resistance (Heddleson et al., 1991). Frozen

cells may be more susceptible to heat due to formation of ice crystals that disrupt the cell

51

membrane (Doyle and Cliver, 1990; Smith, 1995). These factors must be taken into
account when using challenge studies to verify lethality performance standards.

In the early 1980’s, a enterohemorrhagic strain of Escherichia coli, strain 01 57:H7,
emerged as a serious threat to the safety of meat products. Since 1982, undercooked
ground beef has been the most common source of E. coli 01572H7 infection (CDC,
1999a). Because E. coli 01 57:H7 has been the cause of the most serious outbreaks with
contaminated ground beef patties, thermal inactivation studies are warranted.

A time-temperature integrator (TTI) undergoes an irreversible time/temperature
dependent change, which can be used to verify adequacy of thermal processing. In its
simplest form, a TTI must exhibit the same time-temperature dependent response as the
target pathogen when temperature is the only factor. The z-value (temperature increase
needed to decrease the thermal death time or D-value by 90%) can be used to determine
the viability of a TTI, such as an enzyme, if it mimics the z-value of the target pathogen.
Six endogenous enzymes were tested to determine if any could be used as TTIs in ground
beef (Orta-Ramirez et al., 1997). Triose phosphate isomerase (TPI) had a z-value of
556°C, whereas the z-value for E. coli 01 57:H7 was 559°C, and the z-value of S.
Senftenberg was 625°C. Because the temperature dependence of the thermal
inactivation rate constant of TPI was very similar to E. coli 01 57:H7 and S. Senftenberg,
the researchers suggested that TPI had potential for use as a TTI in ground beef.

The objective of this study was to 1) evaluate the effect of microbial grth phase
and freezing of inoculated ground beef on the heat resistance of an 8-strain Salmonella
cocktail; 2) compare the heat resistance of the Salmonella cocktail, E. coli 01 57:H7, and

triose phosphate isomerase in ground beef of two fat contents; and 3) assess the

52

applicability of using triose phosphate isomerase as an endogenous TTI to determine that

ground beef is adequately processed.

4.3 MATERIALS AND METHODS
4.3.1. Ground beef preparation

Low fat beef was obtained from Michigan State University Meat Laboratory (East
Lansing, MI 48824); high fat beef was obtained from a local processor. The beef was
ground twice through a 4 mm-diameter plate attached to a Kitchen Aid grinder (Model
KS-A, Hobart, Troy, OH), vacuum packaged in 60 g portions and frozen at -12°C. The
meat was transported on dry ice to Iowa State University and irradiated at 10kGy.
Sterility of the meat was tested by combining 1 g of meat with 9 mL of 0. 1% sterile
peptone water (Difco, Detroit, MI) and plating on Petrifilm® aerobic count plates (3M,
St. Paul, MN).

Fat, protein, and moisture analyse were conducted in triplicate using AOAC
(1996) Methods 991.36, 981.1, and 950.46B respectively. To determine pH, 10 g of
ground beef was combined with 90 g of distilled water and homogenized with a Polytron
homogenizer (Model PT 10/35, Brinkman Instruments, Westbury, NJ) for 30 s at speed
setting 3. The pH of the homogenate was measured in triplicate using a combination

electrode (Model 145, Coming, Medfield, MA).

4.3.2. Thermal inactivation experiments
Ground beef was thawed under cold running water for 5 min prior to each thermal

inactivation study. The meat was placed into a 60 mL syringe and 1 g meat was extruded

53

into a 5 x 25.5 cm polyethylene-laminated nylon pouch (Butcher and Packer Supply Co.
Detroit, MI 48207). The pouch was heat-sealed using a soldering iron, refrigerated and
used within 4 h. Bacterial inactivation experiments were performed using irradiated meat
that had been inoculated with either E. coli 01 57:H7 or the 8-strain Salmonella cocktail
in either log or stationary phase. Enzyme inactivation experiments were performed using
sterile meat. A type-T (copper-constantan) thermocouple (Part # TJ48-SCPSS-03ZG-
3.25-OST-M, Omega Engineering, Inc, Stamford, CT) was inserted into a bag of sterile
meat and sealed on both sides. The thermocouple was connected to a Danook Data
acquisition system equipped with a DBK 19 Thermocouple Card (Omega Engineering,
Inc.) The bags were placed into a Polystat circulator bath (Model 1268-52, Cole-Farmer
Instrument Company, Chicago, IL). The pouches were removed at the time intervals
required and immediately placed into an ice-water bath. Each experiment was replicated

three times using 8 sampling intervals at each cooking temperature.

4.3.3. Protein extraction for TPI activity

Cooked turkey was transferred to scintillation vials (Research Products
International Corp, Mount Prospect, IL 60056), combined with 3 volumes (w/v) ofO. 15
M NaCl, 0.01 M sodium phosphate buffer, pH 8.0 (PBS) and mixed for 30 s using a
Fisher Vortex Genie2TM (Scientific Industries, Inc., Bohemia, NY 11716). The extracts
were then stirred with a magnetic stir bar, without foaming, for 15 min at 4° C and
immediately centrifuged at 6000 x g for 10 min at 4°C. The supernatant was filtered
through Whatman no. 1 filter paper, held on ice and assayed for TPI activity within

24 h.

