PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES Mum on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU I. An Affirmative Action/EM Opportunity Initiation
W

m1

 

IDENTIFICATION AND VERIFICATION OF AN ENDOGENOUS
TIME-TEMPERATURE INTEGRATOR TO DETERMINE
PROCESSING ADEQUACY OF ROAST BEEF
By

Ivy Yih-Chih Hsu

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1997

ABSTRACT
IDENTIFICATION AND VERIFICATION OF AN ENDOGENOUS
TIME-TEMPERATURE INDICATOR TO DETERMINE
PROCESSING ADEQUACY OF ROAST BEEF
By

Ivy Yih-Chih Hsu

The USDA-F SIS requires specific thermal processing schedules for roast beef
products to ensure the destruction of pathogens. A thermal process to achieve a 7-D
reduction in Salmonella was recently proposed by the F818 as the lethality performance
standard for roast beef and cooked beef products. Triose phosphate isomerase (TPI) had
a 2 value Similar to that of Salmonella and was identified as a potential endogenous time-
temperature integrator (TTI) to verify adequacy of roast beef processing based on
previous studies. The overall goal of this study was to determine if TPI could be used as
a TN to verify compliance to USDA processing schedules. The ability of TPI to indicate
thermal processing adequacy was investigated in a ground beef water bath model system
and roast beef pilot studies at MSU. Three adequate temperature schedules, [62.2 °C/5
min (high), 58.3 °C/24 min (medium), and 54.4 °C/ 121 min (low)] and corresponding
inadequate processing schedules (0.5 and 1 log cycle reduction in holding time from each
adequate process) were used. In the ground beef water bath model system, TPI activity
averaged 2.6 U/kg in adequately processed ground beef and increased when processing
time was inadequate. In the MSU pilot plant study, TPI activities were similar when
roasts were adequately processed at low and high temperatures, but TPI activity increased

as proCessing time was decreased in the high temperature process only. Bovine muscle

TPI was purified and polyclonal antisera (PAb) were raised in rabbits. A sandwich
ELISA was developed and cross reactivities of antibodies with TPI from difierent animal
species, muscle protein concentrates, and common meat starter cultures were examined.
Ground beef was heated in tubes to internal temperatures of 48.9 °C to 76.7 °C in 5.5 °C
increments and extracted with PBS buffer. TPI concentration and activity in extracts of
cooked ground beef decreased as temperature of water bath was increased. Both the
ELISA and enzyme assay were able to detect differences in TPI due to cooking
temperature of beef. In a commercial pilot plant study, no differences in TPI
concentration and activity were found between adequately and inadequately processed
roast beef. This may be due to inconsistent processing conditions within the smokehouse
facilities or to variations in size and shape of roasts. Further research is needed to
determine if TPI can be used as a 'I'I‘I under commercial processing conditions.

KEYWORDS- beef, endpoint temperature, processing

Copyn'ght by
Ivy Yih-Chih Hsu
l 997

This work is dedicated to my beloved parents,

Mr. Ling-Yun Hsu and Mrs. Jean-Da Lee,

Mr. John Chang, who supported this “dreamer ”

and the One I believe in.

Their unlimited love made this work happen.

ACKNOWLEDGMENTS

I wish to acknowledge all those who made this dissertation a reality. I would like
to express my deepest appreciation to my major professor, Dr. Denise M. Smith for her
guidance throughout my doctoral program. I “enjoyed” the 30 min meeting every Friday
morning with her and believe that made constructive impact toward my academic and
future goal. Sincere appreciation is given to my guidance committee, Dr. Alden M.
Booren, Dr. James J. Pestka, and Dr. Steven Bursian for their guidance, understanding,
encouragement, inspiration, and support throughout her study period. Sincere thanks are
extended to Dr. Robert Tempelman for his assistance in statistical analysis of this study.

Very special thanks are given to two wonderful friends, Dr. Stephanie Smith, for
her strong moral support, and shared, learned, experiened this part of our life together.
The author experiences an unique level of understanding with Mr. Arnie Sair that allowed
me to share a delightful and memorable fiiendship working together on the “TPI” project.

Sincere thanks are given to Dr. Cheng-Hsing Wang and Dr. Virginia Vega-
Wamer for their expertise, fiiendship and inspiration during the author’s study.

Special thanks are also given to Dr. A.M Booren and Mr. Tom Forton for their
help in roast beef processing. I thank Dr. William Schwartz, Dr. Paul Benthal, and Mr.

Tom Mattawitz fi'om Bil-Mar Foods for kindly donating roasts and allowing me to

vi

conduct my research in a commercial facility. The author also thanks Dr. James Price for
his help in the microbial Study.

Special thanks are also given to Dr. Arti Arora, Ms. Mridvika, MS. Alicia Orta-
Ramirez, Dr. Giri Veeramuthu, Ms. Shelli Pfeifer, Dr. James R. Clarke, Ms. Sarah Smith,
Ms. Tammy Zielinski, Dr. Shaun Chen, Ms. Manee Vittayanont, Dr. Choi-Lan Ha, Ms.
Vance Chonhenchob, Mr. Jay Chick, Mr. Matthew Rarick, Mr. Mike Miller. Their
friendship will never be forgotten.

Finally, I wanted to thank my sisters, Ms. Judy Hsu and Ms. Grace Hsu for having

such significant impacts on my life.

 

This material is based upon work supported by the Cooperative State Research,
Education and Extension Service, US. Department of Agriculture, under Agreement Nos.
96-35201-3343 and 92-37201-8100. Any opinions, findings, conclusions, or
recommendations expressed in this document are those of the author and do not

necessarily reflect the view of the US. Department of Agriculture.

 

vii

TABLE OF CONTENTS

List of Tables ..................................................................................... xiv
List of Figures .................................................................................... xvii
Chapter 1. Introduction .. ................................................................... 1
Chapter 2. Literature Review ............................................................... 8

2.1 USDA-FSIS Thermal Regulations for Meat Products 8
2.2 The USDA-F SIS Proposed Regulations to Amend the Current
Meat and Poultry Inspection Regulations 10

2.3 Current USDA Methods for Endpoint Temperature Determination 11

2.3.1 Bovine Catalase Activity Test ....................................... 12
2.3.2 Protein "Coagulation Test" ............................................ 12
2.3.3 Acid Phosphatase Activity Test ...................................... 13

2.4 Alternative Methods for EPT Determination 14

2.5 Enzymatic Methods for EPT Determinations in Meat Products ........... 14
2.5.1 Bovine Meat ............................................................ 15
2.5.2 Porcine Meat ....................................................... 16
2.5.3 Poultry Meat ............................................................ 16

2.5.4 Comparison of EPT Indicators in Bovine, Porcine and Poultry Meats 18

2.6 Immunoassays for EPT Determinations in Meat Products 18

viii

2.7

2.8

2.9

Chapter 3.

3.1
3.2

3.3

2.6.1 Bovine Meat ............................................................ 19

2.6.2 Poultry Meat ............................................................ 20
Thermal Death Time (TDT) Studies .......................................... 21
2.7.] Definitions .............................................................. 22

2.7.2 TDT Studies in Various Meat Systems 24

2.7.2.1 Ground Beef ................................................... 24
2.7.2.2 Water-cooked Beef ........................................... 31
2.7.2.3 Dry Roasted Beef ............................................. 33
2.7.2.4 Canned Ham ................................................... 33
2.7.2.5 Pork Sausage .................................................. 34

2.7.2.6 Ground Turkey Breast Muscle 34
2.7.2.7 Chicken Muscle .............................................. 35
Time-Temperature Integrators (TI'I) As Thermal Processing
Indicators ........................................................................ 35
The Biochemical Properties of Triose Phosphate Isomerase (TPI) ..... 38
2.9.1 Determination of TPI Activity in Muscle Tissues 40
2.9.2 Purification of TPI from Muscle Tissues ......................... 41

Thermal Inactivation of Acid Phosphatase and Peroxidase in Ground

Beef ................................................................................. 45
Abstract ........................................................................ 45
Introduction ..................................................................... 46
Materials and Methods ........................................................ 48

3.4

Chapter 4.

4.1

4.2

4.3

4.4

3.3.1 Sample Preparation .................................................. 48
3.3.2 Thermal Treatments .................................................. 48

3.3.3 Preparation of Protein Extracts and Enzyme Assays 49

3.3.4 Calculations of D and 2 Values ..................................... 50
3.3.5 Proximate Analysis ................................................... 50
Results and Discussion ....................................................... 50

Identification of Marker Proteins in Roast Beef to Verify Compliance

to USDA Processing Schedules .............................................. 56
Abstract ......................................................................... 56
Introduction .................................................................... 57
Materials and Methods ........................................................ 59
4.3.1 Pilot Studies .......................................................... 59

4.3.1.1 Roast Beef Processing ...................................... 59

4.3.1.2 Proximate Composition ..................................... 64

4.3.1.3 Preparation of Protein Extracts and Enzyme Assays 64
4.3.2 Model System Studies ................................................ 65
4.3.2.1 Ground Beef Cooking ....................................... 65
4.3.2.2 Preparation of Protein Extracts and Enzyme Assays 66
4.3.3 Statistical Analysis ................................................... 67

Results and Discussion ....................................................... 67

Chapter 5.

5.1

5.2

5.3

5.4

Development of an ELISA to Quantify Triose Phosphate Isomerase

in Cooked Beef .................................................................. 77
Abstract ........................................................................... 78
Introduction ...................................................................... 78
Materials and Methods ......................................................... 80
5.3.1 Purification of TPI ..................................................... 80
5.3.2 TPI Enzyme Assay .................................................... 82

5.3.3 Bradford Soluble Protein Concentration Determination 82

5.3.4 Immunization and Polyclonal Antibody Production 84

5.3.5 Indirect ELISA (Titer Determination) ............................... 84
5.3.6 Electrophoresis ......................................................... 85
5.3.7 Western Blotting ....................................................... 86
5.3.8 Biotinylation of Polyclonal Antibodies ............................ 87
5.3.9 Sandwich ELISA ...................................................... 88

5.3.10 Cross Reactivities of TPI from Different Animal Species, and
Commercial Whey Protein and Plasma Protein Concentrates 89

5.3.11 Cross Reactivity ofTPI from Starter Cultures 90

5.3.12 Ground Beef Water Bath Model Study ............................ 91

5.3.13 Preparation ofProtein Extracts and Enzyme Assays . . 92

5.3.14 Statistical Analysis .................................................... 92
Results and Discussion ......................................................... 92
5.4.1 Purification of TPI ...................................................... 92

xi

5.4.2 Production of Polyclonal Antibody and ELISA Development .. 96

5.4.3 Cross Reactivity of TPI Polyclonal Antibodies with TPI from
Different Species ......................................................... 98

5.4.4 Cross Reactivity of TPI Polyclonal Antibodies with TPI from
Different Sources .................................................... 103

5.4.5 Cross Reactivity of TPI Polyclonal Antibodies with TPI from

Starter Cultures ....................................................... 104
5.4.6 Ground Beef Water Bath Model Study ........................... 105
5.5 Conclusions ................................................................... 107

Chapter 6. Verification of Triose Phosphate Isomerase (TPI) Enzyme-Linked

Immunosorbent Assay (ELISA) to Determine Processing Endpoint

Temperatures of Roast Beef in a Pilot Study .............................. 111

6.1 Abstract ........................................................................ 111
6.2 Introduction .................................................................. 112
6.3 Materials and Methods ...................................................... 114
6.3.1 Processing Schedules .............................................. 114

6.3.2 Extraction of TPI ................................................... 119

6.3.3 TPI Enzyme Activity and Concentration ........................ 119

6.3.4 Statistical Analysis ................................................. 119

6.4 Results and Discussion ...................................................................... 120
6.5 Conclusions ......................................................................................... 124

Chapter 7. Conclusions .................................................................. 125

xii

Chapter 8. Future Research ................................................................ 127
References ...................................................................................... 129

Appendix .................................................................................................................... 1 3 8

xiii

Chapter 8. Future Research ................................................................ 127
References ...................................................................................... 129

Appendix .................................................................................................................... 138

xiii

LIST OF TABLES

Table
2.1 Minimum USDA thermal processing schedules for cooked beef and roast

beef ................................................................................................................... 9
2.2 Thermal death time curve data and 2 values for Salmonella serotypes in

a ground beef system ....................................................................................... 25
2.3 Time-temperature schedules for a 7-D reduction of Salmonella in cooked

beef .................................................................................................................. 26
2.4 Thermal death time curve data and 2 values for L. monocytogenes Scott A

in lean (2% fat) and fatty (30.5% fat) ground beef systems ........................... 28
2.5 Thermal death time curve data and 2 values for Escherichia coli: 0157:

H7 in lean (2% fat) and fatty (30.5% fat) ground beef systems .................... 30
2.6 Thermal death time curve data and 2 values for Escherichia coli 0157:

isolate 204P in ground beef, pork sausage, turkey breast and chicken meat .. 32

3.1 D values and regression analysis for acid phosphatase activity from

ground beef heated at 53, 58, 63 and 68 0C ................................................... 51

3.2 D values and regression analysis for peroxidase activity fiom ground

beefat 53, 58, 63 and 68 ° C ......................................................................... 52

3.3 Z values and regression analysis from thermal death time studies of acid

xiv

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8

4.9

5.1

phosphatase (AP) and peroxidase (PO) from ground beef .......................... 54
Processing times and temperatures used to prepare adequately and
underprocessed beef roasts ......................................................................... 60
Smokehouse schedule used to prepare low temperature (54.4 °C; 130 °F)
processed roast beef ..................................................................................... 61
Smokehouse schedule used to prepare medium temperature (58.3 °C;

137 °F) processed roast beef ........................................................................ 62
Smokehouse schedule used to prepare high temperature (62.2 °C; 144 °F)
processed roast beef .................................................................................... 63
Peroxidase activity (U/kg meat) in adequately processed and

underprocessed roast beef prepared in a smokehouse ................................ 69
Acid phosphatase activity (U /kg meat) in adequately processed and
underprocessed roast beef prepared in a smokehouse ................................ 70
Lactate dehydrogenase concentration (ug/ g meat) in adequately

processed and underprocessed roast beef prepared in a smokehouse ....... .. 72
Triose phosphate isomerase activity (U /kg meat) in adequately processed

and underprocessed roast beef prepared in a smokehouse ......................... 74
Triose phosphate isomerase activity (U /kg meat) of beef processed using

three USDA approved schedules for roast beef and underprocessed by
reducing the processing time by 0.5 and 1.0 log cycle in a ground beef

water bath model system ............................................................................. 75

Purification of triose phosphate isomerase from bovine semimembranosus

XV

5.2

5.3

6.1

6.2

6.3

6.4

6.5

6.6

muscle ........................................................................................................ 94
Polyclonal antibody titers (serum dilution) against bovine triose

phosphate isomerase from semimembranosus muscle .............................. 97
Effect of cooking temperature on triose phosphate isomerase activity

(U/kg meat) and concentration (ug/ g meat) of bovine meat cooked

in a water bath ....................................................................... 106
Processing times and temperatures used to prepare adequately and
inadequately processed beef roasts ............................................................ 115
Smokehouse schedule used to prepare low temperature (54.4 °C;

130 °F) processed roast beef ...................................................................... 116
Smokehouse schedule used to prepare medium temperature (5 8.3 °C;

137 °F) processed roast beef ...................................................................... 117
Smokehouse schedule used to prepare high temperature (62.2 °C;

144 °F) processed roast beef ...................................................................... 118
Triose phosphate isomerase activity (U/kg meat) of roast beef processed
using three USDA approved schedules for roast beef and inadequately
processed by using the processing time by 0.5 and 1.0 log cycle in
commercially processed roasts .................................................................. 121
Triose phosphate isomerase concentration ( ug/ g meat) of roast beef
processed using three USDA approved schedules for roast beef and
inadequately processed by decreasing the processing time by 0.5 and 1.0

log cycle in commercially processed roasts ............................................. 122

xvi

Figure

5.1

5.2

5.3

LIST OF FIGURES

Representative sodium dodecyl sulfate-polyacrylamide gel electrophoretogram
of muscle extracts from the triose phosphate isomerase (TPI) purification
procedures. Proteins were stained with Coomassie Blue. (lane 1) molecular
weight standard; (lane 2) porcine TPI (from Sigma); (lane 3) bovine muscle
homogenate; (lane 4) 55% acetone precipitate fraction; (lane 5) 90% ammonium
sulfate precipitate fraction; (lane 6) afier carboxylmethyl cellulose (CMC)
column chromatography ................................................................................... 95
Cross-reactivity of bovine anti-triose phosphate isomerase (TPI) polyclonal
antibodies with TPI from different animal species by sandwich ELISA. The
standard deviation values that were less than 0.02 absorbance units were not
plotted on the figure ......................................................................................... 99
Western blot of bovine, dog, rabbit, and porcine muscle triose phOSphate
isomerase (TPI) and bovine muscle extract electrophoretically transferred

from a native polyacrylamide gel to a nitrocellulose membrane hybridized

with anti-bovine TPI polyclonal antibodies; (lane 1; 5 pg protein) dog

xvii

5.4

5.5

5.6

muscle TPI; (lane 2; 5 pg protein) rabbit muscle TPI; (lane 3; 5 pg protein)
porcine muscle TPI; (lane 4; 5 pg protein) bovine muscle T'PI; (lane 5; 100 pg
protein and lane 6; 200 pg protein) bovine raw muscle

extracts ....................................................................................................... 100
Western blot of bovine, dog, rabbit, and porcine muscle triose phosphate
isomerase (TPI) electrophoretically transferred from a sodium dodecyl
sulfate-polyacrylamide gel to a nitrocellulose membrane hybridized with
anti-bovine TPI polyclonal antibodies; (lane 1; 100 pg protein) dog muscle

TPI; (lane 2; 5 pg protein) rabbit muscle TPI; (lane 3; 5 pg protein)

porcine muscle TPI; (lane 4; 5 pg protein and lane 5; 10 pg protein)

increasing concentration of bovine TPI; (lane 6 - 9) decreasing amount of

bovine raw muscle extracts (lane 6; 200 pg protein; lane 7; 100 pg protein;

lane 8; 75 pg protein; lane 7; 30 pg protein) ................................................ 101
Cross-reactivity of bovine triose phosphate isomerase (TPI) polyclonal

antibodies with TPI from difl‘erent protein concentrates determined by

sandwich ELISA. Bovine TPI = TPI purified from bovine

semimembranosus muscle, WPI = whey protein isolate, AMP600N =

hydrolyzed protein fi'om meat and plasma, AMP800 = whey protein

concentrate, bovine raw extract = ground bovine semimembranosus

muscle extracted with phosphate buffer saline (1:3). The standard deviation values
that were less than 0.02 were not plotted on the figure 102

Representative sodium dodecyl sulfate-polyacrylamide gel

xviii

5.7

electrophoretogram of muscle extracts of bovine semimembranosus meat
heated to different end-point temperatures. Proteins were stained with
Coomassie Blue. (Lane 1) Molecular weight standards; (Lane 2)

Unheated bovine muscle extracts; (Lane 3) 48.9 °C; (Lane 4) 54.4 0C;
(Lane 5) 60.0 °C; (Lane 6) 65.6 °C; (Lane 7) 71.1 °C; (Lane 8) 76.7 °C;
(Lane 9) Purified bovine TPI ......................................................................... 106
Western blot of bovine muscle extract with anti-triose phosphate isomerase
polyclonal antibodies electrophoretically transferred from a

sodium dodecyl sulfate-polyacrylamide gel to a nitrocellulose membrane.
Bovine semimembranosus muscle was heated to different end-point
temperatures. (lane 1) Bovine TPI; (lane 2) Unheated bovine muscle
extracts; (lane 3) 48.9 °C; (Lane 4) 54.4 °C; (Lane 5) 60.0 0C; (Lane 6)

65.6 °C; (Lane 7) 71.1 0C; (Lane 8) 76.7 °C ................................................... 107

xix

CHAPTER 1

INTRODUCTION

Since the January 1993 E. coli 01572H7 outbreak of foodbome disease caused by
consumption of undercooked ground beef patties at a fast food restaurant chain in several
Western states, the general public has become highly concerned about the safety of meat
and poultry products (U SDA-FSIS, 1996a). Salmonella, hemorrhagic Escherichia coli,
Campylobacter and Staphylococcus are the major pathogenic bacteria found to cause
foodbome disease outbreaks associated with domestic and imported precooked meats. A
recent study estimated that foodbome outbreaks result in 5.5 to 6.2 million cases costing
5.8 to 8.6 billion US. dollars each year (Todd, 1994). Enteric pathogens from animal
food products cause from 6.5 to 81 million outbreaks of illness each year in the US. as
reported by the Centers for Disease Control and Prevention (CDC) (Bean and Griffin,
1990). Salmonella is one of the most frequent causes of foodbome disease in the U. S.
Although 45,000 cases of Salmonella outbreaks are reported every year in the United
States, foodbome cases of salmonellosis are estimated to range from 790,000 to 3.69
million annually, with a median number of 1.92 million cases (Todd, 1994). Most cases
of foodbome disease attributed to E. coli 01572H7 were caused by consumption of
undercooked and contaminated hamburger. About 10,000 to 20,000 cases of E. coli

0157:H7 induced illness and 200 to 500 deaths occur in the U. S. every year (CDC,

1996). In fact, foodbome outbreaks associated with meat products continue to occur
frequently in the U. 8., thus effective methods to prevent and control foodbome
pathogens are needed.