54

4.3.4. TPI activity determination

TPI activity was determined as described by Sair (1997). The reaction buffer
contained 1.0 mL 0.2M triethanloamine buffer, pH 8.0, 0.2 mL 15 mM glyceraldehyde-3-
phosphate, 10 ul B-nicotinamide adenine dinucleotide (10 mg/ml), 10 pl glycerol-3-
phosphate dehydrogenase (1.5 mg/ml), and 10 ul protein extract. The change in
absorbance was measured at 340 nm for 1 min at 25°C. The protein extract was diluted
with water as needed so that change in absorbance per minute was less than 0.2 (Norton

et al., 1970). TPI activity was calculated using the following equation:

Activity (U/L) = 1000 x A x v

 

TxExdx(0/D)

A = absorbance

V = assay volume, mL

T = time (min)

E = extinction coefficient (6.310 L x mmol-l x cm)
D = distance of light path (1 cm)

0 = sample volume (0.01 mL)

D = dilution of sample

4.3.5. Bacterial cultures and inoculation

Escherichia coli 01 57:H7 (ATCC 43 894) was frozen in TSB with 20% glycerol
at -80°C. The following eight Salmonella serovars were supplied by V.K. Juneja at the

Agricultural Research Service, Eastern Regional Research Center of the USDA: S.

55

thompson FSIS 120 (chicken isolate), S. enteriditis H3 527 (clinical isolate phage type
13A), S. enteriditis H3502 (clinical isolate phage type 4), S. Typhimurium H3380 (human
isolate DT104), S. hadar MP 60404 (turkey), S. copenhagen 8457 (pork), S. montevia’eo
FSIS 051 (beef) and S. heidelberg F503 8BG1 (human isolate). These strains were held at
-80°C in vials containing tryptic soy broth with 10% glycerol. The cultures were grown
separately in tryptic soy broth (TSB) (Difco Laboratories, Detroit, M1) at 37°C for 24 h
prior to use. New cultures were grown from the frozen stock cultures every 2 weeks. For
the thermal inactivation studies, 24 h cultures were transferred twice. Preliminary studies
showed that at 18-24h, the cultures were in log phase. Stationary phase occurred at 48h.

The E. coli 0157:H7 culture was pelletized at 6,000 x g for 10 min at 4°C. The
pellets were resuspended in sterile 0.1% peptone water. For the cocktail, equal amounts
of each strain were combined and pelletized at 6,000 x g for 10 min at 4°C and then
resuspended in sterile 0.1% peptone water to a concentration of 1 x 108 CFU/mL.

Meat (120 g) was inoculated to contain 1.0 x 107 CFU/g. The meat was mixed
manually under aseptic conditions to ensure even distribution. Dye was added to ensure
that mixing was adequate. Inoculated meat was used within 2 h or frozen at -12°C for 14
days prior to the thermal studies. For the stationary phase studies, 48 h old cultures were

inoculated into sterile ground beef.

4.3.6. Bacterial counts

Within 4 h of heat treatment, the cooked beef (1 g) was transferred into sterile 18
oz Whirlpak bagsTM and manually homogenized with 9 mL of 0. 1% sterile peptone water.
Appropriate dilutions were made with 0.1% sterile peptone water before both the

Salmonella cocktail and E. coli 01 57:H7 were plated in duplicate on Petrifilm Aerobic

56

Count Plates (3M, St. Paul, MN). All plates were counted after 24 h of incubation at

37°C.

4.3.7. Identification of Salmonella survivors

Preliminary experiments were conducted to determine the heating
time/temperature combination required to obtain <5 CFU/plate. After heating at 55°C for
75 min and 63°C for 50 s, survivors were plated on tryptic soy agar. Colonies were
isolated and then grown at 37°C for 24h on tryptic soy agar. Survivors were identified
using API 20 E test strips (Biomerieux Vitek, Inc., St. Louis, MO), which is an

identification system for Enterobacteriaceae and other Gram-negative rods.

4.3.8. Data analysis

D- and z-values were calculated assuming first-order kinetics (Pflug, 1997). D-
values (time required to reduce the microbial count or enzyme activity by 90%) were
calculated using linear regression analysis of time vs. log activity or log CFU/g with
Microsoft Excel (version 97). At least 4 data points with a correlation coefficient >0.90
using Microsoft Excel were used. The z-values (temperature increase needed to decrease
the thermal death time or D-value by 90%) were calculated using linear regression
analysis of temperature vs. log D-value with Microsoft Excel (version 97).

Statistical analysis of D- and z-values was conducted using one way analysis of
variance (SAS Institute Inc., 1995) on three replicates. The Tukey-Kramer HSD test was
used to compare D- and z-values with the mean square error at the 5% level of

probability.

57

4.4 RESULTS AND DISCUSSION
4.4.1. Proximate analysis

Low fat ground beef contained 4.8:] . 1% fat, 724:0.2% moisture, 15.8:2.6°/o
protein and had a pH of 5.65i0.01. High fat ground beef contained 19.1:0.7% fat,
63.4i0.6% moisture, 14.4i1.9% protein and had a pH of 5.72i0.01.

4.4.2. Thermal resistance of E. coli OlS7:H7

The D-values of E. coli 01572H7 in low and high fat ground beef were 20.08 min
and 22.47 min at 55°C, respectively (Table 4.1). However, values are 53-fold lower than
those found by Orta-Ramirez et al. (1997), which was 6.44 min at 58°C, who used TDT
tubes, which have a longer come-up time. With a longer come-up time, the initial
microbial load would be lowered, resulting in a lower slope of the regression line and a
higher D-value. Juneja et al. (1997) conducted thermal inactivation studies with a 4-
strain E. coli 01 57:H7 cocktail in 10% fat ground beef and obtained a D-value of 21 . 13
min at 55°C. This value is very close to those obtained in this study. However, at 60° C
Juneja et a1. (1997) calculated a D-value of 3.17 min, which is higher than 0.32 min
found at 61°C in the present study. The use of a larger sample size (3 g) could be the
cause of the higher D-value because the come-up time could be greater.

The z-values of E. coli in low and high fat beef were 3.79 and 360°C,
respectively. Orta-Ramirez et a1. (1997) reported a z-value of 559°C in ground beef that
was 3.8% fat and had a pH of 6.0. The higher z-value found by Orta-Ramirez et al.
(1997) could be due to the lower fat content, or the higher pH. The beef used by Orta-

Ranrirez et al. (1997) also contained only the sernitendinosus muscle, as compared to

58

Table 4.1. D- and z-values and regression parameters of E. coli 01 57:H7 in low
(4.8% fat) and high (19% fat) fat ground beefI

 

 

 

 

 

Treatment Temperature (°C) Slope y-intercept R2 D-value (min)"
Low fat 55 -0.05 6.91 0.95 20081045
58 -O.82 6.85 0.86 1.221013
61 -3.11 6.72 0.91 03210.02
63 -6.97 6.32 0.91 01610.02
High fat 55 -0.04 6.91 0.96 22.471078
58 -0.49 6.68 0.92 2.051013
61 -3.10 6.73 0.95 03210.01
63 -6.97 6.32 0.91 0.18+0.01
Treatment Slope y-intercept R2 z-value (°C)
Low fat -0.26 15.67 0.96 37910.10
High fat -0.28 16.53 0.99 36010.03

 

“ Values are expressed as the mean value 1 the standard error of the mean of three

replicate experiments

59

Table 4.2. D- and z—values and regression parameters of Salmonella
cocktail in fresh and frozen inoculated ground beef (19. 1% fat).
Cultures were inoculated when in the log and stationary growth

 

 

 

 

 

 

phasesa

Treatmentb Temperature Slope y-intercept R2 D-value (min)
(°C)

Cocktail 55 -0.06 7.08 0.98 16.341072
58 -0.37 7.04 0.97 2.721021
61 -2.25 6.95 0.98 0.441002
63 -6.54 6.93 0.97 0.15100]

Cocktail-frozen 55 -0.10 6. 58 0.92 9.851063
58 -0.70 6.78 0.86 1.431022
61 -3.41 6.58 0.76 02910.06
63 -7.18 6.21 0.81 0.14+0.03

Cocktail-stationary 55 -0.06 7.09 0.99 18.661065
58 -0.30 6.85 0.97 3.391006
61 -1.75 6.88 0.97 0.571003
63 -5.01 6.88 0.98 0.201001

Treatment Slope y-intercept R2 z-value (°C)

Cocktail -0.26 15 .29 1 3.901003

Cocktail-frozen -0.23 13.74 0.99 4.2910. 18

Cocktail-stationary -0.25 14.91 1 4.081002

 

‘ Values are expressed as the mean value 1 the standard error of the mean of three

replicate experiments

b Treatments are either the cultures in log phase inoculated into ground beef and used
immediately, cultures in log phase inoculated into ground beef and then fiozen at —9°C
for 14 days, or cultures in the stationary phase inoculated into ground beef and then used

in thermal inactivation studies

60

ground beef used in the current study, which contained numerous muscles. characteristic
of ground beef that would be consumed by the general public. These differences could
also cause the higher z-value. Juneja et al. (1997) reported a z-value of 598°C of a 4-

strain E. coli OIS7:H7 cocktail in ground beef.

4.4.3. Thermal resistance of Salmonella cocktail

The D—value for the log phase Salmonella cocktail in ground beef was 16.34 min
at 55°C and decreased to 0.15 min at 63°C (Table 4.2). Using the same 8-strain cocktail,
Juneja et al. (1999) found D-values of 8.65 min at 58°C and 1.5 min at 625°C. The
reason for the higher D-values could be due to a larger sample size (3 g) or different
plating methods used for enumeration. A larger sample size would have longer come-up
times resulting in higher D-values as explained previously. Juneja et al. (1999) used
tryptic soy agar for plating, whereas in the present study Petrifilm® aerobic count plates
were used. Significantly higher numbers of aerobic mesophilic bacteria were recovered
from ground beef by Linton et al. (1997) when using trypticase soy agar than Petrifilm®
aerobic count plates. The pH of the ground beef used in the Juneja study was 6.0. The
optimum pH for growth of Salmonella is 6.6 to 8.2 (Jay, 1996). The pH of the ground
beef used in our study was 5.7. A pH fiirther from the optimum, makes the
microorganisms more sensitive to toxic agents of a wide variety (Jay, 1996). Therefore,
because the pH of the ground beef used in the Juneja study was closer to the optimum pH
for Salmonella, it might be more thermally stable.