On October 1, 1996, a collaborative project to develop food safety strategies to
reduce foodbome illness was initiated (U SDA-F SIS, 1996a). The purpose of this project
was to improve and monitor the implementation of food safety programs to decrease the
number of pathogenic microorganisms, especially Salmonella and E. coli 0157:H7, in
meat, poultry, seafood, dairy, fruit and vegetable products. Several agencies, including
the US. Department of Agriculture-Food Safety and Inspection Service CUSDA-FSIS),
the Food and Drug Administration (FDA), CDC and public health departments, state
health departments and local investigators at five locations in U. S., will collaborate to
develop better methods to track the incidence of foodbome illness.

Adequate cooking is the simplest means of eliminating pathogenic bacteria from
meat and poultry products. The USDA-FSIS (United States Department of Agriculture-
Food Safety Inspection Services) has established specific cooking requirements for
different types of precooked meat and poultry products. The purpose of these
requirements is to ensure the destruction of harmful microorganisms and viruses that
could cause foodbome illness in humans and livestock (Townsend and Blankenship,

1989). The USDA-F SIS (1995) has required that cooked beef and roast corned beef be

heated to 62.8 °C (145 oF) instantaneously or to one of sixteen different time-temperature

combinations. These processing treatments were designed to ensure a 7-D destruction of

b)

Salmonella (a 7-D reduction of Salmonella is to reduce the number of Salmonella from 1
x 107 to 1 in 1 gram of meat) (Goodfellow and Brown, 1978).

On May 2 1996, USDA-F SIS proposed several rules to amend the current federal
meat and poultry inspection regulations to establish “safety margins” for thermally
processed meat and poultry products (U SDA-F SIS, 1996b). The new regulations for
meat and poultry will be applied to the following products: roast beef, cooked beef and
cooked corned beef, uncured meat patties (fully cooked, partially cooked and char-
marked patties), and some poultry products (fully and partially cooked products). The
purpose of the new proposed performance standards for meat and poultry is to ensure the
safety of meat products by eliminating pathogenic microorganisms from the products.
Three major steps to achieve this objective were lethality, stabilization, and handling.
Salmonella is a common pathogen in roast and cooked beef products and is more
thermally resistant than E. coli 015 7:H7. Thus, instead of specific approved processing
schedules, a 7-D reduction in Salmonella was proposed by FSIS as the lethality
performance standard for processing of roast beef, cooked beef and cooked corned beef
products. A 5-D reduction in Salmonella (a 5-D reduction of Salmonella is to reduce the
number of Salmonella from 1 x 105 to 1 in 1 gram of meat) was proposed as the lethality
performance standard for cooked, uncured meat patties. Also, the USDA-FSIS
announced a new series of time-temperature processing requirements based on a 7-D

reduction in Salmonella for poultry, instead of the current single temperature requirement

for uncured (71.1 °C, no holding time) and cured (68.3 °C, no holding time) products.

Meanwhile, F818 is evaluating these lethality standards using different processing

conditions to accommodate more flexibility and to provide more realistic processing
schedule combinations. This proposed regulation Should allow for the use of more
sophisticated and flexible thermal processing treatments and provide a higher standard of
safety and wholesomeness in meat and poultry products.

A time-temperature integrator (TTI) is needed to verify that meat products receive
adequate thermal treatment to eliminate pathogenic microorganisms. A TTI is “a
measuring device that shows a time-temperature dependent, easily, accurately and
precisely measurable irreversible change that mimics the changes of a target attribute
undergoing the same variable temperature exposure” (Hendrickx et al., 1995). The 2
value or activation energy of a TTI in the revelant range of a time-temperature processing
schedule should have a z-value or activation energy (kinetics of rate constant) similar to
that of the target index in the food system. A TTI can be prepared from enzymes,
proteins, microorganisms or chemicals. Time-temperature integrators can exist in either
extrinsic or intrinsic forms. Extrinsic TTIs are added to the food system and are retrieved
back from the food after thermal processing is completed. Extrinsic TTIs can be
encapsulated to prevent changes in inactivation kinetics of the indicator due to reactions
with the food material. However, intrinsic TTIs are prepared from compounds in the
target food. If the TTI is an endogenous component of the food system, extraction or
recovery procedures are necessary to analyze its concentration after processing.

An endogenous indicator protein used as a TTI to verify roast beef processing
should meet two criteria. First, its concentration or activity should differ between optimal

and sub-optimal processing treatments. Second, the indicator should have the same

concentration or activity in all optimal processes. For these criteria to be met, the marker
protein must have a 2 value similar to that of the pathogen used by the USDA-FSIS to
establish roast beef processing schedules.

Currently, the F SIS is using the "Bovine Catalase Test" developed by Eye (1982)
for the detection of underprocessing of rare beef and cooked beef. A protein

"Coagulation Test" is used for monitoring the maximum internal temperature of both beef

and pork products heat processed to temperatures lower than 65 0C (149 °F) (U SDA-
FSIS, 1986a). These methods for verification of heating endpoint temperatures in
precooked meat products have been shown to be subj ective and inaccurate for predicting
the actual heating endpoint temperature achieved during processing (Townsend and
Blankenship, 1989). The greatest disadvantage of these tests is that none uses an
indicator with a 2 value similar to the 2 value of Salmonella used by the USDA to define
roast beef processing schedules.

The overall goal of this study was to examine the thermal inactivation kinetics of
selected endogenous enzymes present in bovine semimembranosus muscle and then to
identify a candidate marker protein as a TN to monitor thermal processing adequacy of
roast beef. A marker protein, triose phosphate isomerase (TPI), was identified as an
enzymatic TTI in this study since it had a 2 value similar to that of Salmonella used to
establish the USDA roast beef processing schedules. Therefore, TPI was purified and
used to produce anti-TPI polyclonal antibodies. The final objective of this Study was to
establish an endogenous enzymatic TTI using an immunochemical assay (enzyme -

linked immunosorbent assay; ELISA) to verify adherence to safe processing schedules.

Theoretically, the ELISA should provide an accurate, rapid, inexpensive and convenient
method for thermal monitoring by regulatory agencies and processors for determining the
adequacy of thermal processing of precooked roast beef products.

Four specific objectives of this study are listed below:

Study I - To compare the thermal inactivation kinetics of acid phosphatase and
peroxidase to the D and 2 values of E. coli OlS7:H7 and Salmonella senflenberg in
ground beef at four different temperatures.

Study 11 - To screen several endogenous bovine enzymes as time-temperature
integrators for precooked roast beef using optimal and sub-optimal time-temperature
processing conditions.

Study III - To purify the marker protein, develop polyclonal antibodies and devise
an ELISA to quantify the marker protein in extracts of roast beef.

Study IV - To verify the ability of the selected marker protein by ELISA and
enzyme assay to determine adequacy of thermal processing of beef cooked in model
systems and commercial meat processing plants.

Four experiments were designed to evaluate the objectives of this study:

Experiment I - The first experiment was designed to compare the thermal
inactivation kinetics of acid phosphatase and peroxidase to the D and 2 values of E. coli
OlS7:H7 and Salmonella senfienberg in ground beef at four different temperatures. This
research was done in COOperation with other researchers who also evaluated the thermal
inactivation kinetics of triose phosphate isomerase, glyceraldehyde-3-phosphate

dehydrogenase, phosphoglycerate mutase, E. coli 01 57:H7 and Salmonella senfienberg.

Enzymes with a 2 value Similar to that of Salmonella could be used as TTIs to verify that
roast beef was adequately processed.

Experiment 11 - The second experiment was designed to screen several
endogenous bovine enzymes as TTI indicators for precooked roast beef using different
optimal and sub-optimal time—temperature processing conditions. Beef top round roasts
were processed using two or three time temperatures combinations in a smokehouse used
as a heat processing oven. Each combination included one or two sub-optimal processes
and one optimally cooked treatment. Low salt soluble proteins were extracted and
enzyme activities or concentrations of peroxidase, TPI, acid phosphatase, and lactate
dehydrogenase were determined.

Experiment III - The third experiment was to purify the marker protein, TPI, and
to develop polyclonal antibodies against the marker protein. Polyclonal antibodies were
made against purified bovine TPI using rabbits. A TPI sandwich ELISA was developed
and optimized to quantify the marker protein in extracts of roast beef in a model system.

Experiment IV - The fourth experiment was to verify the ability of TPI, measured
by ELISA and enzyme assay, to determine processing adequacy of beef cooked in a
commercial meat processing plant. This final experiment was designed to determine if
the TPI sandwich ELISA could be used to differentiate between optimal and sub-optimal
time-temperature combinations of commercially processed roast beef to confum its

applicability.

CHAPTER 2

LITERATURE REVIEW

2.1 USDA-FSIS THERMAL REGULATIONS FOR MEAT PRODUCTS

To ensure the destruction of pathogenic bacteria and viruses, Title 9 of the Code of
Federal Regulations lists required thermal treatments for a variety of precooked meat and
poultry products. Regulations also exist for imported precooked products to prevent
foodbome diseases. These regulations were implemented to ensure the destruction of
harmful microorganisms (e. g., Salmonella, hemorrhagic Escherichia coli,
Campylobacter, Clostridium, and Staphylococcus) and viruses (e.g., Velogenic-
Viscerotropic Newcastle Disease Virus) that cause illness in humans and animals
(Townsend and Blankenship, 1989).

The United States Department of Agriculture - Food Safety and Inspection
Service (U SDA-FSIS) (1995) requires that cooked beef and rare roast beef be heated in
accordance with one of 16 time-temperature combinations (Table 2.1). For instance, beef
can be beat processed to 62.8 °C (145 °F) with no holding time or processed to 54.4 0C
(130 °F) with 121 min holding time. Cured/smoked and ready-to-eat pork products must
be heated to at least 58.3 0C (137 °F) or frozen to destroy T richinae. For poultry
products, minimum internal temperatures of71.1 0C (160 °F) and 68.3 0C (155 °F) are

required for uncured and cured poultry products, respectively.

Table 2.] Minimum USDA thermal processing schedules for cooked beef and roast beef

 

 

Temperature (°C (°F)) Time (min)
54.4 (130) 121
55.0 (131) 97
55.6 (132) 77
56.1 (133) 62
56.7 (134) 47
57.2 (135) 37
57.8 (136) 32
58.4 (137) 24
59.5 (139) 15
60.0 (140) 12
60.6 (141) 10
61.1 (142) 8
61.7 (143) 6
62.2 (144) 5
62.8 (145) o

 

Code of Federal Regulations, Title 9 (USDA-F818, 1995)

10

2.2 THE USDA-FSIS PROPOSED REGULATIONS TO AMEND THE
CURRENT MEAT AND POULTRY INSPECTION REGULATIONS

On May 2, 1996, USDA-FSIS proposed several rules to amend the current federal
meat and poultry inspection regulations to establish “safety margins” for thermally
processed meat and poultry products (U SDA-FSIS, 1996b). The new regulations for
meat and poultry would be applied to the following products: roast beef, cooked beef and
cooked corned beef, uncured meat patties (fully cooked, partially cooked and char-
marked patties), and some poultry (fully and partially cooked products). The purpose of
the new proposed performance standards for meat and poultry products are to ensure the
safety of meat products by eliminating pathogenic microorganisms from the products.
Three major steps to achieve this objective were lethality, stabilization, and handling.

A 7-D reduction in Salmonella was proposed by FSIS as the lethality performance
standard for roast beef, cooked beef and cooked corned beef products (U SDA-F SIS,
1996b). There are three reasons that Salmonella was chosen as the indicator strain by
F SIS. First, Salmonella is a major and commonly found pathogen in roast and cooked
beef products. Second, Salmonella is more thermally resistant than E. coli 01 57:H7.
Third, even though Listeria monocytogenes is more heat resistant than Salmonella, it is
mostly found in post-process contaminated meat and at levels much lower than typical for
Salmonella in meat products. In general, a processing scheduled based on 7-D reduction
in Salmonella for roast and cooked beef was designed to produce pathogen-flee and safe
cooked beef products, however, it may cause over-processed beef products. Since

Salmonella is usually found in raw beef at concentration below 103 to 104

ll

microorganisms/gram of meat, a 3-D or 4-D reduction in Salmonella should be sufficient
to offer safe cooked beef products. At this time, FSIS is still evaluating this lethality
standard by comparing different commercially applicable processing conditions.

In addition, a 7-D reduction in Salmonella was proposed for cooked poultry
products. This is the first time USDA-F SIS has pr0posed a series of time-temperature
schedules for poultry instead of a single temperature requirement for uncured (71.1 °C, no
holding time) and cured poultry (68.3 °C, no holding time) products.

A 5-D reduction in Salmonella was proposed by FSIS as the lethality performance
standard for cooked, uncured meat patties (U SDA-F SIS, 1996b). A processing schedule
sufficient to cause a 5-D reduction should provide for pathogen-free products and avoid

dry, burned and low quality products caused by a more severe thermal process.

2.3 CURRENT USDA METHODS FOR ENDPOINT TEMPERATURE (EPT)
DETERMINATION

Current procedures used by the FSIS for monitoring the adequacy of heat
treatment of meat and poultry include three techniques. They are the Bovine Catalase
Activity Test for cooked and roast beef, the Protein Coagulation Test for beef and pork
products, and the residual Acid Phosphatase Activity Test for canned hams, picnics and
luncheon meats. These USDA approved methods for EPT determination are inaccurate,
subjective, impractical, and time-consuming. Since the number of foodbome illness
outbreaks are increasing yearly, these EPT detennination methods are crucial for the

health of the general public. However, the current methods have shortcomings and

12

effective assays are urgently needed. Many alternative methods have been investigated to

determine EPT in various meat products.

2.3.1 Bovine Catalase Activity Test

The Bovine Catalase Activity Test was developed approximately 11-15 years ago
(Townsend and Blankenship, 1989) and is used for cooked and roast beef. The detection
of residual endogenous catalase activity is commonly utilized at heating EPTs slightly
above 60 °C (140 °F). Foam is produced as oxygen and is liberated from a reaction of
hydrogen peroxide in the presence of sodium lauryl sulfate (shampoo). “Strong, medium,
weak and no activity” must be interpreted from the quantity of foam produced (U SDA-
FSIS, 1989). Research has indicated the highly subjective nature of this test and endpoint

temperatures interpretations varied by operator (Stalder et al., 1991).

2.3.2 Protein "Coagulation Test"

A protein "Coagulation Test" is used to monitor the maximum internal
temperature of both beef and pork products heat processed to temperatures lower than 65
°C (149 °F) (U SDA-F SIS, 1986a). The Protein Coagulation Test was developed about
35-40 years ago (Townsend and Blankenship, 1989).

The "Protein Coagulation Test" is based on measurable loss in protein solubility
as product temperature is increased. The test involves extracting soluble muscle proteins
and observing the temperature at which the first signs of cloudiness or turbidity (54-57

°C; 129-135 °F) appear when the filtrate is heated. This is considered to be the maximum

a

1.)

internal cooking temperature of the product. For products heat processed to 63-71 °C
(145-160 0F), the temperature at which the filtrate becomes cloudy may differ by 8-10 oC
(46.4-50 °F) from the known internal temperature of the product (U SDA—F SIS, 1986a).
This method is empirical, subjective, time—consuming and difficult to perform outside of

an established laboratory (Townsend etal., 1984).

2.3.3 Acid Phosphatase Activity Test

The Acid Phosphatase Activity Test for determining the EPT was also developed
35-40 years ago (Townsend and Blankenship, 1989). The residual "Acid Phosphatase
Activity Test" (U SDA-F SIS, 1986b) is used to determine heat treatment of canned hams,
picnic hams and luncheon meat.

Enzyme activity is determined on water extracts of ground meat samples utilizing
disodium phenyl phosphate as substrate. The enzymatic reaction produces phenol- and
mono-hydrogen phenyl phosphate. Phenol is subsequently reacted under alkaline
conditions with 2, 6-dibromoquinonechlorimide to produce a blue chromophore which
absorbs at 610 nm. The residual activity of the enzyme after cooking is expressed as p
mole of phenol formed per 1000 g meat after reacting with the substrate, disodium
phenylphosphate, for 60 min at 37 0C (99 °F), pH 6.5. The residual activity of acid
phosphatase in hams does not depend directly on temperature alone, but on
time/temperature history. Results are poorly correlated with actual EPT (Cohen, 1969;

Kormendy et al., 1987; Townsend and Blankenship, 1989).

14

2.4 ALTERNATIVE METHODS FOR EPT DETERMINATION

Several alternative testing procedures for meat and poultry have been developed.
They include methods based on solubility of sarcoplasmic proteins (Davey and Gilbert,
1974), dominant spectral wavelengths of beef juice to predict the maximum internal
temperatures in cooked beef (Nusimovich et al., 1979), intrinsic protein fluorescence to
examine the structural changes of muscle protein during heating (Oreshkin et al., 1968),
near infrared reflectance to measure protein denaturations (Osborne and F earn, 1986),
differential scanning calorimetry to monitor protein denaturation temperature (Quinn et
al., 1980; Wright and Wilding, 1984), sodium dodecyl sulfate (SDS) gel electrophoresis
of proteins (Caldironi and Bazan, 1980), residual enzyme activity (Townsend and
Blankenship, 1987a, 1987b; Collins et al., 19913, b; Stalder et al., 1991), and
immunoassays (Wang et al., 1992; Abouzied et al. 1993; Wang et al., 1993, 1994, 1995).
Based on a literature review, enzyme assays and immunoassays offer powerful techniques

for selecting protein markers and verifying EPT in processed products of various meat

species.

2.5 ENZYMATIC METHODS FOR EPT DETERMINATIONS IN MEAT
PRODUCTS
Several candidate enzymes have been screened and identified as potential protein

markers in different meat products.

15

2.5.1 Bovine Meat

Lactate dehydrogenase (LDH) has been evaluated as a potential protein marker in
both bovine and porcine muscles (Collins et al., 1991a, b; Stalder et al., 1991). The
influence of pH, salt, phosphate, cooking temperature, muscle type, carcass gender and
maturity on LDH activity in water extracts of bovine muscle have been examined (Stalder
et al., 1991). Cooking decreased LDH activity from greater than 1000 U/g muscle in raw
meat to almost undetectable levels at 66 0C (151 °F), regardless of pH. Extracts heated
at pH 5.6 showed sharper decreases in activity with increasing temperature than samples
heated at pH 6.4. When different levels of salt (0 to 4.5% NaCl) and phosphate (0 to
0.3% sodium triophosphate) were added, LDH activity decreased between 5 7 and 63 0C
(135-145 0F) at pH 5.6. Gender of carcasses did not affect LDH activity.