The Salmonella cocktail was more thermally stable when inoculated into ground

beef in the stationary phase than the log phase. Stationary phase occurs in

61

microorganisms when nutrients are depleted, following exponential growth (Kolter et al..
1993). Upon being starved, microorganisms enter a phase characterized by
metabolically less active and more resistant cells (Kolter et al., 1993). Changes that may
occur within the cell are a decrease in cell size as well as structural changes within the
cell wall (Kolter et al., 1993). E. coli cells that had been starved were more resistant to
heat shock (Jenkins et al., 1991). Staphylococcus aureus in retentates of milk
ultrafiltrates (Komacki and Marth, 1989), Pediococcns (Annous et al., 1999) and other
Salmonella spp. in a liquid menstruum containing food components (Heddleson et al.,
1991) also have been reported to be more stable in the stationary phase.

The log phase Salmonella cocktail was less heat resistant when tested in beef that
was held at -9°C for 14 days as compared to unfrozen fresh ground beef. At 58°C, the D-
value of the cocktail frozen in ground beef was 1.9 fold less than those not fi'ozen after
inoculation. Intracellular and extracellular ice crystals form during freezing that disrupt
the cell membrane and may lower the thermal stability of the cells (Doyle and Cliver,
1990; Smith, 1995).
4.4.4. Survivor identification

The API 20 E test strips contain 20 rnicrotubules filled with dehydrated
substrates. Identification of individual organisms is based on the reactions of the
individual substrates. The reaction results of API 20E test strips for the 8 strains of
Salmonella used in the cocktail are found in Table 4.3. The API 20E test strips could
differentiate between strains S. Thompson, S. Enteritidis H3527, and S. Montevideo.
However, S. Enteritidis H3 502 and S. Heidelberg (designated Group A) could not be

differentiated from each other. S. Typhimurium, S. Hadar, and S. Copenhagen

62

5:82 032%? n 1 £6382 328a u + ..

 

 

+ + + + 8:888 Noz .2
1 1 1 1 83686885030 .3
+ + + + 385983 notatioioaficosgm .om
-- -- -- -- 38283 coeweaecoeecoéfi .2
+ + + + 683:8: coseumonouficoEcom .E
1 1 1 1 A8833 cocevioEoseEoEuom .5
+ + + + 38:83: coceEonoueEoEom .e.
+ + + + £25.53 :ofiEonowflcoEuom .2
+ -- + + 2686:: ceeweaecoeewoesm .w.
+ + + + 23258 88362288882 .2
+ + + + 38623 someeEonowflaoEcom .m—
1 1 1 1 0358—00 .2
1 1 1 1 2.26365 8922. .o.
1 1 1 1 2.28.605 222: .m
1 1 1 1 82553 gunmen—bu. .w
1 1 1 1 082: s
+ + + + 88368 mf .e
+ + 1 + 533:3: 28:0 .m
+ + + + owe—xxofieoov 2.2250 .v
+ + + + owe—269.86% 2:93 .m
-- + + + 822356 2.8sz .m
1 1 1 1 omeEmoaoflewéom .—
8858? ..4 82: 22:26 ..e 82.: eeececm ..e coaeofi ..e .8862

 

2:25 536.63 336.56% 3 2:5 62 Mom 7?. mo 2288M .né «Sun.

63

532.0. 0>3em2. n 1 .5308. 33.50 H +

 

+ + + + 53050... «OZ .mm
1 1 1 1 owe—038-080.5030 ..N
+ + + + 350583 53.....on033550... .om
1 1 1 1 3.3.3383 535383038550". .0.
+ + + + $53.38. 003528538550... .m—

1 1 1 1 95.003 5323530380050... .2
+ + 855.... 53.....on038550... 0.
+ + 20.3.8. 532353038550... .2
1 1 + + 20.30.... 535.532.8550... .3

+ + 90.3.5.5 535383538550... .m .

+ + @5023 53528538550”. .2
1 1 1 1 82.3200 .2
1 1 1 1 53260.0 £0.03. .0.
1 1 1 1 53050... 0.0.2.. .0
1 1 1 1 82.3503 002.002.»...
1 1 1 1 080.3
+ + + + 53050... mum
+ 1 + + 533:3: 28.5
+ + + + owe—5.0.3.803 2.33.50
+ + + + owe—$003003 2.3.3
+ + -- -- 8266...... 2.5....
1 1 .. 1 ome3_m0.0e.em-e.0m

ANMV‘vixo'h'oo'

 

$2.320: .... 02030.52 ..m. 55:02.00 ..4 83.... .... .5383.