Further evaluation of LDH as a heating EPT indicator in whole cooked roast beef
products was performed by Stalder et al. (1991). LDH activity decreased with increased
cooking temperature in roasts prepared with and without brine. Colonnetric and
fluorescent assays were successfully used on juice squeezed from intact roasts to
determine LDH activity. LDH was inactivated at about 62 0C (144 °F). No relationship
of LDH complianeed to time-temperature thermal processing schedules was investigated.

Townsend and Davis (1992) concluded that transaminase enzymes (glutamate-
oxaloacetate transaminase (GOT), glutamate-pyruvate transaminase) retained
considerable activity at 71.1 0C (160 °F), therefore, GOT could possibly be used for
determining the adequacy of thermal treatment of imported cooked beef which must be

heat processed to 79.4 0C (175 °F). These enzymes would not be good indicators for

l6

ground beef which requires one of seven time-temperature combinations, ranging from

66.1 °C (151 °F) for 41 sec to 69.4 0C (157 0F) for 10 sec (USDA-FSIS, 1995).

2.5.2 Porcine Meat

Pyruvate kinase (PYK) has shown potential as EPT indicators in a model cured
pork system and a commercial canned cured pork product (Davis et al., 1988). In the
model system, high PYK activity was observed at 67.7 0C and activities were decreased
when meat heated to 68.3 or 68.9 °C. When heated to 69.5 or 70.0 °C, PYK activity was
not detectable fi'om the cured pork model system. In a commercial canned cured pork
study, high PYK activity was found at 62.9 °C internal temperature. Gradually decreased
in PYK activities were found when products heated to 65.6 and 68.6 °C. No PYK activity

was observed when canned products heated to 69.9 °C.

2.5.3 Poultry Meat

Bogin et a1. (1992) examined 12 enzymes from turkey breast meat for use as EPT
indicators. The activities of the enzymes were studied after heating 25 - 50 g meat in
glass beakers to different temperatures (50 °C, 60 0C, or 70 oC) and for various times
(15, 30, 45, 90 min) to identify possible EPT indicators. Then, 3 g of meat was removed,
homogenized with 30 m1 Tris-HCl buffer (0.02 M; pH 7.4), and centrifuged at 35,000 x g
for 40 min. The supernatant was collected to assess enzyme activity and concentration of

soluble protein. Asparate-oxoglutarate aminotransferase was found to be most suitable

for verification of an EPT of 70 °C (158 °F). Creatine kinase (CK), malic

l7

dehydrogenase, lactate dehydrogenase, and isocitric dehydrogenase were also found to
decrease proportionally with heating temperature and time and were suggested potential
markers for inactivation of Velogenic-Viscerotropic Newcastle Virus in turkey meat.

Hsu (1993) screened 26 endogenous enzymes extracted from turkey pectoralis
major (breast) and sartorius (thigh) muscles. Lactate dehydrogenase and malic
dehydrogenase from pectoralis major each showed consistent decreases in activity
between 69 0C (156 0F) and 73 0C (163 °F) in three treatments (water treatment: ground
muscle blended with 15% double distilled water; brine treatment: ground muscle blended
with 15% NaCl brine by weight to result in a final concentration of 1.5% NaCl, 0.5%
sodium tripolyphosphate (STPP) and 12.5% water in the tissues; and curing brine
treatment: ground muscle mixed with 15% NaCl curing brine by weight to result in a
final concentration of 2.5% NaCl, 0.5% STPP, 156 ppm sodium nitrite, 550 ppm sodium
erythorbate and 12.5% water in the tissues), at three heating rates (0.25, 0.5, and 1.0
oC/min) in fresh and frozen tissues. Thus, these enzymes could potentially serve as EPT
indicators in poultry for USDA-FSIS minimum temperatures of 68.3 0C (155 °F) (cured)
or 71.1 0C (160 0F) (brined).

Davis and Townsend (1994) examined acid phosphatase activity in both non-
frozen and fiozen ground broiler, turkey breast and turkey dark meat. Meat was tightly
packed in glass tubes (25 x 150 mm) and heated to five different temperatures: 62.8, 65.6,
68.3, 71.1 and 73.9 0C in a water bath system. The authors suggested that residual acid
phosphatase activity may be product dependent; therefore, more research will be

necessary to establish maximum residual concentrations for various products.

18

2.5.4 Comparison of EPT Indicators in Bovine, Porcine and Poultry Meats

Smith (1991) investigated 26 endogenous enzymes as EPT indicators
in bovine, porcine, and poultry meats. In bovine muscle, CK, enolase, glyceraldehyde-3-
phosphate dehydrogenase, malate dehydrogenase, phosphoglucoisomerase,
phosphoglucomutase, and pyruvate kinase underwent rapid denaturation between 60 0C
(140 0F) to 72 °C (162 0F). In porcine muscle, enolase, malate dehydrogenase, fructose-
6-phosphate kinase, pyruvate kinase, CK, aldolase, and TPI were rapidly inactivated near
the federally regulated EPT of 68.8 0C (156 °F) for imported pork and pork products. In
chicken muscle, aldolase, CK, enolase, fructose-6-phosphate kinase, glutamate-pyruvate
transaminase, malate dehydrogenase, and phophoglucoisomerase from chicken muscles
showed significant decreases in activity or no activity at an EPT range of 66-68 0C (151 -

154 °F) or 72-74 0C ( 162-165 0F).

2.6 IMMUNOASSAYS FOR EPT DETERMINATIONS IN MEAT PRODUCTS
Immunoassays provide a sophisticated approach to assure the safety and quality of
foods (Kang’ethe, 1990; Fukal, 1991; Samarajeewa et al., 1991; Morgan et al., 1992).
The use of immunological assays for detection of hormones, pesticides, antibiotics,
mycotoxins, proteins and enzymes have been well documented (Pestka, 1988).
Immunoassays have been used for determining EPT of several meat products, including
ground beef (eg., hamburger patty) and poultry products (eg., turkey roll and turkey ham)
(Wang et al., 1992; Abouized et al. 1993; Wang et al., 1993, 1994, 1995; Orta—Ramirez et

al., 1996). Enzyme-linked immunosorbent assays (ELISA) are the most widely used

l9

immunological assays in the food and agricultural sciences. ELISAS are Simple,
sensitive, highly Specific, more accurate and less time-consuming than many other
detection methods. Immunoassays may be more sensitive and less expensive than

enzymatic methods.

2.6.1 Bovine Meat

Wang et al. (1995) developed an LDH sandwich ELISA to determine the EPT of
ground beef. Ground beef was packed in 10 x 75 mm test tubes and heated from 62 to
74 0C at 2 oC increments in a water bath. Polyclonal antibodies against LDH were used
for capture and biotinylated polyclonal antibodies were used for detection in the sandwich
ELISA. Less than 4 ug LDH/g meat was detected at 69 oC and decreases (p < 0.05) in
LDH were found when ground beef was cooked between 66 and 74 0C at 2 0C
increments. Commercially cooked processed beef patties heated fi'om 68.3 to 71.1 0C
contained 2.78 pg LDH/g meat. The authors suggested that the LDH sandwich ELISA
was able to detect the EPT of commercially cooked beef patties; however, thermal
processing variations and formulations would affect the residual amount of LDH in
processed patties.

A sandwich ELISA, using both anti-LDH monoclonal and polyclonal antibodies,
was developed to verify EPT of ground beef patties (Orta-Ramirez et al., 1996). The
ELISA could detect differences (p < 0.05) in LDH concentration between patties cooked
to EPT of 62.8 0C (145 °F) and 65.6 0C (150 °F) or 62.8 °C (145 °F) and uncooked

‘ patties. No difference in LDH concentration was found among patties cooked to internal

20

temperatures of 65.6 0C (150 0F), 68.3 0C (155 °F) and 71.1 0C (160 0F). At different
fat contents (10.7, 13.6 and 19.0%), no differences were found among patties at EPT of
68.3 0C (155 OF). Orta-Ramirez et al. (1996) also suggested that the LDH sandwich
ELISA detected about 3 pg LDH/g meat in processed ground beef cooked to 70 0C in

both a model study and commercially cooked beef patties.

2.6.2 Poultry Meat

Wang et al. (1992) developed an indirect competitive ELISA to determine EPT of
uncured turkey breast rolls using LDH as the marker protein. Turkey rolls were processed
to internal temperatures of 68.3 0C (155 °F), 69.7 0C (157 °F), 70.9 0C (160 °F) and
72.1 °C ( 162 °F). LDH was not observed in extracts separated by native polyacrylamide
gel electrOphoresis (PAGE) at 70.9 0C (160 °F) and was chosen as the EPT marker
protein for immunoassay development. LDH was identified on the basis of molecular
weight (35 kd) using SDS-PAGE and LDH-specific stain on native PAGE. Polyclonal
antibodies (PAbS) were made against purified turkey muscle LDH and commercially
available chicken muscle LDH. The LDH content in turkey breast roll extracts as
determined by sandwich ELISA decreased as the cooking temperature was increased.

Abouzied et al. (1993) developed a sandwich ELISA to quantify LDH in uncured
poultry products. Four monoclonal antibodies (MAbs) against chicken muscle LDH were
produced. In this LDH sandwich ELISA, monoclonal antibodies were used for capture
and polyclonal antibodies were used as the detecting antibodies. The ELISA could detect

,1 ng LDH/mL in turkey or chicken meat extracts. The LDH sandwich ELISA was tested

on extracts of turkey rolls processed to 68.3, 69.7, 70.9 and 72.1 0C. LDH concentration
was 10 times lower in rolls processed to 70.9 °C as compared to those cooked to 69.7 °C.
The antibodies cross reacted with LDH from chicken and turkey, but not beef or pork.
The authors suggested that the sandwich ELISA should be able to verify the EPT of
turkey breast rolls previously cooked to the required USDA minimum internal
temperature of 71.1 °C.

Wang et al. (1993) further examined the effects of formulation, storage and
processing conditions on LDH concentration of turkey breast rolls as measured by
sandwich ELISA. LDH concentration differed at processing temperatures of 70.0 0C
(1 5 8 °F) and 71.1 0C (160 °F), however extractable protein and LDH activity did not
differ at these temperatures. They concluded that salt concentration, cooking schedule
and product casing diameter did not influence LDH concentration. Frozen storage
decreased LDH content of uncooked rolls. However, the LDH ELISA could not
differentiate EPT of turkey thigh rolls processed between 68.9 0C (156.02 °F) and 71.1
0C (159.98 °F) due to the presence of heat stable LDH isozymes (Wang et al., 1994).
Desrocher (1994) also developed ELISAS to verify EPT of turkey hams. The LDH
sandwich and immunoglobulin G indirect competitive ELISA both showed the potential

to accurately determine EPT of turkey ham.

2.7 THERMAL DEATH TIME (TDT) STUDIES
Thermal death time (TDT) studies are used to determine the thermal resistance of

microorganisms or enzymes in a food system. There are many factors that affect thermal

22

death time, such as water activity, pH, and food components and contents (protein,
carbohydrate, lipid, salt). D value represents the time (min) required to destroy or
inactivate 90% of the organisms or enzyme activity at one temperature. In general,
decreasing water activity increased D value. Decreasing the pH of a food usually reduces
heat resistance and decreases D value. Because of the protective effect provided by
physical aggregation of carbohydrates, proteins, lipids and salt (up to 2-4 %), the higher

the concentration of these ingredients, the higher the D value (Jay, 1992).

2.7.1 Definitions

Mathematical methods can be used to determine the effect of thermal processing
on food components and microorganisms. The thermal destruction of microorganisms,
enzymes, nutrients and quality factors, such as texture, color and flavor, in a food system,
usually follow first order reaction kinetics (Lund, 1975).

The decimal reduction time or D value represents the time (min) needed to
destroy or inactivate 90% of the organisms or food component in a system at a certain
temperature. The 12D concept is applied in the canning industry to reduce the population
of the most heat-resistant spore-forming pathogenic microorganism, Clostridium
botulinum, by 1 x 1012 spores. If 1 x 103 Clostridium botulinum spores (which is the
estimated maximum count) were in a low-acid food; therefore, after a 12D process, there

should be less than one spore in l x 109 cans.

23

The D value is calculated from the equation:

T

 

D =
log a - log b
where
D = decimal reduction time or death rate (min)
T = heating time (min)

a = initial number of microorganisms or initial activity of an enzyme

b = final number of microorganisms or residual enzyme activity

The thermal death time curve can be graphed on a semi-log scale. The X-axis is
the temperature which is the linear scale and the Y-axis is the D value on a logarithmic
scale. The best straight line through these points is the thermal death time (TDT) curve.
The 2 value is calculated from the reciprocal of the slope of the TDT curve and is defined
as the temperature change needed to transverse one logarithmic cycle of the TDT curve.

The F value is the time (min) required to destroy a certain number of
microorganisms at a specific temperature. This value can be used to compare different
thermal processes. The F0 value is the time required to destroy a certain number of
microbial spores at 250 0F when the 2 value is 18 0F.

D and 2 values can be calculated based on microbial destruction or enzyme
inactivation. For instance, catalase and peroxidase were selected as marker proteins for
the blanching process for vegetables (Lund, 1975). Alkaline phosphatase was chosen as
the indicator protein for milk pasteurization, since Mycobacterium tuberculosis, the most

' thermal-stable pathogen in milk, and alkaline phosphatase have the same rate of thermal

24

inactivation (Kay and Graham, 1933). Clostridium botulinum is used as an indicator of
lethality in thermal process calculations system, because it is the most heat-resistant
spore-forming pathogenic microorganism found in canned foods. These indicator
enzymes and microorganisms can be categorized as endogenous time temperature
integrators (TITS) with thermal inactivation constants (2 value or activation energy)

similar to the target attributes.

2.7.2 TDT Studies in Various Meat Systems
2.7.2.1 Ground Beef

Goodfellow and Brown (1978) initiated a study to determine the D values of
Salmonella serotypes, including Salmonella typhimurium strain TMI (as a reference
strain, less heat resistant Strain), Salmonella newport, Salmonella agona, Salmonella
bovis-morbificians and Salmonella muenchen strains (the last 4 strains were isolated fiom
food poisoning outbreaks related to meat sources) in ground beef. A mixed strain
inoculum of six Salmonella serotypes was added to ground beef (fat content not stated).
Plate Count Agar (PCA) and PCA with xylose lysine deoxycholate agar overlay were
used to recover Salmonella spp. D values for Salmonella in ground beef were determined
and Salmonella survival was also evaluated in both a water cooking and dry roasting beef
system. D and 2 values are shown in Table 2.2. This study was the first to determine a
series of industrial processing schedules which could reduce Salmonella by 7 D in ground

beef.

25

Ng et al. (1969) heated individual Salmonella strains in broth (Trypticase Soy
Broth with 2% yeast extract, TSB-YE). The D values at 57 °C of Salmonella
typhimurium Tm-l was 1.2 min, Salmonella blockley 2004 was 5.8 min, and Salmonella
senftenberg 775W was 31 min. The authors suggested that Salmonella senflenberg 775W
strain was more heat resistant than other Salmonella strains. Since Goodfellow and
Brown (1978) used mixed serotype strains of Salmonella, the D values at 57.2 °C were
3.8 - 4.2 min, indicating that mixed strains were less heat resistant than Salmonella

senftenberg 775W but more heat resistant than Salmonella typhimurium Tm-l.

Table 2.2 Thermal death time curve data and 2 values for Salmonella serotypes in a

 

 

 

ground beef system

Fat content D value (min) 2 value (°C)
51.6 0C 57.2 0C 62.7 0C

Not specified 61-62 3.8-4.2 0.6-0.7 5.56

 

From Goodfellow and Brown (1978).

Goodfellow and Brown (1978) also recommended time-temperature schedules
ranging from 53.3 to 62.2 °C to reduce Salmonella by 7-D in roast beef (Table 2.3). In

' addition, other cooking quality factors were also affected by temperature, such as degree

26

Table 2.3 Time-temperature schedules for a 7-D reduction of Salmonella in cooked beef

 

 

Internal temperature (°C (° F )) Processing time (min)
53.3 (128) 195
53.9 (129) 153
54.4 (130) 121
55.0 (131) 97
55.6 (132) 77
56.1 (133) 62
56.7 (134) 47
57.2 (135) 37
57.8 (136) 32
58.3 (137) 24
58.9 (138) 19
59.4 (139) 15
60.0 (140) 12
60.6 (141) 10
61.1 (142) 8
61.7 (143) 6
62.2 (144) 5

 

i From Goodfellow and Brown (1978).

27

of doneness and color of beef roasts. The USDA established thermal processing
schedules for cooked beef and roast beef based on this study, and food processors have
implemented these schedules in their roast beef manufacturing.

Fain et al. (1991) determined the D and 2 values for Listeria monocytogenes strain
Scott A over the same range of temperatures (from 53.3 to 62.2 °C) as Goodfellow and
Brown (1978) in lean (2% fat) and fatty (30.5% fat) ground beef (Table 2.4). Two
recovery mediums were used; Columbia CNA (Columbia Colistin Nalidixic Acid Agar)
agar containing sodium pyruvate with horse blood overlay (CBNA) and Listeria Plating
Medium (LPM). 2 Values were higher in higher fat beef than lean meat. Also, Listeria
monocytogenes had a higher 2 value in LPM recovery medium than in CBNA recovery
medium, suggesting that different recovery techniques for Listeria monocytogenes would
affect D and 2 values in ground beef. Since the 2 value of Listeria monocytogenes (11.4
°F) was higher than that of Salmonella (5 .56 0C) (Goodfellow and Brown, 1978), the

authors suggested that L. monocytogenes was a potential food safety problem.

28

Table 2.4 Thermal death time curve data and 2 values for L. monocytogenes Scott A in

lean (2% fat) and fatty (30.5% fat) ground beef systems

 

Fat content Culture medium D value (min) 2 value (0C)

 

51.7 0C 57.2 0C 62.8 0C

(125 0F) (135 0F) (145 0F)

 

2.0% CBNA ' 81.3 2.6 0.6 5.2
30.5% CBNA 71.1 5.8 1.2 6.3
2.0% LPM 2 56.1 2.4 0.5 5.4
30.5% LPM 34.5 4.6 1.1 7.3

 

‘ Columbia CNA (Columbia Colistin Nalidixic Acid Agar) agar containing sodium
pyruvate with horse blood overlay (CBNA).
2 Listeria Plating Medium (LPM).

From Fain et al. (1991).

Escherichia coli OlS7:H7 has caused serious foodbome disease outbreaks since
1982 and was first found in Oregon and Michigan. A majority of these outbreaks were
associcated with hamburger (ground beef) consumption at fast food restaurants. The
symptoms include hemorrhagic colitis, hemolytic uremic syndrome and death in serious

cases (Dorn, 1993). Doyle and Schoeni (1984) investigated the survival characteristics of

29

E. coli OlS7:H7 in heated and frozen (-20 0C for 0 to 9 months) ground beef (1 7-20%
fat). There was no difference in the survival of E. coli OlS7:H7 in the inoculated beef
patties after 0, 3, 6 and 9 months frozen storage. They found that the D values were
lower for E. coli OlS7:H7 than for Salmonella (Goodfellow and Brown, 1978). The D
values for E. coli OlS7:H7 were 2390, 270, 70, 45, 24 and 9.6 min at 54.4, 57.2, 58.9, 60,
62.8 and 64.3 °C, respectively. The 2 value was 4.1 0C. By comparing the 2 value of
Salmonella (5 .56 °C) (Goodfellow and Brown, 1978) to that of E. coli OlS7:H7 (4.1 °C),
it was suggested that E. coli 01 57:H7 is more thermally sensitive to changes in
temperature than Salmonella.