 

€086. mw 2.3

64

(designated Group B) also could not be differentiated using the API 20E test strips. S.
Thompson, S. Enteritidis H3 527, S. Montevideo, organisms in Group A, and the
organisms in Group B were all present afier being cooked in sterile ground beef at 55°C
for 75 min. At 63°C, S. Thompson, S. Enteritidis H3 527, organisms in Group A, and the
organisms from Group B were detected after cooking for 505. Based on these findings,
the heat resistance of the strains used in this cocktail were similar.
4.4.5. Thermal resistance of TPI

At all four temperatures tested, the D-values of TPI in high fat beef were higher
than in low fat beef (Table 4.4). At 53°C, the D-value of the 19.1% fat ground beef was
2.6 fold higher than the D-value of the 4.8% fat ground beef. Dita-Ramirez et al. (1997)
found higher D-values for TPI in 3.8% fat ground beef of 20. 13 min and 2.25 min at 58
and 63°C, respectively. The higher D-values could be due to differences in heating
methods or meat composition. Orta-Ramirez et al. (1997) used 10 mm diameter glass
tubes that have longer come-up times than the pouches used in our study. It would take
longer for the internal temperature to reach the target temperature, which would decrease
the initial enzyme activity, lower the slope of the regression line, and result in a higher D-
value. Orta-Ramirez et al. (1997) also used ground beef with a higher pH (6.0). TPI has
an optimal activity at pH 7-8. At pH 6.3 enzyme activity was about half that at pH 7-8.
(Beisenherz, 195 5). In ground beef with a higher pH, TPI would have more activity and
be more thermally stable.

The z-value of TPI in high fat beef was 6.54 as compared to 717°C in low fat

beef. Orta-Ramirez et al. (1997) found a z—value of 556°C, which indicates lower

65

Table 4.4. D- and z-values and regression parameters of triose phosphate isomerase in
low (4.8% fat) and high (19.1% fat) fat ground beef”

 

 

 

 

 

Treatment Temperature (°C) Slope y-intercept R2 D-value (min)
Low fat 53 -0.09 5.53 0.88 109212.46
55 -0. 14 3.77 0.97 7.021051
58 -044 3.72 0.98 2261019
61 -l.11 3.52 0.96 0.901008
High fat 53 -003 5.43 0.93 286513.93
55 -0.09 5.46 0.94 11.161128
58 -0.23 5.34 0.96 4.281022
61 -062 5.25 0.97 1.621004
Treatment Slope y-intercept R2 z-value (°C)
Low fat -0.14 8.46 0.99 7.171029
High fat -0. 15 9.51 0.99 6.541034

 

a Values are expressed as the mean 1 the stande error of the mean of three replicate

experiments

66

temperature dependence. Higher pH and fat content of the meat might decrease the
temperature dependence of TPI. The z-value of TPI was 1.74-fold higher than E. coli
0157:H7 in high fat ground beef and 1.6-fold higher than the Salmonella cocktail.
TPI could still be used as a TTI for Salmonella and E. coli OIS7:H7, if a
mathematical model can be developed to predict the thermal inactivation kinetics
(Hendriclor et al., 1995). Orta-Ramirez (1999) developed a non-linear mathematical

model for R-phycoerythrin (a fluorescent protein) for use as a TTI in beef products.

4.5 CONCLUSIONS

The fat content of ground beef did not affect the thermal stability of either E. coli
01 57:H7 or the 8-strain Salmonella cocktail. Therefore, fat concentration does not need
to be considered when designing adequate thermal processes for beef. Currently the
USDA-PSIS requires that a 6-8 strain Salmonella cocktail be used in challenge studies to
determine adequate processing of meat. The time required for a 6.5-log kill for the
Salmonella cocktail used in this study was 106.21 min at 55°C, 17.68 min at 58°C, 2.86
min at 61°C and 0.98 min at 63 °C. However, the time required for a 6.5-log kill in E.
coli 01572H7 was 146.06 min at 55°C. If the thermal processing schedule was based on
the Salmonella cocktail at 55°C, E. coli 01 57:H7 would not be reduced by 6.5 loglo. In
challenge studies, Salmonella serovars with higher thermal resistance should be used to
ensure that thermal processing schedules are adequate.

If the meat is frozen prior to cooking, less severe thermal processing is needed to

ensure a 6.5-log kill for Salmonella. However, if the cells are in stationary phase, more

67

thermal processing is needed to ensure a 6.5-log kill for Salmonella. When conducting
challenge studies, cultures in the stationary phase should be used.

Even though this study showed that TPI does not have the same z-value as
Salmonella, mathematical modeling can be done to predict the thermal inactivation

kinetics of Salmonella using TPI (Hendrickx et al., 1995).

68

CHAPTER 5

CONCLUSIONS

The USDA-F SIS has set lethality standards for thermal processing of meat
products based on destruction of Salmonella. To determine that a thermal process is
adequate, challenge studies using meat inoculated with Salmonella must be conducted by
the processor. Because of obvious safety and health concerns, it is not feasible to bring a
pathogenic microorganism into a meat processing facility to determine appropriate
thermal processing. There is a need for an alternative approach to verify that thermal
processing is adequate in meat processing facilities. This study examined environmental
effects on the thermal inactivation kinetics of Salmonella, E. coli 01 57:H7, and TPI, a
possible time-temperature integrator, in ground turkey and beef.

The effect of fat content, muscle type, salt and phosphate on the thermal
inactivation kinetics of Salmonella Senfienberg, E. coli 01 57:H7, and TPI in ground
turkey was studied. Overall, the D-values for TPI were more susceptible to
environmental effects than both S. Senftenberg and E. coli 01 57:H7. The D-values for
TPI were higher in thigh muscle than breast muscle and increased with higher fat content.
Addition of salt to high fat turkey breast meat decreased the D-values of TPI; however
adding phosphate to high fat breast meat containing 1% NaCl increased the D-values.
The D-values for S. Senfienberg were not affected by changes in muscle type; however,
increased fat content increased the D-values for S. Senftenberg at most of the
temperatures tested. Addition of 2% NaCl and phosphate to high fat turkey breast

increased the D-values for S. Senftenberg. The D-values for E. coli 01572H7 were not

69

affected by different muscle types or fat content. Adding salt increased the D-values for
E. coli 0157zH7, however the addition of phosphate had no effect.

The effect of fat on thermal inactivation kinetics of a Salmonella cocktail, E. coli
01 57:H7, and TPI in ground beef was also studied. At 55°C, the D-value for E. coli
01572H7 in high fat beef was higher than the D-value for the Salmonella cocktail. Fat
content did not have an effect on the D-values of either the Salmonella cocktail, or E. coli
01 57:H7. The D-values for TPI were higher in high fat (19.1%) beef than in low fat
(4.8%) beef. Freezing ground beef for 14 days at ~12°C decreased the D-values for the
Salmonella cocktail. D-values for the Salmonella cocktail in stationary phase were
higher than those for log phase.

In order for TPI to be a TTI, the z-value of TPI must be close to Salmonella in
either turkey or beef. The z-values of TPI in turkey, with the exception of the phosphate
treatments, were all within 05°C of the z-values for Salmonella. Therefore, it appears
that TPI would make a good TTI for cooked turkey products as long as phosphate is not
present. The z-values of TPI in beef were not similar to the z-values for Salmonella.
Hence, a mathematical model would be needed before TPI could be used as a TTI for

beef products.

70

CHAPTER 6

FUTURE RESEARCH

This study examined whether TPI could be used as an endogenous TTI in turkey
and beef products. Further research should be considered in the following areas.

All of the thermal death time studies were conducted using a meat model system.
While this system allows for accurate control of the testing environment, it is not
indicative of a “real world” setting. Studies need to be conducted using ground beef
patties and processed turkey products.

The USDA-FSIS calls for a Salmonella cocktail with 6-8 high heat resistant
serovars to be used in challenge studies. This study showed that the cocktail used in this
study, which contained 8 high heat resistant serovars, was less heat stable than E. coli
0157:H7 at some processing temperatures. Further research needs to be done to
determine what specific serovars should be used in challenge studies that would
guarantee thermal destruction of E. coli 01 57:H7 but not detract from the quality of the
processed meat product.

This study showed that mathematical models must be developed to use TPI as a
TTI in turkey products containing phosphate, or in any beef products. Once
mathematical modeling is complete, trials should be done to verify that the modeling is

accurate.

71

REFERENCES

AOAC. 1996. Official Methods of Analysis, 16‘“ ed. Arlington, Virginia: Association
of Official Analytical Chemists.

Abdul-Raouf, U.M., Beuchat, L.R., Ammar, MS. 1993. Survival and growth of
Escherichia coli 01572H7 in ground, roasted beef as affected by pH, acidulants,
and temperature. Appl. Environ. Microbiol. 59:2364-2368.

Ahmed, N.M., Conner, D.E., Huffman, UL. 1995. Heat-resistance of Escherichia coli
01572H7 in meat and poultry as affected by product composition. J. Food Sci.
60:606-610.

Annous, B.A., Kozempel, M.F., Kurantz, MJ. 1999. Changes in membrane fatty acid
composition of Pediococcus sp. Strain B-2354 in response to grth conditions
and its effects on thermal resistance. Appl. Environ. Microbiol. 65:2857-2862.

Bean, NH. and Grifin, P.M. 1990. Foodbome disease outbreaks in the United States,
1973-1987: Pathogens, vehicles and trends. J. Food Prot. 53:804-817.

Beisenherz, G. 1955. Triosephosphate isomerase from calf muscle. Ch. 57, In Methods
in Enzymology, S.P. Colowick and NO. Kaplan (Ed), p. 387-391. Academic
Press, Inc. New York, NY.

Blackburn, C.W., Curtis, L.M., Humpheson, L., Billon, C., McClure, P.J. 1997.
Development of thermal inactivation models for Salmonella Enteritidis and
Escherichia coli 01 57:H7 with temperature, pH, and NaCl as controlling factors.
Intern. J. Food Microbiol. 38:31-44.

CDC. 1999a. “Escherichia coli 01 57:H7.”

72

<http://www.cdc.gov/ncidod/dbmd/diseaseinfo/escherichiacoli_g.htm> (09 April
1999 )

CDC. 1999b. “Salmonellosis.”
<http://www.cdc.gov/ncidod/dbmd/diseaseinfo/salmonellosis_g.htm> (09 April
1999)

CDC. 1995. Escherichia coli 0157:H7 outbreak linked to commerCially distributed dry-
cured salami -— Washington and California, 1994. Morb. Mortal. Wkly. Rep.

44: 157-160.