Line et al. (1991) conducted similar research to determine the D and 2 values for
E. coli OlS7:H7 at 51.7, 57.2 and 62.8 °C in lean (2% fat) and fatty (30.5%) ground beef
(Table 2.5). They used PCA containing 1% sodium pyruvate media and the 2 hr indole
test to recover E. coli 01 57:H7. Using PCA recovery, higher D values were found in
fatty meat than lean meat, however fat content did not affect the 2 values of E. coli in
beef. Also, different recovery techniques affected the 2 values in lean beef because the
PCA recovery method yielded a higher 2 value (8.3 °C) than the 2 hr Indole recovery
method. The PCA recovery procedure could recover more heat shocked Listeria

monocytogenes after cooking.

30

Table 2.5 Thermal death time curve data and 2 values for Escherichia coli 01 57:H7 in

lean (2% fat) and fatty (30.5% fat) ground beef systems.

 

Fat content Culture medium D value (min) 2 value (°C)

 

51.7 0C 57.2 C’C 62.8 0C

(125 0F) (135 0F) (145 °F)

 

2.0% PCA ' 78.2 4.1 0.30 4.6
30.5% PCA 115.5 5.3 0.47 4.7
2.0% 2-h Indole 80.1 4.0 0.22 4.3
30.5% 2-h Indole 121.0 7.4 ND 2 ND

 

I PCA: Plate count agar (PCA) + 1% sodium pyruvate.
2 ND: Not determined due to insufficient data.

From Line et al. (1991).

Ahmed et al. (1995) examined the D and 2 values for E. coli 01 57:H7 in ground
beef, pork sausage, turkey meat and chicken breast muscle. E. coli 01 57:H7 strain 204P
was inoculated into ground beef containing 7%, 10% and 20% fat and was recovered after
heating on TSA(t1yptic soy agar) (Table 2.6). In general, the higher the fat content, and

the lower the moisture content, the higher the D values. However, the 2 values were

' similar. The 20% fat ground beef had less moisture (61.9%) than that of 10% fat (69.1%)

31

and 7% fat (72.6%) beef, and the D values were higher. Fat levels affect thermal lethality
of E. coli OlS7:H7. In addition to fat content, different recovery techniques, including

the type of media and plating procedures, can cause differences in D and 2 values (Line et

al., 1991).

2.7.2.2 Water-cooked Beef

Goodfellow and Brown (1978) also conducted a thermal death time study of
Salmonella in a water-cooked beef system. Beef rounds were inoculated internally with
Salmonella using dialysis tubing (Willardson et al., 1977). Commercial time-temperature
processing schedules were used. The cooling rate of the 7.3-8.2 kg roasts after cooking
was slower than meat in TDT tubes. Results from the ground beef study (Goodfellow
and Brown, 1978) suggested that a minimum of 100 min was necessary to reduce
Salmonlla by 7-D at 54.4 C’C (130 °F). Due to the slower cooling rate of the water
cooked beef roasts compared to that of a ground beef tube system, water cooking reduced
Salmonella from 1 x 107 CPU/g roast beef to less than 0.3 CF U/g after cooking and
holding for 30 min. No Salmonella were detected after 60 min cooking at 54.4 0C (130

°F).

32

Table 2.6 Thermal death time curve data and 2 values for Escherichia coli 01 57:H7

strain 204P in ground beef, pork sausage, turkey breast and chicken meat

 

 

 

 

 

 

Products Fat content D value (min) 2 value (°C)
50 0C 55 0C 60 0C

ground beef 7% 55.34 11.40 0.45 4.8
10% 80.66 15.30 0.46 4.4

20% 92.67 19.26 0.47 4.4

pork sausage 7% 49.50 6.37 0.37 4.7
10% 62.90 7.83 0.46 4.7

30% 80.64 11.28 0.55 4.6

turkey breast 3% 70.41 6.37 0.55 4.7
11% 115.00 9.69 0.58 4.4

chicken meat 3% 65.24 8.76 0.38 4.5
11% 105.50 9.74 0.55 4.4

 

Form Ahmed et al. (1995).

33

2.7.2.3 Dry Roasted Beef

Goodfellow and Brown (1978) also inoculated the surface of 7.3 - 8.2 kg oven-
roasted roasts with Salmonella. The surface of these round roasts were found to be very
dry (the relative humidity was lower than 0.02). Oven temperatures of 93.2 0C (200 0F),
107.1 0C (225 °F), 121.0 0C (250 °F) and 134.9 0C (275 0F) were used. Salmonella was
eliminated from the surface of the roasts at oven temperature of 121.0 0C (250 °F) which
corresponded to an internal temperature of 54.4 0C (130 0F) with 121 min holding time.
No Salmonella was found on roasts at internal temperatures of 54.4 0C (130 °F) or above
and extremely low numbers of viable Salmonella were found on roasts at internal
temperatures of 51.6 0C (125 °F). The authors concluded that the shape and size of
roasts were the major requirement for processing round roasts to reach a 7-D reduction of
Salmonella when roasts ranging from 2.27 - 4.55 kg and the temperature was a less

important factor compared to the shape and size of roasts.

2.7.2.4 Canned Ham

The USDA method (1986b) for determining the internal temperature of canned
ham measures residual acid phosphatase activity and was adapted from the method by
Kormendy and Gantner (1960, 1967). However, Kormendy et al. (1987) reported that
residual phosphatase activity could vary with can size. The larger the can size, the longer
the time to reach the target internal temperature. The hams were cooked in bags in a
water bath to constant temperatures of 60, 65, 70 and 75 °C for different times. The D

_ and 2 values were determined. They suggested that the thermal death time relationship

34

could offer a more accurate approach to determine thermal processing equivalence. D
values were 3351, 445, 70, and 9 min at 60, 65, 70, and 75 °C, respectively, and a 2 value

of 5.85 °C was found determined using residual acid phosphatase activity in canned ham.

2.7.2.5 Pork Sausage

Ahmed et al. (1995) evaluated the D and 2 values of E. coli OlS7:H7 strain 204P
in pork sausages containing 7, 10 and 30% fat (Table 2.6). In general, as fat content was
increased from 7 to 30%, the moisture content decreased and D values increased. The
authors suggested that microorganisms tended to be more thermally labile at higher water
activities. The authors suggested that 30% fat sausage contained less moisture (51.1%)
than that of 10% fat (70.9%) and 7% fat (73.3%) sausage. The authors suggested that fat
played a protective role for the bacterial cells, so that D values were higher. The authors
concluded that fat content affected heat lethality of E. coli 01 57:H7 in pork sausage

products.

2.7.2.6 Ground Turkey Breast Muscle

Ahmed et al. (1995) also investigated the D and 2 values of E. coli OlS7:H7 strain
204P in turkey meat containing 3 and 11% fat (Table 2.6). The 3% fat turkey breast
contained higher moisture than that of 11% fat (the exact moisture content was not
reported). The D values were higher in meat products with higher fat composition and

lower water content.

35

2.7.2.7 Chicken Muscle

Ahmed et al. (1995) examined the D and 2 values of E. coli OlS7:H7 strain 204P
in chicken meat containing 3 and 11% fat (Table 2.6). As previously reported, fat and
water contents influenced the thermal lethality of E. coli OlS7:H7. Chicken meat with
3% fat had higher moisture than that with 11% fat, and the D values were higher in 11%
fat chicken meat at the same temperature (the exact moisture content was not reported in
the publication).

To summarize the study of Ahmed et al. (1995), meat products from different
species (ground beef, pork sausage, turkey meat, chicken meat) all had higher D values at
higher fat contents. The 2 values of E. coli in different products were similar, ranging
from 4.4 - 4.8 °C. Product composition and different recovery media were the major
factors affecting D values among different meat or poultry products. However, in this
study, meat from various species (Ahmed et al., 1995) did not Show specific differences

in D and 2 values at different temperatures.

2.8 TIME -TEMPERATURE INTEGRATORS (TTI) AS THERMAL
PROCESSING INDICATORS

Three common methods are used to verify that thermal processes have been
properly conducted. They include in situ methods, physical-mathematical methods, and
time-temperature indicators (TTIS) (Hendrickx et al., 1995; Van Loey et al., 1996). Both

the in situ and physical-mathematical methods are well—established and widely used.

36

The in situ method is used to measure changes in a safety or quality attribute
before and after thermal treatment such as microbial counts, nutrient content, color or
texture. This approach is relatively conventional and laborious because data must be
collected before and after the process to verify compliance to thermal processing
schedules. This method may be relatively time consuming and needs trained analysts. It
may be difficult to conduct the experiment due to the detection limit of the analytical
method and or sampling problems.

The physical-mathematical method is used to determine the effect of a thermal
process on a certain attribute based on the time-temperature history of a processed food.
A safety or quality parameter in a food can be used to develop a mathematical model and
to calculate the change after the thermal treatment. To improve the accuracy and
applicability of the model system, a SOphisticated physical-mathematical model is
established using the actual time-temperatrrre data.

The third method, the TTI method, was designed to avoid the limitations
associated with the in situ and physical-mathematical methods, such as collecting detailed
time-temperature data during processing. The definition of a TTI is "a small measuring
device that shows a time-temperature dependent, easily, accurately and precisely
measurable irreversible change that mimics the changes of a target attribute undergoing
the same variable temperature exposure" (Hendrickx etal., 1995). A TTI should be
easily recovered from a food, inexpensive and not affect heat transfer within the food.
Theoretically, the 2 value of a TTI (in the computed range of the time-temperature

profile) should have a 2 value similar to that of the target attribute. A TTI has to

37

demonstrate the same time-temperature dependent response as that of the target attribute
in a food and the temperature should be the only rate-limiting factor in the system. TTIS
can be either intrinsic or extrinsic. Intrinsic TITS are prepared from compounds already
in the target food, so extraction or recovery procedures are necessary to analyze the
concentration of the intrinsic TTI after processing. Extrinsic TITS are added to the food
system and are retrieved back from the food after thermal processing is completed.
Extrinsic TTTs can be encapsulated to prevent changes in inactivation kinetics due to

reactions with the food material.

TTIS can be classified into three categories: biological, chemical and physical.
Biological TTTs, include microbial and enzymatic TTIs. The 2 value of an enzymatic
TITS should be similar to the 2 value of the target index in a food when processed in the
same temperature range. Endogenous enzymatic TITS are generally dispersed throughout
a food system. Enzymatic TTIS are usually less expensive, less time consuming and give

more accurate determinations than microbial TTIs (Hendrickx et al., 1995).

Currently, USDA methods to verify processing adequacy are only used to detect
compliance to a single endpoint temperature in roast beef products. No offical method is
available to verify thermal adequacy when roasts are processed using different USDA
approved time-temperature schedules. Accurate and rapid methods are urgently needed
to verify that proper thermal treatment has been achieved in beef roasts. A TTT is needed
to monitor processing adequacy, if the recently proposed USDA-F SIS lethality standard
(1996b) requiring a thermal process that results in a 7-D reduction in Salmonella is

implemented. Orta-Ramirez et al. (1997) investigated the thermal inactivation kinetics of

38

six endogenous enzymes in bovine semitendinosus muscle: acid phosphatase, lactate
dehydrogenase, phosphoglycerate mutase, peroxidase, glyceraldehyde-3-phosphate
dehydrogenase, and triose phosphate isomerase. TPI had a 2 value of 5.71 °C and was
similar to that of S. senftenberg (5.56 0C) which was used to establish the USDA roast
beef processing schedules (Goodfellow and Brown, 1978; USDA-F SIS, 1995). The
authors suggested that TPI might be used as an intrinsic TTI to monitor thermal adequacy

of roast beef processes.

2.9 BIOCHEMICAL PROPERTIES OF TRIOSE PHOSPHATE ISOMERASE
(T P1)

Triose phosphate isomerase (TPI, D-glyceraldehyde-3-phosphate ketol-isomerase,
EC 5.3.1.1) converts dihydroxyacetone phosphate to D-glyceraldehyde-3-phosphate.
This enzyme is active as a dimer with a molecular mass of 53 kd from porcine muscle
and exists in multiple forms in most higher organisms (Darnall and Klotz, 1975). The
molecular mass of TPI from human liver was 44 kd by gel filtration method or 49.1 kd by
analytical ultracentrifugation. The molecular mass of TPI from rabbit muscle was 49.1
kd by gel filtration method or 45.7 kd by analytical ultracentrifugation (Lee et al., 1971).

TPI is present as isoforms. Five isoforms in rabbit muscle, three isoforms in
horse liver and three isoforms in human liver were observed by gel electrophoresis (Lee
et al., 1971). Sawyer et al. (1972) also studied human TPI from erythrocytes. They
found three variants (I, II and III) and TPI was a dimer with a molecular mass of about 56

kd. The isoelectric points for variants I, II and III were 6.7, 6.5 and 6.1, respectively.

39

Variants I, II and III accounted for 1-5%, 70-75% and 20-25% of total TPI, respectively.
Variants I and III were dimers, AA and BB. Variant II was a heterodimer, AB, and
dissociated and reassociated to AA and BB forms. The amino acid composition of
variant I and III differed in histidine, serine, glycine, valine and leucine. Due to its higher
content of histidine, variant I was assumed to be more basic than III.

Dabrowska et al. (1978) reported that human skeletal muscle TPI, with a
molecular mass of 57.4 kd, existed as a dimer. Eber and Krietsch (1980) isolated TPI
isoforms from human skeletal muscle and compared their immunological properties.
Three isoforms (A,B,C) were isolated by diethylaminoetlryl (DEAE) cellulose
chromatography and identified on gel electrophoresis. The A and C isoforms were
homodimers (AA) and the B isoform was a heterodimer (AB) agreeing with the
conclusions of Sawyer et al. (1972). These three isoforms were examined by
immunodiffusion method and found to have the same immunological properties. They
suggested that A and B polypeptide chains had the same amino acid composition,
molecular weight and antigenicity.

Beisenherz (1995) reported that TPI activity may be reduced to 25% of initial
activity in 0.05 M phosphate solution. Therefore, phosphate ions are inhibitors of TPI.
The optimum pH of TPI is between 7 and 8 and TPI activity decreased by half at pH 6.3
compared to its activity at pH 7-8. Turnover rate at 26 °C was 945,000 moles

substrate/min and the turnover rate was doubled at 38 0C.

40

2.9.1 Determination of TPI Activity in Muscle Tissues

TPI activity can be determined by coupling the following two reactions:

TPI
glyceraldehyde-3-phosphate ------- > dihydroxyacetone phosphate (DHAP)

a-glycerophosphate dehydrogenase
DHAP + NADH > a-glycerophosphate + NAD

 

An unit (U) was defined as 1 pmole of substrate converted per minute. TPI from
horse and liver tissues had specific activities of 3183 and 2397 U/mg (Lee et al., 1971).
Porcine muscle contained 2 mg TPI/g tissue and 1,500U/g muscle (Scopes, 1973);
however, TPI is unstable at pH 5.5 and 55 0C. Smith (1991) reported TPI activity in
different species. TPI had high activity in chicken pectoralis major or white muscle
(1260.0 3: 104.0 U/kg muscle) and sartorius or red muscle (957.0 :1: 65.0 U/kg muscle)
TPI activity decreased when heated to 60, 70 and 80 OC. TPI also had high activity in the
following muscles: 984.0 i 163.0 U/kg muscle in bovine infraspinatus; 1210.0 :1: 156.0
U/kg muscle in bovine semimembranosus; 1720.0 i 47.0 U/Kg muscle in porcine
longissimus dorsi; and 1420.0 :1: 81.4 U/Kg muscle in porcine psoas major. Hsu (1993)
reported that TPI activity of turkey pectoralis major muscle with either 15% water, 15%

brine or 15% curing brine decreased by 99% of its initial activity when heated to 67 °C.

41

- 2.9.2 Purification of TPI from Muscle Tissues

TPI purification procedures have been reported for calf muscle (Beisenherz,
1955), horse and human liver (Lee et al., 1971), human skeletal muscle (Dabrowska et al.,
1978), rabbit muscle (Scopes and Stoter, 1982), frozen chicken breast muscle (Petell et a1.
1982) and chicken breast muscle (Reiss and Schwartz, 1987).

Beisenherz (1955) purified TPI from calf muscle. Muscles were extracted with
0.05% disodium ethylenediaminetetraacetate (EDTA) solution (pH of solution was not
stated) for 30 min at 3 °C. The homogenate was then centrifuged at 2,200 x g for 30 min
and the supernatant was decanted and saved. The pellet was re-extracted. TPI was
fractionated in 30% and 50% acetone. Then, TPI was precipitated in 60% acetone. The
pellet was collected and further precipitated using 70 and 85% ammonium sulfate. After
washing and recrystallization, the authors reported that the purity was increased from
2.5% (initial crude extract) to 100% (purified TPI) with a recovery of 24%.

Lee et al. (1971) purified TPI from horse and human liver tissues. Tissues were
extracted with 0.05% EDTA for 4 hr (pH 6.8), followed by fiactionation in 35% and 50%
acetone. Proteins in the supernatant were precipitated in 60% acetone. The pellet was
dialysed against 0.05% EDTA to eliminate acetone and heated at 40 0C for 15 min, then
50 0C for 30 min. TPI was eluted with 0.1 M NaCl using a Sephadex QAE [diethyl(2-
hydroxypropyl)aminoethyl] A-50 (anion exchange) column. The protein was crystallized
in 3.5 M ammonium sulfate. The crystals were dissolved in phosphate buffer (0.002 M,
pH 7.8) and filtered through a Sephadex G-75 (superfine) column. The specific activities

. of purified TPI were 3,183 I.U.lmg for horse liver and 2,397 I.U./mg for human liver.

42

The specific activities of purified TPI from horse and human tissues were increased 999
and 361 fold, respectively.

Dabrowska et al. (1978) purified TPI from human skeletal muscle. The major
steps included: (1) extraction in EDTA solution (1.5 mM, pH 5.6). (2) acetone
fractionation (3 5%, 45% and 60%), (3) heating to 50 0C for 30 rrrin, then centrifugation
to remove the denatured protein, (4) precipitation in 54% ammonium sulfate. The pellet
was dissolved in a 0.05 M Tris (Tris[hydroxymethyl]-aminomethane hydrochloride)
buffer containing 3 mM EDTA and 1.0 mM 2-mercaptoethanol, pH 7.5, (5) fractionation
on a Sephadex G-100 gel filtration column, (6) fractionation on a DEAE
(Diethylaminoethyl)-cellulose column, and (7) precipitation with ammonium sulfate.
Crystals were formed after 7 days. Crystallized TPI had a specific activity of 7,200
U/mg.

Scopes and Stoter (1982) developed a scheme for purification of several
glycolytic enzymes from rabbit muscle. Rabbit muscle was homogenized and extracted
in a buffer containing 30 mM K phosphate, 1 mM EDTA and 10 mM 2-mercaptoethanol,
pH 7.0. Several fractions were collected depending on the concentration of ammonium
sulfate added to the meat extract. TPI was soluble in 2.44 M ammonium sulfate, but
precipitated in 3.02 M ammonium sulfate at pH 7.0. This precipitate contained five
enzymes: TPI, CK, adenylate kinase, enolase and phosphoglycerate kinase.
Carboxylmethyl-cellulose column chromatography was used to absorb adenylate kinase,
enolase and phosphoglycerate kinase on the column while TPI and CK were eluted. At

_ room temperature, CK was denatured and lost enzymatic activity at pH 6.0 on a

43

carboxylmethyl-cellulose column allowing for the separation of TPI from CK. Finally,
TPI was eluted with MES (2-[N-morpholino]ethane-sulfonic acid) buffer by increasing
the pH to 6.5.

Petell et al. (1982) also reported a method to isolate TPI fi'om chicken breast
muscle. Proteins in the muscle extract were fractionated with 70% and 90% ammonium
sulfate. The pellet was solubilized in phosphate buffer containing 40% ammonium
sulfate. CK was precipitated by adjusting the pH to 5.4 with 1 M acetic acid. The
supernatant was then dialysed against MES buffer (1 mM MES, 1 mM magnesium
acetate, 1 mM EDTA, 1 mM 2-mercaptoeathanol). TPI was eluted from a
phosphocellulose column with a phosphate buffer (pH 6.0, 50 mM phosphate). TPI was
95% pure as determined using SDS-PAGE with Coomassie blue stain.

Reiss and Schwartz (1987) developed a procedure to purify certain glycolytic
enzymes from chicken breast muscle. These enzymes included TPI, aldolase,
glyceraldehyde phosphate dehydrogenase, phosphoglycerate kinase, phosphoglycerate
mutase, enolase, pyruvate kinase, lactate dehydrogenase and CK. Chicken pectoralis
major muscle was homogenized in a MEMT buffer (5 mM MgSO4, 0.4 mM EDTA, 7
mM 2-mercaptoethanol and 50 mM Tris-HCI, pH 6.8). The TPI purification procedure
included: ammonium sulfate fractionation and pH fractionation. Then, TPI and CK were
eluted using a pH gradient in a phosphocellulose column with increasing ionic strength.
Affinity chromatography with Cibacron blue-2 agarose was used as the final purification
step. TPI was eluted at pH 6.8 when applied using a pH 5.0 buffer, whereas, CK was

eluted at pH 6.8 when applied using a pH 5.5 buffer. The specific activity of the crude

44

extract was 7.0 U TPI/mg muscle and the purified TPI had a specific activity of 1200
U/mg muscle. TPI was purified 170 fold from chicken pectoralis major muscle. TPI was
about 99% pure.

Reiss and Schwartz (1987) also prepared polyclonal antibodies for each enzyme.
Antibodies were purified further using immunoadsorbent techniques to yield higher
specific binding between antigens and antibodies. Anti-TPI antibodies were not as
specific as anti-CK, phosphoglycerate kinase, aldolase, glyceraldehyde phosphate
dehydrogenase, enolase, pyruvate kinase and lactate dehydrogenase antibodies to their

own antigen. In addition to TPI, anti-TPI antibodies cross reacted with aldolase.

CHAPTER 3
THERMAL INACTIVATION OF ACID PHOSPHATASE AND

PEROXIDASE IN GROUND BEEF

3.1 ABSTRACT

The USDA-FSIS requires Specific thermal processing schedules for roast beef
products to ensure the destruction of pathogenic bacteria, such as Salmonella and E. coli.
An endogenous muscle enzyme with a thermal inactivation rate constant (2 value) Similar
to that of Salmonella could be used as a time-temperature integrator (TTI) to monitor the
adequacy of roast beef thermal processing. A thermal process sufficient to cause a 7-D
reduction in Salmonella was recently proposed by F SIS as the lethality performance
standard for roast beef, cooked beef and cooked corned beef products. This study
investigated the thermal inactivation kinetics of acid phosphatase (AP) and peroxidase
(PO) to evaluate their potential use as time-temperature indicators of processing adequacy
in roast beef. The D values of AP and PO were determined from 53 to 68 °C in 5.5 °C
increments and their 2 values were calculated from the relevant temperature range. The D
values for AP were 352.9, 26.3, 5.6 and 3.3 min at 53, 58, 63 and 68 °C, respectively,
with a 2 value of 7.41 0C. The D values for P0 were 3871.0, 2678.6, 769.2 and 42.9 min

at 53, 58, 63 and 68 °C, respectively, with a 2 value of 7.80 °C. Neither of these enzymes

45

46

had a 2 value similar to that of Salmonella which was the indicator microorganism used
by the USDA to establish the time-temperature processing schedules for roast beef.

Keywords: time-temperature integrators, roast beef, thermal death time study

3.2 INTRODUCTION

Salmonella and hemorrhagic Escherichia coli are the major pathogenic bacteria
found to cause foodbome disease outbreaks from precooked meats. Enteric pathogens
from animal food products cause from 6.5 to 81 million outbreaks of illness each year in
the US. as reported by the Centers for Disease Control and Prevention (Bean and Griffin,
1990). Although 45,000 cases of Salmonella outbreaks are reported every year in the
United States, foodbome cases of salmonellosis are estimated to range from 790,000 to
3.69 million annually with a median number of 1.92 million cases (Todd, 1994). E. coli
01 57:H7 is an emerging pathogenic bacteria in meat products. There are about 10,000 to
20,000 cases and 200 to 500 deaths attributed to E. coli in the United States each year
(CDC, 1996).

Proper cooking is one way to eliminate pathogenic bacteria from meat products.
Title 9 of the Code of Federal Regulations (U SDA-F SIS, 1995) requires cooked beef and
rare roast beef be heated to one of 16 time-temperature combinations. For example, beef
can be heated to 62.8 0C (145.04 °F) with no holding time or to 54.4 0C (129.92 0F) with
121 min holding time. In May 1996, USDA-FSIS proposed several rules to amend the
current federal meat and poultry inspection regulations to establish “safety margins” for

thermally processed meat and poultry products (USDA-FSIS, 1996b). Salmonella was

47

selected as the indicator organism as it is more thermally resistant than E. coli OlS7:H7.
A 5-D reduction in Salmonella was also proposed by FSIS as the lethality performance
standard for cooked, uncured meat patties. A thermal process sufficient to cause a 7-D
reduction in Salmonella was recently proposed by FSIS as the lethality performance
standard for roast beef, cooked beef and cooked corned beef products. The proposed
regulation allows processors to use more sophisticated thermal treatments and more
flexible time-temperature processing combinations.

Current endpoint temperature tests used by the USDA cannot predict processing
adequacy when different approved time-temperature processing combinations are used.
Accurate and rapid methods are urgently needed to verify that meat products have
received proper thermal processing. The proposed USDA-FSIS lethality standard based
on a 7-D reduction in Salmonella allows for the use of a time-temperature integrator to
monitor processing adequacy. A time-temperature integrator (TTI) is "a small measuring
device that shows a time-temperature dependent, easily, accurately and precisely
measurable irreversible change that mimics the changes of a target attribute undergoing
the same variable temperature exposure" (Hendrickx et al., 1995). A TTT should have the
same 2 value as that of Salmonella, indicating similar thermal inactivation kinetics
(Hendrickx et al., 1995, Van Loey et al., 1996). Therefore, the objective of this research
was to investigate the thermal inactivation kinetics of acid phosphatase and peroxidase
muscle enzymes to find a marker which has a 2 value similar to that of Salmonella. This
endogenous protein could then be used as a TTI to verify adequacy of roast beef

processing.

48

3.3 METHODOLOGY
3.3.1 Sample Preparation

Fresh semitendinosus meat (eye of round) purchased from a local store was
trimmed of external fat and connective tissue. Muscles were cut into 1 cm cubes and 4 or
5 pieces (500-600 g) were randomly placed in plastic pouches and vacuum packaged in
Cryovac® packaging bags (W.R. Grace Co., Duncan, SC 29681). Meat was used
immediately or stored in the blast freezer (-35 0C) for two weeks or less before use. Meat

was thawed overnight at 4 0C (39.2 0F) prior to use.

3.3.2 Thermal Treatments

Meat was ground twice using a 3.175 mm diameter grinder plate in a Hobart
grinder (Model 84181D, Hobart Mfg. Co., Troy, OH 45374) and placed into a 60 ml
syringe (Becton Dickinson and Co., Franklin Lakes, NJ 07417). Two gram of meat was
extruded through a plastic tube (7 cm in length and 0.5 cm in diameter) into a 10 x 75
mm thermal death time (TDT) tube. The glass tubes were then sealed using a flame. The
TDT tubes were heated at four temperatures [53 0C (127.4 °F), 58 0C (136.4 °F), 63 0C
(145.4 °F) and 68 0C (154.4 °F)] for different holding times (53 0C (127.4 °F): 0, 1.5, 3,
4, 5, 6, 7, 8, 9, 10, 11,12 hr; 58 0C (136.4 °F): 0, 2, 4, 8, 10, 15, 20, 30, 35, 45, 50, 60
min; 63 0C (145.4 °F): 0, 60, 75, 90, 105, 120, 135, 150, 165, 180, 195, 210, 240, 260
sec; 68 0C (154.4 CF): 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110 sec) in a water bath
(Model 1268-52, Cole-Farmer Instrument Company, Chicago, IL 60648) connected to a

V digital programmer (Model 1268-62, Cole-Farmer). 'The holding times at each

49

temperature were selected to achieve at least a one log cycle reduction in enzyme activity.
To monitor thermal processing, a RTD (Resistance Temperature Detector) thermocouple
(platinum Pt100 temperature probe, i 0.1 °C accuracy, Solomat Partners LP, Stanford,
CT 06906) was inserted into the center of a TDT tube containing 2 g meat in each
replicate. The water bath temperatures was set 0.2 0C higher than the target temperature.
The thermocouple was connected to a Solomat MPM 200 Modumeter. Temperature and
time were recorded using a Solomat MPM Logger connected to the modumeter. Zero
time was defined as when the internal temperature reached the target temperature. Tubes
were held for the designated holding time and then removed fiom the water bath and
immersed in an ice-water bath for 10 min. Tubes were held at 4 0C (3 9.2 °F) until

assayed within 12 hr. Each experiment was conducted in triplicate.

3.3.3 Preparation of Protein Extracts and Enzyme Assays

The TDT tubes were broken and cooked meat was transferred into scintillation
vials (Research Products International Corp., Mount Prospect, IL 60056). Four milliliters
of cold phosphate buffer saline (PBS, 0.15 M NaCl, 0.01 M sodium phosphate buffer, pH
7.2) was added to each vial. Samples were vortexed for l min, then samples were stirred
on a magnetic plate stirrer for 15 min in a cold room at 4°C. Samples were centrifuged at
4,500 x g for 5 min (Sorvall Superspeed Automatic Refrigerated Centrifuge, Model
RCZB, Norwalk, CT 06852). The supernatant was collected and held at 4°C until used

within 8 hr. Enzyme activities of acid phosphatase (AP) and peroxidase (PO) were

50

determined as described by Bergmeyer (1974). Three replications were performed at

each temperature and enzyme activity was assayed within 12 hr of cooking.

3.3.4 Calculations of D and 2 Values

The D values were calculated at each temperature based on the Laboratory
Manual for Food Canners and Processors (National Canner Association, 1968) except
that survivor curves were calculated using linear regression analysis (LOTUS 1-2-3,
Version 1.0, Lotus Development Corp., Cambridge, MA 02142). TDT curves were
constructed by plotting D values (min) vs. temperature (° C). 2 Values were calculated as
the absolute value of the reciprocal of the slope of the TDT curve using regression

analysis as described above.

3.3.5 Proximate Analysis
Fat and moisture contents of ground beef were determined using AOAC (1990)
methods 960.39 and 950.46B, respectively. The pH was determined by homogenizing 50

g ground beef with 50 ml of double distilled water using a Waring“ blender for 30 s.

3.4 RESULTS AND DISCUSSION

Raw ground semitendinosus meat contained 72.8% moisture, 3.8% fat and the pH
was 6.0. Acid phosphatase had a D value of 352.9 min (5.88 hr) at 53 °C, which
decreased to 3.3 min at 68 °C (Table 3.1). Peroxidase activity was stable at 53 and 58 °C

. with the D values of 3871.0 min (64.52 hr) and 2678.6 min (44.64 hr) (Table 3.2).

51

Table 3.1 D values and regression analysis for acid phosphatase activity from ground

beef heated at 53, 58, 63 and 68 °C

 

 

Temperature (° C) y-intercept slope R2 D value (min)
53 1.134 -0.170 0.69 352.9
58 1.488 -0.038 0.94 26.3
63 0.791 -0.003 0.98 5.6
68 1.296 -0.005 0.81 3.3

 

52

Table 3.2 D values and regression analysis for peroxidase activity from ground beef at

53, 58, 63 and 68 °C

 

 

Temperature (° C) y-intercept slope R2 D value (min)
53 1.983 -0.016 0.88 3871.0
58 1.570 -0.022 0.70 2678.6
63 1.712 -0.001 0.60 769.2
68 1.655 -0.023 0.79 42.9

 

53

Peroxidase activity decreased rapidly at 63 and 68 °C; the D values were 769.2 and 42.9
min, respectively (Table 3.2). The D values of PO at 53 and 68 0C were at least ten times
higher than that of AP, whereas, the D values of PO at 58 and 63 °C were more than one
hundred times higher than that of AP (Table 3.2). Thus, P0 was a more thermally stable
enzyme than AP from 53 to 68 °C.

Orta-Ramirez et al. (1997) determined the D values of glyceraldehyde-3-
phosphate dehydrogenase, lactate dehydrogenase, phosphoglycerate mutase, PO, triose
phosphate isomerase in ground beef. PO had higher D values at 53, 58, 63 and 68 0C
than all enzymes examined by Orta-Ramirez et al. (1997) except for lactate
dehydrogenase at 53 °C. Lactate dehydrogenase had a D value of 6968.6 min at 53 °C.
Acid phosphatase and phosphoglycerate mutase had similar D values at 53 0C (352 min
and 325 min). However, AP was less stable at 58 °C than all enzymes, except triose
phosphate isomerase (Orta-Ramirez et al., 1997).

The apparent 2 values of AP and P0 in ground beef were 7.41 and 7.80 °C (Table
3.3), respectively. Orta-Ramirez et al. (1997) reported the 2 values for glyceraldehyde-3-
phosphate dehydrogenase, lactate dehydrogenase, phosphoglycerate mutase, and triose
phosphate isomerase of 4.71, 3.98, 5.18, and 5.56 0C, respectively. Levieux et al. (1995)
reported that the 2 values of lactate dehydrogenase M4 and lactate dehydrogenase H4,
(lactate dehydrogenase has five isoforms which are comprised of two polypeptide
subunits, H and M: H4, H3M, HZMZ, HM3, and M4 [Holbrook et al., 1975]) were 6.1 and
5.7 °C, respectively using irnmunodiffusion. In general, AP and PO had higher 2 values

(thermal inactivation rates) than those of other enzymes. The apparent 2 values of

54

Table 3.3 2 Values and regression analysis from thermal death time studies of acid

phosphatase (AP) and peroxidase (PO) from ground beef

 

Marker protein y-intercept slope R2 2 value (° C)

 

AP 9.477 -0.l35 0.92 7.41

PO 10.636 -0.128 0.87 7.80

 

55

Salmonella senftenberg and E. coli OlS7:H7 in ground beef were 6.25 °C and 5.57 °C,
respectively (Orta-Ramirez et al., 1997). The 2 value of E. coli OlS7:H7 was close to
that of the USDA roast beef process. Goodfellow and Brown (1978) reported a 2 value of
5.56 °C in ground beef inoculated with Salmonella serotypes. The 2 value of Salmonella
from Goodfellow and Brown (1978) was less than that reported by Orta-Ramirez et al.
(1997). The sixteen time-temperature processing schedules allowed by the USDA were
established using a 7-D reduction in Salmonella in roast beef (Goodfellow and Brown,
1978). A 7-D reduction in Salmonella was recently proposed by the USDA-FSIS to
develop a new thermal processing regulation for roast beef products. Our results indicate
that AP and PO could not be used as endogenous TIT to evaluate the processing

adequacy of roast beef since their 2 values were greater than that of Salmonella.

CHAPTER 4
IDENTIFICATION OF MARKER PROTEINS IN ROAST BEEF TO

VERIFY COMPLIANCE TO USDA PROCESSING SCHEDULES

4.1 ABSTRACT

Our objective was to identify a marker protein that could determine the adequacy
of roast beef processing. Three adequate (high, medium and low temperature processes
were 62.2 °C/5 min, 58.3 °C/24 min, and 54.4 “0121 min, respectively) and
corresponding inadequate processing (0.5 and 1 log cycle reduction in holding time from
adequate process) schedules were used. The residual enzyme activity of peroxidase (PO),
acid phosphatase (AP), and triose phosphate isomerase (TPI) were determined. Lactate
dehydrogenase (LDH) content was determined using immunoassay. In the ground beef
water bath model system, TPI activity averaged 2.6 U/kg in adequately processed ground
beef and increased to 4.9 and 13.3 U/kg when inadequately processed by reducing the
holding time 0.5 log and 1.0 log, respectively. TPI activity was similar within all
adequately processed beef and TPI activity increased (P < 0.0001) when processing time
was inadequate. In the MSU smokehouse pilot study, TPI activity in adequately
processed roast beef averaged 1.6 U/kg and increased to 3.7 and 7.8 U/kg when
inadequately processed by reducing the holding time 0.5 and 1.0 log, respectively. TPI

was identified as the best marker protein in both ground beef and roast beef cooked using

56

57

medium to high temperature processes, as TPI activity was the same when compared
within adequate cooking treatments, but increased as cooking time was decreased. The
other enzymes were not suitable markers as either differences in activity were found
within adequately processed beef treatments or no differences were found between
adequately and inadequately processed beef.

KEYWORDS~ beef, endpoint temperature, processing

4.2 INTRODUCTION

The United States Department of Agriculture - Food Safety Inspection Service
(U SDA-FSIS) has established specific processing requirements for precooked meat and
poultry products. The purpose of these regulations is to ensure the destruction of
pathogenic bacteria and viruses that could cause foodbome illness in humans and live-
stock (Townsend and Blankenship, 1989). Cooked beef and roast beef are required to be
heated to 62.8 °C (145 °F) with no holding time or to one of sixteen different time-
temperature combinations outlined by USDA-F SIS (1995). The schedules were based on
the thermal inactivation in Salmonella. In May 1996, USDA-FSIS proposed several rules
to amend the current federal meat and poultry inspection regulations to enhance the safety
of thermally processed meat and poultry products (U SDA-F SIS, 1996b). A 7-D
reduction in Salmonella was proposed by F SIS as the lethality performance standard for
processing of roast beef, cooked beef and cooked corned beef products.

The current methods for endpoint temperature (EPT) determination used by

g USDA-F SIS are a Protein Coagulation Test (U SDA-FSIS, 1986a) and a Bovine Catalase

58

Test (Eye, 1982; USDA-F SIS, 1989). These methods have been shown to be subjective
and inaccurate in predicting the actual EPT achieved during processing (Townsend and
Blankenship, 1989). In fact, the major disadvantage of the current tests is that they
cannot detect adequacy of processing when roast beef is processed using current USDA
time-temperature schedules.

Orta-Ramirez et al. (1996) determined the 2 value of six potential marker enzymes
in bovine semitendinosus muscle. The 2 values of acid phosphatase, lactate
dehydrogenase (LDH), phosphoglycerate mutase, peroxidase, glyceraldehyde-3-
phosphate dehydrogenase and triose phosphate isomerase (TPI) were 7.41, 3.99, 4.11,
7.80, 4.56 and 5.71 °C, respectively. TPI had a 2 value similar to that of Salmonella
senfienberg (5.56 °C) which was used to establish the USDA approved roast beef
processing schedules. The authors suggested that TPI might be used as an endogenous
time-temperature integrator (TIT) to determine the adequacy of thermal processing of
roast beef and beef patties. The definition of a TTI is "a small measuring device that
shows a time-temperature dependent, easily, accurately and precisely measurable
irreversible change that mimics the changes of a target attribute undergoing the same
variable temperature exposure" (Hendrickx et al., 1995).

The objective of this study was to identify endogenous marker enzymes that could
be used to determine if roast beef has been adequately processed to meet the USDA time-
temperature schedules using model and pilot studies. Adequately and inadequately
processed beef roasts were used to evaluate the ability of several potential marker

, proteins to differentiate among the processing schedules.