Carter, A.O., Borczyk, A.A., Carlson, J .A., Harvey, B., Hockin, J .C., Karmali, M.A.,
Krishnan, C., Korn, D.A., Lior, H. 1987. A severe outbreak of Escherichia coli
0157:H7 associated with hemorrhagic colitis in a nursing home. N. Engl. J. Med.
317: 1496-1500.

Conner, DE, and Kotrola, J .S. 1995. Grth and survival of Escherichia coli 0157:H7
under acidic conditions. Appl. Environ. Microbiol. 61 :382-385.

Doyle, M.P., Cliver, DO. 1990. Salmonella. Ch. 11 in Foodborne Diseases. D.O.
Cliver. (Ed), p. 186-204. Academic Press, Inc., NY.

Doyle, M.P., Schoeni, J .L. 1984. Survival and grth characteristics of Escherichia coli
associated with hemorrhagic colitis. Appl. Environ. Microbiol. 48:855-856.

Fain, A.R., Line, J .E., Moran, A.B., Martin, L.M., Lechowich, R.V., Carosella, J .M.,
Brown, W.L. 1991. Lethality of heat to Listeria monocytogenes Scott A: D-
value and z-value determinations in ground beef and turkey. J. Food Prot.

54:756-761.

73

Goodfellow, S.J., Brown, W.L. 1978. Fate of Salmonella inoculated into beef for
cooking. J. Food Prot. 41:598-605.

Griffin, P.M., Tauxe, R.V. 1991. The epidemiology of infections caused by Escherichia
coli 0157:H7, other enterohemorrhagic E. coli and the associated hemolytic
uremic syndrome. Epidemiol. Rev. 13:60-98.

Heddleson, R.A., Doores, S., Anantheswaran, R.C., Kuhn, G.D., Mast, MG. 1991.
Survival of Salmonella species heated by microwave energy in a liquid
menstruum containing food components. J. Food Protec. 54:63 7-642.

Hendrickx, M., Maesmans, G., De Cordt, S., Noronha, J ., Van Loey, A., Tobback, P.
1995. Evaluation of the integrated time-temperature effect in thermal processing
of foods. Crit. Rev. Food Sci. Nutr. 35:231-262

Hsu, Y.C. 1997. Identification and verification of an endogenous time-temperature
indicator to determine processing adequacy of roast beef. Ph.D. Dissertation.
Michigan State University, East Lansing, MI.

Hsu, Y.C., Sair, A.I., Booren, A.M., Smith, D.M. 2000. Triose phosphate isomerase as
an endogenous time-temperature integrator to verify adequacy of roast beef
processing. J. Food Sci. 65:236-240.

Jay, J.M. 1996. Modern Food Microbiology, 5th ed. Chapman & Hall, New York, NY.

Jenkins, D.E., Auger, E.A., Matin, A. 1991. Role of RpoH, a heat shock regulator
protein, in Escherichia coli carbon starvation protein synthesis and survival. J.
Bacteriol. 173: 1992-1996.

Juneja, V.K., Eblen, B.S., Manner, B.S., Williams, A.C., Palumbo, S.A., Miller, A].

1995. Thermal resistance of nonproteolytic type B and type B Clostridium

74

botulinum spores in phosphate buffer and turkey slurry. J. Food Prot. 58:75 8-
763.

Juneja, V.K., Snyder, O.P., Marmer, BS. 1997. Thermal destruction of Escherichia coli
0157:H7 in beef and chicken: determination of D- and z-values. Intern. J. Food
Microbiol. 35:23 1-23 7.

Juneja, V.K., Marmer, BS. 1999. Lethality of heat to Escherichia coli 0157:H7: D-
and z-value determinations in turkey, lamb, and pork. Food Research Intern.
32:23-28.

Juneja, V.K., Eblen, B.S., Ransom, GM. 1999. Thermal inactivation of Salmonella spp.
in chicken broth, beef, pork, turkey, and chicken: determination of D- and 2-
values. Submitted to the Journal of Food Science.

Juneja, V.K., Eblen, BS. 2000. Heat inactivation of Salmonella Typhimurium DT104 in
beef as affected by fat content. Lett. Appl. Microbiol. 30:461-467.

Kolter, R., Siegele, D.A., Tormo A. 1993. The stationary phase of the bacterial life
cycle. Annu. Rev. Microbiol. 47:855-874.

Komacki, J .L., Marth, EH. 1989. Thermal inactivation of Staphylococcus aureus in
retentates from ultrafiltered milk. J. Food Prot. 52:63 1-63 7.

Kotrola, J.S., Conner, DE. 1997. Heat inactivation of Escherichia coli 0157:H7 in
turkey meat as affected by sodium chloride, sodium lactate, polyphosphate, and
fat content. J. Food Prot. 60:898-902.

Kotrola, J .S., Conner, D.E., Mikel, W.B. 1997. Thermal inactivation of Escherichia coli

0157:H7 in cooked turkey products. J. Food Sci. 62:875-877,905.

75

Line, J .E., Fain, A.R., Moran, A.B., Martin, L.M., Lechowich, R.V., Carosella, J.M.,

‘ Brown, W.L. 1991. Lethality of heat to Escherichia coli 0157:H7: D-value and
z-value determinations in ground beef. J. Food Prot. 54:762-766.

Linton, R.H., Eisel, W.G., Muriana, P.M. 1997. Comparison of conventional plating
methods and Petrifilm for the recovery of microorganisms in a ground beef
processing facility. J. Food Prot. 60: 1084-1088.

Ng, H., Bayne, H.G., Garibaldi, J .A. 1969. Heat resistance of Salmonella: The
uniqueness of Salmonella Senftenberg 775W. Appl. Microbiol. 17:78-82.

Norton, I.L., Pfuderer, P., Stringer, C.D., Hartman, EC. 1970. Isolation and
characterization of rabbit muscle triose phosphate isomerase. Biochem. 9:4952-
4958.

Orta-Ramirez, A. 1999. A fluorescence-based time-temperature integrator for
monitoring thermal processing in beef products. Ph.D. Dissertation. Michigan
State University, East Lansing, MI.

Orta-Ramirez, A., Price, J .F., Hsu, Y.C., Veeramuthu, G.J., Cherry-Merritt, J .S., Smith,
D.M. 1997. Thermal inactivation of Escherichia coli 0157:H7, Salmonella
Senftenberg, and enzymes with potential as time-temperature indicators in ground
beef. J. Food Prot. 60:471-475.

Padhye, N.V., Doyle, MP 1992. Escherichia coli 0157:H7: Epidemiology,
pathogenesis, and methods for detection in food. J. Food Prot. 55:555-565.

Pflug, 1.]. 1997. Evaluating a ground-beef patty cooking process using the general

method of process calculation. J. Food Prot. 60: 1215-1223.

76

Riley, L.W., Remis, R.S., Helgerson, S.D., McGee, H.B., Wells, J .G., Davis, B.R.,
Hebert, R.J., Olcott, E.S., Johnson, L.M., Hargett, N.T., Blake, P.A., Cohen, ML.
1983. Hemorrhagic colitis associated with a rare Escherichia coli serotype. N.
Engl. J. Med. 3081681-685.

SAS Institute, Inc. 1995. JMP® Introductory Guide, Version 3. SAS Institute Inc,
Cary, NC.

Sair, A. 1992. Verification of indicator proteins and color measurements as methods to
determine adequate thermal processing of ground beef patties. MS. Thesis.
Michigan State University, East Lansing, MI.

Sair, A.I., Booren, A.M., Berry, B.W., Smith, D.M. 1997. Residual triose phosphate
isomerase activity and color measurements to determine adequate cooking in
ground beef patties. J. Food Prot. 62: 156-161.

Schoeni, J .L., Brunner, K., Doyle, MP. 1991. Rates of thermal inactivation of Listeria
monocytogenes in beef and fermented beaker sausage. J. Food Prot. 54: 334-
337.

Schuman, J.D., Sheldon, B.W. 1997. Thermal resistance of Salmonella spp. and Listeria
monocytogenes in liquid egg yolk and egg white. J. Food Protc. 60:634-638.

Smith, MG. 1995. Survival of E. coli and Salmonella after chilling and freezing in
'liquid media. J. Food Sci. 60:509-512.

Townsend, W.E., and Blankenship, LC. 1989. Methods for detecting processing
temperature of previously cooked meat and poultry products - a review. J. Food

Prot. 52:128-135.

77

USDA. 1986a. Determination of internal cooking temperature (Coagulation). In
Revised Basic Chemistry Laboratory Guidebook No. 3.019, pp. 3-55. Science
Chemistry Div., Food Safety and Inspection Service, US. Dept. of Agriculture,
Washington, DC.

USDA. 1986b. Determination of internal cooking temperature (Acid phosphatase
activity). In Revised Basic Chemistry Laboratory Guidebook No. 3.018, pp. 3-49.
Science Chemistry Div., Food Safety and Inspection Service, US. Dept. of
Agriculture, Washington, DC.

USDA, 1989. Performing the catalase enzyme test. A self-instruction guide. Technical
Services Training Div., Food Safety and Inspection Service, US. Dept. of
Agriculture, Washington, DC.

USDA-FSIS. 1999. Performance standards for the production of certain meat and
poultry products. US. Department of Agriculture, Food Safety Inspection
Service, Washington, DC. Federal Register 64:732-749 (January 6).

USDA-FSIS. 2000. “Interim progress report on Salmonella testing of raw meat and

3’

poultry products. <http://www.fsis.usda.gov/ophs/haccp/salmraw.htm> (09
October 2000)

Veeramuthu, G.J., Price, J .F., Davis, GB, Booren, A.M., Smith, D.M. 1998. Thermal
inactivation of Escherichia coli 0157:H7, Salmonella Senftenberg, and enzymes
with potential as time-temperature indicators in ground turkey thigh meat. J.
Food Prot. 61:171-175.

Wang, S.F., Abouzied, M.M., Smith, D.M. 1996. Proteins as potential endpoint

temperature indicators for ground beef patties. J. Food Sci. 61 :5-7.

78

Zhao, T., Doyle, M.P., Besser, RE. 1993. Fate of enterohemorrhagic Escherichia coli
0157:H7 in apple cider with and without preservatives. Appl. Environ.

Microbiol. 59:2526-2530.

79

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

11111111111111le11111111111