59

4.3 MATERIALS AND METHODS
4.3.1 Pilot Studies
4.3.1.1 Roast Beef Processing

Choice top round roasts (semimembranosus, SM) purchased from Ada Beef
Company (Ada, MI 49301) were vacuum packaged and transported in ice to the MSU
Meat Laboratory. All roasts were trimmed of external fat and connective tissue. Each
roast was divided into two pieces, cut into a rectangular shape (about 40 cm x 25 cm) and
vacuumed packaged in Cryovac packaging bags (W.R. Grace Co., Duncan, SC, 29681).
Roasts were stored at 4 °C for less than 16 hr before processing.

Three difi‘erent USDA approved time-temperature schedules were used to
determine the effect of processing on peroxidase activity in roast beef (Table 4.1).
Adequately cooked roast beef was processed to internal temperatures of 62.2 °C (144
°F)/5 min, 58.3 °C (137 °F)/24 min, and 54.4 0C (130 0C)/121 min. Under-cooked roasts
were prepared by reducing the processing time by 0.5 log cycle at each temperature. For
acid phosphatase, LDH and TPI, two different processing schedules were used. Roast
beef was adequately cooked to internal temperatures of 62.2 °C (144 °F)/5 min and 54.4
°C (130 oC)/121 min. Under-cooked roasts were processed by reducing the holding time
(min) by 0.5 and 1.0 log cycle at each temperature.

The smokehouse conditions used for the low, medium and high temperature
schedules are listed in Tables 4.2, 4.3 and 4.4, respectively. A microprocessor controlled

smokehouse was used as the oven or cooking chamber. Internal temperatures were

60

 

EE side 2.: Co San

as crap 2: O6 4.:

ES Nine 33 Do v.%

ES :56 R: Oo 3m

58 wkmo um: Do Own

ES véfie hm: Uo Own

EE ace 3.: O6 Se 368 22232
38: E 52268 m2 We

58 mdkme New: Uo _. G woxooodopcz
GE: 5 gauge“: wo_ o. 3

58 m. CEO on: Do Yon coxoooLopcD

 

30er 238388 Bet—

mmooea 238388 8:622

8303 238888 :9:

 

Emma ween 3882932: use .3833? 089.5 8 com: .6123anan was $8: mammooecm :4 2an

61

.88: Homes 3:88 .85 mm @3082 263 named a

 

 

a. 2: O. 2% a. 9.: O. :e a. 3.: O. 3% .8. a
a. 3: O. 2% a. me: O. .3 a. 8: O. Ex 2: m
a. 3: O. 4% E. 8: O. _.: E. 8: O. as. 8 m
a. a: O. as E. a: O. new E. 8: O. ”.2 oe _
838382 £3 33 2282.83 £3 in 238888 12:25 €25 08C. 03%

 

683 ammo.— pemmoooa Ea o2 Moe *4.va 2382.88 32 282m S cows 3.60:8 Omsonoerm N44 033.

 

62

.38: Homes p282 >2: mm @9588 203 named a

 

 

a. me: O. Se A... S: O. «.8 a. 2: O. 3% . em a
C. an: O. fie a. S: O. toe a. 2: O. in as m
E. S: O. a? a. 2.: O. he. a. 8: O. a? 8 m
a. a: O. 3» a. a: O. Sm E. 8: O. :m 8 _
838880315 83 238388 £3 in 238888 E585 AEEV 08$. 03%

 

Leon awe @3885 E. >2 no. 0me 05:83.5“ 6:62: 280:. 8 wow: 2.60:8 managed—08m me Dish

 

63

.88: $99 3:82 >2: mm 3382 203 398M a

 

 

G. en: O. 36 a. B: O. QR S. me: O. fie . ea 4
a. a: O. fie a. a: O. 4% a. 9.: O. ode 8 m
a. me: O. 3. a. E: O. Ben S. 8: O. $4 8 N
E. a: O. 3» a. a: O. new a. 8: O. 3m 8 E
838383 £3 83 288388 £39 in 238388 3525 AEEV DEC. Owfim

 

m8: came ecmmooea £6 3: moo NNS 838595“ :9: 2303 2 p8: 3.62.8 Omzogcxofim 34 033.

64

monitored by platinum Resistance Temperature Detector (RTD) probes, 0.33 cm
diameter, i 0.1 0C accuracy (Omega, Stanford, CT, 06906), inserted in the geometric
center of the roasts. The thermocouples were calibrated at the target temperature using
manufacturer’s guidelines. Two additional thermocouples of the same type were also
placed in the geometric center of additional roasts to monitor internal temperatures.
When processing was completed, the roasts were removed and the center core (about
150-200 g) of the roast was immediately excised, wrapped in a plastic bag and then
placed in ice slush for about 30 min until the internal temperature decreased to 32.2 °C
(90 °F). The cooled roasts were stored at 4 °C (39.2 °F) before being analyzed within 12
hr. Twenty-five grams of meat close to the thermocouple position was removed from
each roast for extraction. Thermal processing was conducted using three separate

smokehouse runs for each processing schedule.

4.3.1.2 Proximate Composition

Proximate composition of semimembranosus muscle was analyzed for protein, fat
and moisture contents using AOAC (1990) standard methods 981.10, 960.39 and
950.46B, respectively. The pH was determined by homogenizing 10 g ground beef with

90 ml of double distilled water using a WaringTM blender for 30 s.

4.3.1.3 Preparation of Protein Extracts and Enzyme Assays
Meat (25 g) was minced and extracted in 3 volumes (75 ml) of 0.15 M NaCl, 0.01

M sodium phosphate buffer, pH 7.2, for 90 sec in a Waring Blenderm. After stining with

 

65

a motorized propeller for 1 hr, the suspension was centrifuged at 4,500 x g for 10 min to
obtain the supernatant containing the soluble sarcoplasmic proteins. Acid phosphatase
and LDH activities were determined using commercial Sigma® kits (acid phosphatase,
104-AL; lactate dehydrogenase, DG1340-K). Peroxidase activity was determined as
described by Bergmeyer (1974). TPI activity was determined based on Bergmeyer
(1983) except that the assay mixture contained 1.0 ml triethanolarnine buffer (TEA, 0.2
M, pH 8.0), 0.2 ml glyceraldehyde-3-phosphate (15 mM), 10 pl NADH, sodium salt (10
mg/ml), 10 pl glycerophosphate dehydrogenase (G-6751, Sigma®; 19 mg protein/ml, 170
units/mg protein) and 10 pl meat extract (Wang et al., 1996). The change in absorbance
at 340 nm was followed for 1 min at 25 °C. Enzyme activity was expressed as units per

liter of meat extract (U/L meat extract) and units per kilogram of meat (U/kg meat).

4.3.2 Model System Studies
4.3.2.1 Ground Beef Cooking

Fresh SM meat purchased fi'om a local store was trimmed of external fat and
connective tissue. Meat was cut into 1 cm cubes and ground twice through a 3.175 mm
diameter plate in a Hobart grinder (Model 8418D, Hobart Mfg. Co., Troy, OH, 45374).
The ground beef was placed into a 60 ml syringe (Becton Dickinson and Co., Franklin
Lakes, NJ 07417). Two grams of ground meat were compressed through a 7 cm long
plastic tubing into a 10 x 75 mm thermal death time (TDT) tube (60825-538, VWR,
Kirnax® 51, South Plainfield, NJ 07080). The glass tubes were then sealed using a flame.

TDT tubes were cooked in a water bath (Model 1268-52, Cole-Parmer, Chicago, Ill,

 

66

60648) connected to a digital programmer (Model 1268-62, Cole-Partner). The
temperature of the water bath was set at the target temperature or 1 °C higher than the
target temperature. An RTD (Resistance Temperature Detector) thermocouple (platinum
Pt 100 temperature probe, 0.15 cm in diameter, i 0.1 °C accuracy, Solomat Partners LP,
Stanford, CT 06906) was inserted into the center of a TDT tube containing 2 g meat.
Three different USDA approved time-temperature schedules were used (Table 4.1).
Ground meat was adequately processed to internal temperatures Of 62.2 °C (144 °F)/5
min, 58.3 °C ( 137 °F)/24 min, and 54.4 °C (130 °F)/121 min and under cooked by
reducing the processing time by 0.5 and 1.0 log cycle. Zero time was defined as when the
internal temperature reached the target temperature. After cooking, tubes were removed
from the water bath, immersed in an ice bath for 10 min and held at 4 °C until assayed

within 12 hr. Each experiment was conducted in triplicate.

4.3.2.2 Preparation of Protein Extracts and Enzyme Assays

TDT tubes were broken and cooked meat was transferred into scintillation vials
(Research Products International Corp., Mount Prospect, IL 60056). Eight milliliters of
0.15 M NaCl, 0.01 M sodium phosphate buffer, pH 7.2 were added to each vial. Samples
were vortexed for 1 nrin, then stirred on magnetic plates for 15 min at 4 °C. Samples
were centrifuged at 4,500 x g for 5 min. The supernatant was collected and held at 4 0C
until used within 8 hr. TPI enzyme activity was determined as described previously. The
LDH content was determined by enzyme-linked immunosorbent assay (ELISA) (Orta-

Ramirez etal., 1996).

 

67

4.3.3 Statistical Analysis

In the pilot study, top round roasts, semimembranosus muscles, were smokehouse
processed in triplicate using three different processes (low, medium and high temperature
processes). Each process was conducted in triplicate for a total of 9 separate smokehouse
processing runs. In the model system, ground bovine semimembranosus muscles, were
thermally processed in triplicate using three different processes (low, medium and high
temperatures). Cooking was conducted in triplicate for a total of 9 separate water bath
processing runs. Data were analyzed using two way AN OVA (analysis of variance)
(treatment x replication). The linear or curvilinear change in enzyme activity or

concentration with temperature was determined using the polynomial contrast test by

SAS software (SAS Institute Inc., Version 6.1, 1995, Cary, NC 27513).

4.4 RESULTS AND DISCUSSION

A marker protein must meet two criteria to function as a TTI in roast beef. First,
its concentration or activity should decrease as processing time at the same temperature is
increased. Second, the marker protein concentration or activity within each time-
temperature cooking schedule (adequately vs. inadequately processed) should be
constant. An ideal indicator should be present at the same amount or activity in
adequately processed beef regardless of the time-temperature schedule used.

In the pilot smokehouse study at MSU, roast beef from semimembranosus meat

contained 59.7% moisture, 10.1% fat and 21.8% protein. The pH of the meat was 5.8. In

 

68

the water bath model system, raw ground semimembranosus meat contained 68.9%
moisture, 3.5% fat and 24.8% protein. The pH of the meat was 6.0.

In roast beef pilot studies at MSU, peroxidase activity was the same in adequately
cooked roasts processed at medium and high temperatures, but was different in roasts
adequately processed using the low temperature process (Table 4.5). Peroxidase activity
was different when roasts were adequately processed and undercooked by reducing the
processing time by 0.5 log cycle in the low temperature process. Peroxidase activity was
the same in adequately processed and inadequately processed roasts cooked to medium
and high temperatures. These results indicated that peroxidase could not be used as a
time-temperature integrator for roast beef because it could not differentiate between
adequately and inadequately processed roasts. These results are supported by Orta-
Ramirez et al. (1997) who reported a 2 value of 7.80 °C for peroxidase. The 2 value is
different from that of Salmonella (5.65 0C) used to establish the USDA roast beef
processing schedules (Goodfellow and Brown, 1978; USDA-FSIS, 1995).

Acid phosphatase activity was Similar between equivalent processing schedules
for both low and high temperature processes and averaged 223.6 U/kg meat (Table 4.6).
Acid phosphatase activity decreased as cooking time was increased within each process.
Based on our pilot study, acid phosphatase might be a promising marker protein.
However, acid phosphatase is a glycoprotein bound to the cell membrane. It is present in

only small amounts in animal tissues and milk systems and is difficult to purify

69

Table 4.5 Peroxidase activity (U/kg meat) in adequately processed and underprocessed

roast beef prepared in a smokehouse

 

Low temperature Medium temperature High temperature

(54.4°C;130°F) (58.3 °C; 137°F) (622°C; 144°F)

 

Under-cooked 74 :t 7Aa 48 :1: 7Ab 51 i 14Ab
(0.5 log reduction in time)

Ba

Adequately cooked 35 i 14 40 :i: 15A?“1 38 i SAa

 

as
Means (:1: standard deviation) in the same column followed by the same letter are not

different (p > 0.05).

ab
Means (i standard deviation) in the same row followed by the same letter are not

different (p > 0.05).

70

Table 4.6 Acid phosphatase activity (U/kg meat) in adequately processed and

underprocessed roast beef prepared in a smokehouse

 

 

Low temperature a High temperature a
(54.4 °C; 130 °F) (62.2 °C; 144 °F)
Under-cooked 490.7 a 45.07b 370.5 :1: 37.63b
(1.0 log reduction in time)
Under-cooked 312.6 a 50.65b 258.8 a 102.05b

(0.5 log reduction in time)

b b

Adequately cooked 275.7 i 65.12 171.4 i 70.99

 

5 Means (i standard deviation) in the same column showed a linear response to

temperature; 54.4 °C (p < 0.0035) and 62.2 0C (p < 0.0017).

b
Means (i standard deviation) in the same row followed by the same letter are not

different (p > 0.05).

7l

(Bingham and Zittle, 1963; Chaimovich and Nome, 1970; Debruyne, 1983; Farrell at al.,
1988; Bingham and Garver, 1990).

The 2 value (7.41 °C) of acid phosphatase reported by Orta-Ramirez et al. (1997)
was different from the 2 value (5.56 °C) of Salmonella senftenberg used to establish the
USDA roast beef processing schedules. However, we cannot rule out acid phosphatase as
a potential TIT based on 2 values alone. Further experiments are needed to verify the
applicability of acid phosphatase as a TIT. Pilot studies need to be conducted to confirm
the applicability of acid phosphatase as a TTI.

LDH concentration decreased as processing time was increased within low and
high temperature processes (Table 4.7). However, the concentration of LDH differed
when equivalent processing schedules were used at low (54.4 °C) and high (62.2 0C)
temperatures. For example, the LDH concentration at the adequately cooked schedule of
low temperature was 1098 pg/ g meat and of high temperature was 5.6 pg/g meat. LDH is
a tetrameric protein consisting of two polypeptide subunits, H and M. LDH exists as five
isoforms H4, H3M, HzMz, HM3, and M4 (Holbrook et al., 1975). The 2 value of LDH
(3.98 0C) from Orta-Ramirez et al. (1997) was different from that of LDH M4 (6.1 0C) and
LDH H, (5.7 0C) reported by Levieux et al. (1995). Orta-Ramirez et al. (1997) used an
LDH enzyme assay, whereas Levieux et al. (1995) used single radical immunodiffusion
(SRID) assay to quantify changes in LDH. Levieux et al. (1995) reported that epitopes of
LDH M4 and LDH H4 recognized by anti-LDH M4 or anti-LDH H4 polyclonal antibodies
were both conformational and sequential. Therefore, only native forms of LDH M4 and

LDH H, were quantified using the SRID method. The conformation of LDH required for

72

Table 4.7 Lactate dehydrogenase concentration (pg/ g meat) in adequately processed and

underprocessed roast beef prepared in a smokehouse

 

 

Low temperature3 High temperature21
(54.4 °C; 130 °F) (62.2 °C; 144 °F)
Under-cooked 1495.0 3: 562.5b 616.3 a 71.9c
(1.0 log reduction in time)
Under-cooked 1241.0 :1: 201.5b 37.5 a 13.9c
(1.0 log reduction in time)
Adequately cooked 1098.0 1 64.6b 5.6 :1: 57°

 

a Means (i standard deviation) in the same column Showed a linear response to

temperature; 54.4 °C (p < 0.0727) and 62.2 °C (p < 0.0105).

be
Means in the same row followed by the same letter are not different (p > 0.05).

73

enzyme activity may not be related to the epitopes recognized by the polyclonal
antibodies, thus causing resulting in different 2 values. Since LDH concentrations were
different at equivalent processing schedules in the pilot study (Table 4.7) and the 2 values
of LDH (Orta-Ramirez et al., 1997) are different from that of Salmonella (z = 5.65 °C)
used to establish the USDA roast beef processing schedules (Goodfellow and Brown,
1978; USDA-F SIS, 1995), this protein marker may not be able to verify the thermal
adequacy of roast beef.

In pilot smokehouse study at MSU, TPI activity decreased as cooking time
increased when processed using the high temperature schedule (P < 0.0160) (Table 4.8).
However, TPI activity did not decrease as cooking time increased using the low
temperature schedule (P < 0.2076). TPI activity was the same when equivalent processes
were used at low and high temperatures. For example, TPI activity of adequately
processed schedules at low temperature was 2.05 U/kg meat and at high temperature it
was 1.58 U/kg meat. In the ground beef model system, TPI activity increased (P <
0.0001) as processing time decreased in all three processing schedules (Table 4.9). TPI
activity was similar in the low, medium and high temperature processes when roasts were
adequately processed and inadequately processed by reducing the processing time by 0.5
log cycle. TPI had similar activity in both medium and high temperature processes when
roasts were inadequately processed by reducing the processing time by 1.0 log. Our
results support those of Orta-Ramirez et al. (1997) who found that the 2 value (5.56 0C) of
TPI was similar to the 2 value of Salmonella (5.65 0C) used by the USDA to establish

processing schedules for roast beef. Based on both pilot and model studies, TPI might be

74

Table 4.8 Triose phosphate isomerase activity (Ii/.g meat) in adequately processed and

underprocessed roast beef prepared in a smokehouse

 

 

Low temperature3 High temperature8
(54.4 0C; 130 °F) (62.2 °C; 144 °F)
Under-cooked 5.0 a 1.99b 7.8 a 3.33b
(1.0 log reduction in time)
Under-cooked 4.5 i 3.89b 3.7 i 3.33b
(0.5 log reduction in time)
Adequately cooked 2.1 :1: 1.26b 1.6 :t 1.00b

 

a Means (d: standard deviation) in the same column showed a linear response to

temperature; 54.4 °C (p < 0.2076) and 62.2 °C (p < 0.0160).

b
Means (1 standard deviation) in the same row followed by the same letter are not

different (P > 0.05).

75

Table 4.9 Triose phosphate isomerase activity (U/kg meat) of beef processed using three USDA
approved schedules for roast beef and underprocessed by reducing the processing time by 0.5 and

1.0 log cycle in a ground beef water bath model system3

 

Low temperature Medium temperature High temperature

 

(54.4 °C; 130 °F) (58.3 °C; 137 °F) (62.2 °C; 144 °F)
Under-cooked 18.2 a- 247° 10.3 :1: 231° 11.5 a 214°
(1.0 log reduction in time)
Under-cooked 4.7 a 039° 3.7 a: 098° 6.2 a 089°
(0.5 log reduction in time)
Adequately cooked 2.1 a 018° 2.3 a. 082° 3.5 a: 056°

 

a Means (:1: standard deviation) in the same column showed a linear response to temperature (p <

0.0001).

be Means (i standard deviation) in the same row followed by the same letter are not different (p

> 0.05).

76

a potential TTI for evaluation of the thermal processing adequacy using medium (58.3

0C) and high (62.2 0C) temperature schedules for beef roasts.

Acknowledgment
The authors gratefully acknowledge the financial support of the USDA-CSREES,

American Meat Institute and MSU Agricultural Experiment Station.

CHAPTER 5
DEVELOPMENT OF AN ELISA TO QUANTIFY TRIOSE PHOSPHATE

ISOMERASE IN COOKED BEEF

5.1 ABSTRACT

Triose phosphate isomerase (TPI) was identified as a potential endogenous time-
temperature integrator (TTT) in beef roasts to verify the adequacy of processing in
previous studies. Our objectives were to (I) prepare an ELISA using anti-TPI antibodies
and (2) verify that the ELISA can quantify TPI in cooked ground beef. TPI was purified
from bovine semimembranosus muscle by a series of procedures including 55% acetone
fractionation, 90% ammonium sulfate fractionation, and carboxylrnethyl cellulose cation
exchange chromatography. The activity of purified TPI was increased 112 fold by
purification. A single band of TPI was observed on SDS-PAGE. Polyclonal antisera
(PAb) were raised in rabbits against purified TPI and yielded titers of 108 after 11 weeks
of immunization. A sandwich ELISA was developed using PAb for capture and
biotinylated PAb for detection. Cross reactivities of antibodies with TPI originating from
different animal species, muscle protein concentrates, and common meat starter cultures
were examined using the ELISA and Western blots. Ground beef was heated in 10 x 75
mm tubes to temperatures between 48.9 °C and 76.7 °C in 5.5 °C increments, then

proteins extracted using 0.15 M NaCl, 0.01 M Na phosphate buffer, pH 7.2. The band

7"!

78

representing TPI on SDS-PAGE gels and Western blots decreased in intensity as
processing temperature was increased. Both TPI concentration and activity in extracts of
cooked ground beef decreased (P < 0.0001) as temperature was increased. Taken
together, the results indicated that both the ELISA and enzyme assay were able to detect
differences in TPI due to cooking temperature of beef.

Keywords: cooking, beef, triose phosphate isomerase, immunoassay

5.2 INTRODUCTION

PrOper cooking is an easy method to effectively eliminate pathogens from meat

products. Currently, the USDA-FSIS requires that roast beef be processed to one of 16
time-temperature combinations. For instance, beef can be heat processed to 62.8 0C

(145.04 °F) with no holding time or processed to 54.4 0C (129.92 °F) with 121 minutes
holding time. A 7-D reduction in Salmonella, the major enteric pathogen causing
foodbome illness in meat, was proposed as the lethality standard for thermal processing
of roast beef, and corned beef products (Goodfellow and Brown, 1978; USDA-FSIS,
1996b). The ultimate purpose of the 7-D reduction in Salmonella is to destroy pathogenic
bacteria in the roast beef.

A time-temperature integrator (TTI) is "a small measuring device that shows a
time-temperature dependent, easily, accurately and precisely measurable irreversible
change that mimics the changes of a target attribute undergoing the same variable
temperature exposure" (Hendrickx et al., 1995). A TTT can be used to verify roast beef

processes. A TTI should have a 2 value or activation energy similar to that of the

79

microorganism used to establish the roast beef processing schedules required by the
USDA.

Orta-Ramirez et al. (1996) investigated the thermal inactivation kinetics of six
endogenous enzymes, acid phosphatase, lactate dehydrogenase, phosphoglycerate mutase,

peroxidase, glyceraldehyde-3 ~phosphate dehydrogenase and triose phosphate isomerase

(TPI), from bovine semitendinosus muscle. TPI had a 2 value (5.71 0C) similar to that of
S. senfienberg (z = 5.56 °C) which was used to establish the USDA approved roast beef
processing schedules (Goodfellow and Brown, 1978; USDA-FSIS, 1996). The authors
suggested that TPI might be used as an endogenous TTI to determine the adequacy of
thermal processing of roast beef and beef patties. Thus, TPI was suggested as an intrinsic
TTT to verify compliance to the USDA roast beef processing schedules.

TPI was identified by Hsu et al. (Chapter 4) as the best marker protein in both
ground beef and roast beef processed from 58.3 to 62.2 0C as TPI activities were the same
when compared within cooking treatments, but increased as cooking time was decreased.
One adequate cooking (high, medium and low temperature processes of 62.2 0C/5 nrin,
58.3 °C/24 min, and 54.4 °C/121 min, respectively) and two inadequate cooking
schedules were evaluated at each temperature. In a ground beef water bath model system,
TPI activity was similar when the meat was adequately processed using both medium and
high temperature processes and TPI activity decreased (P < 0.0001) as processing time
was increased within each process. In a roast beef pilot smokehouse study at MSU, TPI

activity was similar when the meat was adequately processed using low and high

80

temperatures, but TPI activity only decreased (P < 0.05) as processing time was increased
in the high temperature process.

Immunoassays have been used for determining endpoint temperatures of several
meat products, including ground beef (Wang et al., 1995; Orta-Ramirez et al., 1996;
Smith and Desrocher, 1996) and poultry products (Wang et al., 1992; Abouzied et al.,
1993; Wang et al., 1993; Desrocher 1994; Wang et al., 1994; Smith et al., 1996; Smith
and Desrocher, 1996). Immunoassays are more sensitive and less expensive than
enzymatic methods and highly specific. The objectives of this study were to (1) develop
an ELISA using anti-TPI antibodies, and (2) verify the applicability of the sandwich

ELISA to quantify TPI in a cooked ground beef model system.

5.3 MATERIALS AND METHODS
5.3.1 Purification of TPI

Triose phosphate isomerase was purified from bovine semimembranosus muscle
(top round choice muscle) purchased from a local meat retailer, afier removing the cap
muscle. All procedures were performed in a cold room at 4 0C unless otherwise
indicated. Meat was diced into 1 cm cubes and visible connective tissue was removed.
Meat was homogenized with 2.5 volumes of cold extraction buffer (30 mM potassium
phosphate, pH 7.0, containing 1 mM EDTA and 10 mM 2- mercaptoethanol) in a Waring
blenderTM at high speed for two intervals of 45 sec. The homogenate was stirred at room
temperature for 30 min and then centrifuged at 10,400 x g for 15 min. F at was removed

from meat extracts by filtering through cheesecloth.

81

Acetone (~15 0C) was added to the meat extract to reach a final acetone
concentration of 40 % and the mixture was stirred on a magnetic plate for 30 min. Then,
meat extracts were centrifuged at 10,400 x g for 10 min and the precipitate discarded.
The supernatant was adjusted to a final acetone concentration of 55 %. The solution was
stirred on a magnetic plate for 30 min at -15 0C. The solution was centrifuged for 10 min
at 10,400 x g at 4 oC. The pellet was solubilized in 30 mM MES buffer (pH 6.5,
including 30 mM (2-[N-morpholino]ethanesulfonic acid), 10 mM KOH, 0.5 mM
magnesium acetate, 0.1 mM EDTA) and dialyzed against MES buffer overnight.

Saturated ammonium sulfate solution (pH 7.0) was added dropwise to this
solution to achieve 40% saturation. The pH of the solution was adjusted to pH 5.1 with 1
M HCl to precipitate creatine kinase from the supernatant. The supernatant was stirred
for 30 min at 25 OC and centrifuged at 38,000 x g for 10 min. Saturated ammonium
sulfate solution (pH 7.0) was added to the supernatant to achieve 70 % saturation over a
30 min period. The solution was stirred for an additional 30 nrin and centrifuged at
12,000 x g for 10 min. The supernatant was collected and saturated ammonium sulfate
solution (pH 7.0) was added to 90% saturation while stirring slowly for 30 min, and then
centrifuged at 12,000 x g for 10 min. The pellet was collected and dissolved in a
minimum amount of MES buffer (pH 6.5) and dialyzed for 48 hr against MES buffer.
The buffer was replaced every 4 - 6 hr.

After dialysis, the protein solution was loaded onto a carboxylrnethyl cellulose
(CMC) ) (156-0070, Bio-Rad®, Hercules, CA) ion exchange column (100 mL of CMC

resin, 21 cm bed height and 2.5 cm diameter of column). Protein was eluted at a rate of

1'1 .1! II

82

1.0 mL/min and 1 mL fractions were collected. Adenylate kinase, enolase, and
phosphoglycerate kinase were all positively charged and were bound to the CMC column
at pH 6.5 (Scopes and Stoter, 1982). TPI was negatively charged at pH 6.5 and did not
bind to the CMC column. Also, it was reported that creatine kinase was denatured at
room temperature when eluted from the CMC column at pH 6.5 (Scopes and Stoter,
1982). TPI activity and protein concentration were determined during each Stage of the
purification process. Purity of TPI was determined by SDS-PAGE as described below.
TPI concentration was determined at 280 nm using an extinction coefficient, Em = 13.1

(Norton et al., 1970).

5.3.2 TPI Enzyme Assay

Triose phosphate isomerase activity was assayed as described by Bergmeyer
(1974) with modifications (Wang et al., 1996). The assay components of Bergmeyer
(1974) contained 2.5 mL triethanolamine buffer (0.243 mol/L; pH 7.6), 0.5 mL
glyceraldehyde-3-phosphate (3.8 mmol/L), 50 pl NADH, Na salt (10 mg/mL), 10 p1
glycerophosphate dehydrogenase (1.5 mg/mL) and 10 pl of- meat extract. The modified
enzyme assay included 1 mL trietlranolamine buffer (0.2 mol/L; pH 7.6), 0.2 mL
glyceraldehyde-3-phosphate (15 mmol/L), 10 pl NADH, Na salt (10 mg/mL), 10 pl
glycerophosphate dehydrogenase (1.5 mg/mL, 170 units/mg protein, cat. no. G67 51) and
10 pl of meat extract. The initial velocity was read at 340 nm for 1 min at 25°C.
Samples were diluted in PBS (pH 7.2) so that the optical density change was less than 0.2

units per minute (Beisenherz, 1955). TPI activity was determined using an extinction

83

coefficient ofNADH (E340 = 6.317 x 10 2 l x mol " x mm '1). One unit (U) of activity
was defined as 1 pmol substrate converted/min (Bergmeyer, 1983). TPI enzyme activity

was calculated based on the following formula:

A x 1000
Activity (UfL) = pmol x min'l x L 'l
t x s x d x (p

 

A = absorbance

t = time (min)

a = absorption coefficient (6.317), L x mmol " x mm '1
d = the distance of light path, 10 mm

0) = volume fraction of sample in assay (incubation) mixture, vN (no units)

v = volume of sample used in assay, pl

V = assay volume, pl

5.3.3 Bradford Soluble Protein Concentration Determination

Protein concentration of meat extracts was determined following the method of
Bradford (1976). Bovine serum albumin (BSA) was used to prepare a standard curve
ranging fi'om 10 to 913 pg protein/mL. Ten nricroliters of meat extract and 200 p1
Coomassie blue G-250 dye reagent (500-0006, Bio-Rad®) were added to each microwell
(lmmulon® 1, Dynatech, Chantilly, VA 22021), and incubated for 5 min at room

temperature. Absorbance was read at 590 nm using a Minireader 11 (Molecular Devices,

84

Menlo Park, CA 94025). Protein concentration was determined using the lowest dilution

in PBS (pH 7.2) which fell on the BSA standard curve.

5.3.4 Immunization and Polyclonal Antibody Production

Three white New Zealand rabbits (3 months old; code no. 56851, 56881 and
56883) were immunized subcutaneously at 10 different locations with 500 pg of purified
bovine TPI (in 0.5 mL MES buffer) in Freund's complete adjuvant (Difco®, Detroit, MI
48232). A 1.0 mL total injection volume contained 0.5 mL of TPI in MES buffer and 0.5
mL F reund's complete adj uvant. After five weeks, rabbits were boosted by subcutaneous
injection with 500 pg TPI protein emulsified with Freund's incomplete adjuvant (Difco®)
as described above. Ten days after each boost, rabbits were bled via marginal ear veins.
Total volume of blood obtained was based on the body weight of each rabbit. Four
boosts at 4, 8, 11, 18, and 23 wk were used for each rabbit. The polyclonal antibodies
were purified from sera by ammonium sulfate precipitation as described by Harlow and
Lane (1988) except that polyclonal antibodies were precipitated in a solution of 33%

saturated ammonium sulfate solution instead of 40% saturation at pH 7.2.

5.3.5 Indirect ELISA (Titer Determination)

An indirect ELISA was used to determine the titer of anti-bovine TPI polyclonal
antibodies. Titers were determined using sera collected fi'om each rabbit after each boost.
Microtiter wells (Immulon® 2, Dynatech) were coated with 100 pl bovine TPI (1 pg/mL)

in carbonate buffer (0.1 M, pH 9.6) overnight at 4 °C. Microtiter plates were washed 3

 

‘ I III-III

 

85

times with PBS-Tween (0.01 M PBS with 0.05% Tween 20, pH 7.2). Three hundred
microliters 0.5% casein (wt/vol) in PBS was added to each well as the blocking reagent to
decrease nonspecific binding and incubated for 30 min at 37 °C. Wells were washed 4
times with PBS-Tween, and 50 pl of serum diluted from 101 to 108 times was added to
each well and incubated at 37 0C for 30 min. Those polyclonal antibodies not bound to
the TPI antigen on the plates were removed by washing 4 times with PBS-Tween (pH
7.2). Then, 100 pl of goat anti-rabbit (GAR) IgG peroxidase conjugate from Sigma®
(1:500 in 0.5% casein-PBS) was added to the wells and incubated at 37 °C for 30 min.
Each plate was washed 8 times to remove unbound GAR IgG peroxidase. The bound
peroxidase was reacted with ABTS substrate (2,2’-azino-bis(3-ethylbenzthiazoline-6-
sulfonic acid)) to develop color and absorbance was read at 405 nm using a ThennoMax
(Molecular Devices, Menlo Park, CA 94025) (Pestka et al., 1982). Titers were reported
as the highest dilution that showed twice the absorbance of the same dilution of

preirnmune serum.

5.3.6 Electrophoresis

Both sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)
(Wang et al., 1992; 1995) and native PAGE (Wang et al., 1992; 1995) were performed
using a Mini-Protein Gel assembly (Bio-Rad®) with a 4 % acrylanride stacking and 12 %
acrylarrride resolving gel. For SDS-PAGE, protein extracts were combined 9 : 1 (v/v)
with sample buffer (0.5 M Tris-HCl, pH 6.8, 10% (w/v) SDS, 5% (v/v) 2-[3-

mercaptoethanol), placed in boiling water and held for 15 min. For native PAGE, the

86

protein extracts were combined with sample buffer (0.5 M Tris-HCl, pH 6.8) at a 9 : 1

(v/v) ratio, and no heat was applied. Samples were held at -10 C’C and thawed at room
temperature before SDS-PAGE or native PAGE was performed. For both SDS-PAGE
and native PAGE, 15 pl prepared meat extracts were loaded per well. Broad range
molecular weight markers (161-0318, Sigma®) were used to determine the molecular
masses of the unknown protein bands on SDS-PAGE. Purified porcine TPI (type XII, T-
7526, Sigma®) was used (4pg TPI/well) as a marker protein for both native PAGE and
SDS-PAGE. Electrophoresis was performed using Tris-glycine electrode buffer (0.025
M Tris, 0.192 M Glycine, pH 8.3) at a current of 55 milliamps and a voltage of 200 V for
45 min. The gels were stained with 0.2% Coomassie Brilliant Blue R-250 (161-0400,

Sigma®) which has a detection limit of 0.1 to 1 pg protein per band (Wilson, 1983).

5.3.7 Western Blotting

Relative antibody specificity to purified bovine TPI and meat extracts were
evaluated by the western blot method (Wang et al., 1993). Purified bovine TPI, raw meat
extracts, heated meat extracts, and commercially available TPI from different species,
(rabbit (type IIIS), porcine (type x11), and dog (type v1) TPI purchased from Sigma°)
were electrophoretically transferred (1 hr at 100 V) from either native or SDS-PAGE gels'
to nitrocellulose membranes. The membranes were washed thoroughly with PB S-Tween
(Tween 0.05% (v/v)). Ten milliliters of 0.5% casein-PBS solution was used as a blocking
agent for 30 min at room temperature. The membranes were washed with PBS-Tween

again, then 10 mL of bovine TPI PAb (diluted 1:1,000 in 0.5% casein-PBS) was added

87

and incubated at room temperature for another 10 min. Unbound antibodies were
removed by washing with PBS-Tween and 10 mL of goat anti-rabbit IgG (GAR-IgG)
peroxidase conjugate (1 :2,000 in 0.5% casein-PBS) was added to the membrane,
followed by incubation for 10 min at ambient temperature. Several washes with PB S-
Tween were used to remove unbound GAR-IgG peroxidase conjugate. To prepare the
color developing substrate, 24 mg of 3, 3’, 5, 5’ tetrarnethyl benzidine (T-2885, Sigma®)
and 80 mg of dioctyl sulfosuccinate (D-0885, Sigma®) were dissolved in 10 mL Of
ethanol. Then, the above solution was added to 30 mL of 0.1 M citrate-phosphate buffer
(pH 5.0) and 20 pl of 30% H202 was added. This substrate was used to determine bound
peroxidase (Wang et al., 1992). Twenty milliliters of this substrate solution was used to
stain one nitrocellulose membrane. The staining reaction was stopped using double

distilled water when sufficient color was developed.

5.3.8 Biotinylation of Polyclonal Antibodies

Polyclonal antibodies were purified using a protein A/G column following
instructions provided by the manufacturer (Pierce, Rockford, IL 61105). Polyclonal
antibodies were biotinylated as described by Harlow and Lane (1988). The IgG
concentration was determined at 280 nm using an extinction coefficient (E mymL = 1.4).
Three milliliters of antibody solution (1 mg/mL) was diluted in 0.1 M sodium borate
buffer (pH 8.8). The antibody solution (1 mg/mL) was concentrated using a Centriprep
membrane (membrane size of 10 kd; Amicon, Inc., Beverly, MA 01915). Five mg of

biotin-amidocaproate N-hydroxysuccinimide ester was dissolved in 1 mL dimethyl

88

sulfoxide. This biotin ester solution (0.1 mg of biotin ester per gram of antibody) was
added slowly to the antibody solution, followed by stirring for 4 hr at 25 OC. Twenty
microliters of 1 M NH4C1 solution (based on 0.08 pl of 1 M NH4C1 per pg of biotin
ester) was added and the solution incubated for 10 min at 25 °C. The antibody solution
was dialyzed against PBS (0.01 M, pH 7.2) for 48 hr. The buffer was changed every 6 hr.

After dialysis, biotinylated antibodies were aliquoted and stored at -18 °C.

5.3.9 Sandwich ELISA
The sandwich ELISA was performed by coating microtiter wells (Immulon® 2,

Dynatech) with 100 pl of polyclonal antibody #56851 diluted (1:500) in 0.1M carbonate

buffer (pH 9.6) and drying overnight at 37 0C (98.6 °F) in an oven. Wells were washed 4
times with PBS containing 0.05% Tween 20 (PBS-Tween), and 300 p1 of 0.5% casein-
PBS was added to each well to block remaining protein binding sites and incubated for 30
min at 37 °C. After washing 4 times with PBS-Tween, meat extracts or standard bovine
TPI diluted in 0.5% casin-PBS (50 pl) were added to each well and incubated for 30 min
at 37 °C. Plates were washed another 4 times with PBS-Tween, and 50 pl of biotinylated
polyclonal antibodies diluted (1:100) in 0.5% casin PBS was added. After incubation for
30 min at 37 °C and washing 4 times with PBS-Tween, 100 pl avidin-horseradish
peroxidase (HRP) from Sigma® diluted 1:8000 in PBS-casein was added to each well and
incubated for 30 min. Plates were then washed 8 times with PB S-Tween. Bound
peroxidase activity was determined with ABTS substrate (2,2’-azino-bis(3-

ethylbenzthiazoline-6-sulfonic acid)) (Pestka et al., 1982) and absorbance was read at 405

89

nm using a ThennoMax (Molecular Devices, Menlo Park, CA 94025). Bovine TPI
ranging fi'om 0 to 40 pg TPI/mL was used to construct the standard curve. Results were
expressed as pg TPI/mL buffer solution and also expressed as pg TPI/kg meat. The
coefficient of variation was used to evaluate between-run and within-run precision. The
cofiicient of variation (Deshpande, 1996) for between-run precision was based on 8
replicates using 5, 10, 20, and 40 pg TPI/mL. The cofficient of variation of within-run

precision was based on 16 replicates at three TPI concentrations (16, 18, and 20 pg/mL).

5.3.10 Cross Reactivities of TPI Antibodies with TPI from Different Animal
Species, Commercial Whey Protein Concentrates and Plasma Protein Concentrates

Rabbit (type 1118, T-2391), porcine (type XII, T-7526) and dog (type VI, T-6635)
TPI were purchased from Sigma® (muscle types were not specified). Protein
concentration of TPI protein fiom each species or meat extracts was determined using
Bradford method (1976). Cross-reactivities of bovine TPI polyclonal antibodies with
rabbit, porcine, and dog TPI were examined at TPI concentrations ranging from 0.001 to
100 pg TPI protein/mL using the sandwich ELISA. The cross reactivity of a raw extract
of ground beef extracted with PBS (0.1 M NaCl; pH 7.2) was also examined.

Five spray dried protein concentrates of Beef Stock, Pork Stock, Chicken Broth,
AMP 800® (whey protein concentrate), and AMP 600N® (hydrolyzed protein from beef
and plasma) were donated by AMPC, Inc. (Ames, IA 50010). Whey protein isolate (type
A) was obtained from Davisco International® (Le Sueur, W 5605 8). Protein

concentration of each protein concentrate solubilized in PBS (0.1 M NaCl; pH 7 .2) was

90

determined following the method of Bradford (1976). Cross reactivities against bovine
TPI polyclonal antibodies were examined by sandwich ELISA at total soluble protein

concentrations ranging from 0.24 to 250 pg/mL.

5.3.11 Cross Reactivity of TPI from Starter Cultures

Three common meat starter cultures, Lactobacillus plantarum, Pediococcus
pentosaceus, and Pediococcus acidilactici were donated by ABC Research Corporation
(Gainesville, FL 32602). Starter cultures were inoculated separately in a medium of 80%
lactose broth and 20% APT (all purpose tween) agar (for cultivating heteroferrnentative

lactobacilli (Difco®) and incubated for about 72 hr at 37 °C. Total cell counts reached

about 1012 CPU per gram for each strain. Cells were harvested (about 1 g) after
centrifugation of the bacterial suspension at 1,000 x g for 5 min. Ten volumes (about 10

mL) of cold peptone buffer (0.1%) was added to the pellet and mixed thoroughly. Then,
cold acetone (-18 °C, 250 mL) was added slowly to the bacterial suspension, to avoid

protein denaturation, and stirred on a magnetic plate at -18 0C for 10 min. The bacterial
solution was then centrifuged at 5,000 x g for 10 min. Bacterial masses were precipitated
in the pellet. The pellet was vacuum filtered through a 0.2 pm membrane to obtain the
mass. A small amount of ether was added to the filter membrane to remove residual
moisture from the pellet. The bacterial pellet was lyophilized for storage. Total cell
counts were about 5 x 10" cells/2 mL of PBS (0.01 M, pH 7.2). The TPI enzyme activity
and soluble protein concentration were determined using TPI enzyme assay described in

5.3.2 and Bradford method (1976), respectively.

91

5.3.12 Ground Beef Water Bath Model Study

Protein, fat and moisture contents of semimembranosus muscle was determined
using AOAC (1990) standard methods 981.10, 960.39 and 950.468, respectively. The
pH was determined by homogenizing 10 g ground beef with 90 mL of double distilled
water using a WaringTM blender for 30 5.

Fresh semimembranosus meat (top round choice roast), purchased from a local
store, was cut into 1 cm cubes and ground twice through a 3.175 mm diameter grinder
plate in a Hobart grinder (Model 8418D, Hobart Mfg. Co., Troy, OH, 45374). The
ground beef was placed into a 60 mL syringe (Becton Dickinson and Co., Franklin Lakes,
NJ 07417). Two grams of ground meat was compressed through a 7 cm plastic tubing
into a 10 x 75 mm TDT borosilicate glass tubes (60825-538, Kimax® 51, VWR, West
Chester, PA 19380). Glass tubes were then sealed using telfon tape. Meat was cooked to
target endpoint temperatures in a Polystat circulator water bath (Model 1268-52, Cole-
Parmer) set at 82.2 °C. The water bath was connected to a digital programmer (Model
1268-62, Cole-Parmer). A Resistance Temperature Detector thermocouple (platinum Pt
100 temperature probe, Solomat Partners LP, Stanford, CT 06906) was inserted into the
center of a TDT tube containing 2 g meat. Internal temperatures were determined by a
thermocouple using a Solomat MPM 200 Modumeter (Solomat Partners LP) inserted into
the meat. After tubes reached the target temperatures, 48.9 °C (120 °F), 54.4 0C (130 °F),
60.0 °C (140 °F), 65.6 °C (150 °F), 71.1 0C (160 °F), and 76.7 °C (170 °F), meat was

placed in a ice bath for 10 min.

92

5.3.13 Preparation of Protein Extracts and Enzyme Assays

Glass tubes (10 x 75 mm) were broken and cooked meat was transferred into
scintillation vials. Six milliliters 0.01 M, 0.05 M, 0.1 M PBS, pH 7.2 containing 0.1 M
NaCl or 0.01 M MOPS, pH 7.2, 0.1 M NaCl, was added to each vial. Samples were
vortexed for 1 min, then stirred on magnetic plates for 30 min at 4 oC. Samples were
centrifuged at 4,500 x g for 10 min at 4 OC (Micro-centrifuge, Model 59A, Fischer
Scientific). The supernatant was collected and held at 4 0C until used. TPI activity and
concentration of extracts were determined within 24 hr. Preliminary experiments

performed to devise this extraction protocol are described in the Appendix section.

5.3.14 Statistical Analysis

Ground semimembranosus muscles were heated in water bath to six different
cooking temperatures. Each cooking schedule was conducted in triplicate. Data were
analyzed using two way AN OVA (analysis of variance) (treatment x replication). The
linear or curvilinear change in enzyme activity or ELISA with temperature was
determined using orthogonal polynomial contrast test by SAS software (SAS Institute,

Inc., Version 6.1, 1995, Cary, NC 27513).

5.4 RESULTS AND DISCUSSION
5.4.1 Purification of TPI
TPI was purified from bovine semimembranosus muscle using 40% and 55%

acetone fractionations, 40%, 70% and 90% ammonium sulfate fractionations and

93

carboxylrnethyl cellulose (CMC) chromatography. Total protein, total activity, specific
activity, yield, and purification factor of selected fiactions during purification were
reported in Table 5.1. Selected purification fractions can be observed on a representative
SDS-PAGE electrOphoretogram (Figure 5.1). Two major contaminant proteins, creatine
kinase and bovine serum albumin (BSA), were eliminated using specific procedures
described in the following section.

Creatine kinase was the major contaminant during TPI purification since creatine
kinase and TPI were both negatively charged at pH 6.5 and were not separated by CMC
chromatography. Adjusting the pH to 5.1 at 40% ammonium sulfate saturation was
necessary to precipitate creatine kinase from the supernatant. About 90% of the
contaminating creatine kinase precipitated at this stage. The TPI specific activity
increased 111.8 fold after CMC column chromatography. A single band in lane 6 (Figure
5.1) is the purified bovine TPI with a molecular mass of 23 kd, and TPI existed as dimer
forms. In preliminary experiments, TPI and creatine kinase co-eluted when
chromatographed using size exclusion chromatography (50-100 kd size exclusion beads)
(151-1030, Bio-Gel-lSmL-A, Bio-Rad®). Less than 50% of TPI and CK were separated
using the size exclusion column since the shapes of the two proteins were similar, even
though the molecular masses of TPI and creatine kinase were different (46 and 86 kd,
respectively).

BSA was another major contaminant during TPI purification. BSA was

eliminated using a series of procedures. The concentration of BSA was gradually

94

Table 5.1 Purification of triose phosphate isomerase from bovine semimembranosus

 

 

muscle
Protein Volume Specific Purification
Fraction Concentration Activity Activity Factor
(g/L) (U/L) (U/g)
Homogenate 4.91 102.7 20.9 1
55% acetone 9.25 280.7 30.4 1.5
precipitate
90% ammonium 8.47 9595.3 1133.0 54.2
sulfate precipitate
CMC column a 0.98 2290.4 2337.7 111.8

 

" Carboxylmethyl cellulose (CMC) chromatography column.

b Estimated volumes for homogenate, 55% acetone precipitate (solubilized in minimum

amount of 30 mM MES buffer and dialyzed), 90% ammonium sulfate precipitate

(dissolved in minimum amount of MES buffer and dialyzed) and the solution before

loaded into CMC column were 750, 30, 25, and 8 mL.

a”, H
kDa -"
66 - .-
45 m ‘
36 -—
29 - -tt._!i
24 u...- . .8.

.- t

14 I... --‘-.

l 2 3 4 5 6

Figure 5.1 Representative sodium dodecyl sulfate-polyacrylamide gel
electrophoretogram of muscle extracts from the triose phosphate isomerase (TPI)
purification procedures. Proteins were stained with Coomassie Blue. (lane 1) molecular
weight standard; (lane 2) porcine muscle TPI (from Sigma); (lane 3) bovine muscle
homogenate; (lane 4) 55% acetone precipitate fraction; (lane 5) 90% ammonium sulfate
precipitate fi'action; (lane 6) after carboxylrnethyl cellulose (CMC) column

chromatography.

96

decreased in 55% acetone pellet (Figure 5.1; lane 4), 70% ammonium sulfate supernatant

and 90% ammonium sulfate pellet (Figure 5.1; lane 5) fractions.

5.4.2 Production of Polyclonal Antibody and ELISA Development

Anti-bovine TPI polyclonal antibodies were produced in rabbits to devise a TPI
sandwich ELISA. Antibody titers to bovine TPI reached 107.108 in all three rabbits after
the first boost (Table 5.2). Antibody titers decreased to 106-107 after the second boost
and increased to 108 after the third boost. However, after 18 weeks, titers decreased to
104 which might be due to the irnmuno-tolerance built up against TPI (Kuby, 1994).
Serum of rabbit C was used for further studies since it had the highest titer afier three
injections.
A sandwich ELISA was developed and optimized. Purified bovine TPI from
semimembranosus muscle was used to prepare a standard curve from 5 to 40 pg TPI/mL.
The limit of detection of the ELISA was 5 pg TPI/mL. The coefficients of variation
between-runs and within-runs were determined for the ELISA. In the between-run test,
TPI concentrations of 5, 10, 20, and 40 pg/mL had coefficients of variation of 8.42, 9.42,
12.77, and 6.39 %, respectively. At TPI concentrations of 5, 10, 20, and 40 pg/mL, the
respective coefficients of variation were 8.06, 8.35, 3.62, and 6.94 % when precision was

tested within a run.

97

Table 5.2 Polyclonal antibody titers a (serum dilution) against bovine triose phosphate

isomerase from semimembranosus muscle

 

 

 

Antibody titer
Weeks after initial immunization b

Rabbit A Rabbit B Rabbit C

4 7.5x107 3.2x 107 1.0x10"

8 9.0x 10° 8.3x 106 9.1 x107

11 1.0x108 1.1x108 1.1x108

18 1.0x104 1.0x10“ 1.0x104

23 9.4 x 104 9.0 x 104 9.0 x 104

 

3' Titer is defined as the serum dilution at which the absorbance is twice that of the pre-

irnmune serum.

b Booster injections were given at 4, 8, 11, 18 and 23 wk. Sera were analyzed by indirect

enzyme-linked immunosorbent assay.

98

5.4.3 Cross Reactivity of TPI Polyclonal Antibodies with TPI from Different
Species

A sandwich ELISA was used to investigate the cross reactivity of the antibodies
with TPI from the muscles of different animal species (Figure 5.2). Bovine TPI had
highest cross reactivity with bovine anti-TPI polyclonal antibodies when compared to TPI
from other animal species. Bovine TPI was about 0.3 absorbance units higher than
porcine TPI at 25 pg TPI/mL. Porcine TPI had higher cross reactivity when compared to
rabbit and dog TPI ranging from 10 to 25 pg TPI/mL; rabbit TPI and dog TPI had similar
cross reactivities with bovine anti-TPI polyclonal antibodies.

Bovine anti-TPI polyclonal antibodies were highly specific to native and
denatured TPI as shown on western blots (Figures 5.3 and 5.4) of raw bovine muscle
extracts. Only one hybridized band was observed on the membranes, although many
muscle proteins are observed on native and SDS-PAGE gels (Figure 5.1, lane 3). The
substrate intensity increased as volume of meat extracts was increased on the native and
SDS-PAGE gels (Figure 5.3, lanes 5 and 6; Figure 5.4, lanes 6 to 9). Cross reactivity of
raw beef extracts with TPI polyclonal antibodies was also examined using TPI sandwich
ELISA (Figure 5.5). A bovine raw extract of ground beef extracted with PBS (pH 7.2)
cross reacted with TPI polyclonal antibodies from 15 to 250 pg total soluble protein/mL
meat extract.

Western blotting was also conducted to determine the specificity of bovine anti-

TPI polyclonal antibodies with native and denatured TPI from different animal species.

99

 

 

 

 

 

 

 

 

 

 

 

 

1.4
1.3
[+llovinc‘l'l’l vl-l’urcinc'l'l’l e-Rnhhit '1'1’1 alkg'l'l’l]
1.2
1.1
I 1

E 0.9 '
O
:5
3 0.8
t:
.8
3 0.7 .
3 i]
<1:

0.6

0.5 1'

-- fl

0.4

0.3 ‘1’ I

0.2

.1 ' 10 20 30

TPI concentration (ug/mL)

Figure 5.2 Cross-reactivity of bovine anti-triose phosphate isomerase (TPI) polyclonal
antibodies with TPI from different animal species determined by sandwich ELISA. The
standard deviation values that were less than 0.02 absorbance units were not plotted on

the figure.

100

 

Figure 5.3 Western blot of bovine, dog, rabbit, and porcine muscle triose phosphate
isomerase (TPI) and bovine muscle extract electrophoretically transferred from a native
polyacrylanride gel to a nitrocelluolse membrane hybridized with anti-bovine TPI
polyclonal antibodies; (lane 1; 5 pg protein) dog muscle TPI; (lane 2; 5 pg protein) rabbit
muscle TPI; (lane 3; 5 pg protein) porcine muscle TPI; (lane 4; 5 pg protein) bovine
muscle TPI; (lane 5; 100 pg protein and lane 6; 200 pg protein) bovine raw muscle
extracts using MEMT buffer (50 mM Tris-HCl, pH 6.8 containing 0.4 mM EDTA, 7 mM

or-mercaptoethanol and 5 mM MgSO,).

101

 

Figure 5.4 Western blot of bovine, dog, rabbit, and porcine muscle triose phosphate

isomerase (TPI) electrophoretically transferred from a sodium dodecyl sulfate-
polyacrylamide gel to a nitrocellulose membrane hybridized with anti-bovine TPI
polyclonal antibodies; (lane 1; 100 pg protein; lane 2; 50 pg protein) raw bovine muscle
extracts using phosphate buffer saline (0.14 M NaCl, 0.01 Na phosphate, pH 7.2 ); (lane
3; 50 pg protein) raw bovine muscle extracts using MEMT buffer (50 mM Tris-HCI, pH
6.8, containing 0.4 mM EDTA, 7 mM or-mercaptoethanol and 5 mM MgSO,); (lane 4; 5
pg protein) bovine muscle TPI; (lane 5; 5 pg protein) porcine muscle TPI; (lane 6; 5 pg

protein) rabbit muscle TPI; (lane 7; 5 pg protein) dog muscle TPI.

 

 

 

 

 

 

 

 

 

1.2
-- Bovine TPI
1-1 +WP1
1 +AMP600N
E *AMP800
g 0-9 All-Beef Stock
3; 0 8 *Pork Stock .
§ . s-Chicken Stock 1 / g
E 0.7 e-Bovine Raw Extract °/“ "'5
it” A -
0.6 “I. 2.4.! ‘ T7
0.5 A. ._ .A_§_ V
0.4 4.... . -.. 2
0.1 1 10 100 1000

Protein concentration (ug/mL)

Figure 5.5 Cross-reactivity of bovine triose phosphate isomerase (TPI) polyclonal
antibodies with TPI from different protein concentrates determined by sandwich ELISA.
Bovine TPI = TPI purified from bovine semimembranosus muscle, WPI = whey protein
isolate, AMP600N = hydrolyzed protein from meat and plasma, AMP800 = whey protein
concentrate, bovine raw extract = ground bovine semimembranosus muscle extracted

with phosphate buffer saline (1:3). The standard deviations less than 0.02 were not

plotted on the figure.

103

Bovine anti-TPI polyclonal antibodies cross reacted with TPI from bovine, porcine, rabbit
and dog muscles electrophoretically transferred from native gels (Figure 5.3). Single
bands were observed on nitrocellulose membranes containing native TPI from different
species when reacted with TPI polyclonal antibodies. Western blotting revealed that
bovine, porcine and rabbit TPI had higher cross reactivities with anti-TPI polyclonal
antibodies than dog TPI. The results were consistent with the results of the ELISA that
showed bovine, porcine and rabbit TPI had higher cross reactivities with anti-TPI
polyclonal antibodies when compared to dog TPI at 25 pg/mL.

Single bands were also observed when denatured TPI from rabbit, porcine and
dog species reacted with anti-TPI polyclonal antibodies transferred from SDS-PAGE
(denatured) gels (Figure 5.4). Bovine TPI (lane 4) showed higher substrate intensity
compared to that of porcine (lane 3), rabbit (lane 2), and dog TPI (lane 1) when 5 pg TPI

was loaded in each lane.

5.4.4 Cross Reactivity of TPI Polyclonal Antibodies with TPI from Different
Sources

Some protein concentrates are widely used in commercially processed roast beef
products as functional ingredients to improve sensory properties and yields. The study
was designed to examine the cross reactivity of six commercial protein concentrates with
bovine anti-TPI polyclonal antibodies. In general, all commercial protein concentrates
had lower cross reactivities with bovine TPI polyclonal antibodies when compared to

pure bovine TPI (Figure 5.5). AMP 600N® (hydrolyzed protein fi'om beef and plasma)

104

and AMP 800® (whey protein concentrate) had higher cross reactivities with bovine anti-
TPI polyclonal antibodies compared to whey protein isolate, beef stock, chicken stock,
and pork stock at the same concentration.

AMP 600N® showed the highest cross reactivity when compared to the other
protein concentrates (Figure 5.5). AMP 600N® is a hydrolyzed meat protein product (70
% protein content) containing hydrolyzed beef plasma. Beef plasma probably contains
some bovine TPI which may have caused the cross reaction (Sawyer et al., 1972).

No cross reactivity was observed with whey protein isolate (95% protein) at
concentrations up to 250 pg protein/mL, suggesting that whey proteins did not cross react
with TPI polyclonal antibodies. Although AMP 800® is a whey protein concentrate (80
% protein content), a greater cross reactivity was observed compared to whey protein
isolate in the same concentration range (15 to 250 pg protein/mL). Unknown ingredients
in AMP 800®, such as contaminating plasma proteins, could lead to higher TPI cross
reactivities. Plasma proteins were observed when AMP 600N was separated by SDS-
PAGE (data not shown). Since ingredients in AMP 600N® and AMP 800® cross-reacted
with TPI polyclonal antibodies, the use of these concentrates above concentration of 15

pg protein/mL could interfere with the accurate determination of processing adequacy.

5.4.5 Cross Reactivity of TPI Polyclonal Antibodies with TPI from Starter
Cultures
Fermented meat products are produced using bacterial starter cultures to provide

their characteristic sensory properties (Bacus, 1984). Lactobacillus plantarum,

105

Pediococcus pentosaceus, and Pediococcus acidilactici are common commercial meat
starter cultures (Bacus, 1984). The cross reactivities of microbial TPI from meat starter
cultures with TPI polyclonal antibodies was examined to determine if the TPI sandwich
ELISA could be applied to verify the processing adequacy in the fermented meat
products. TPI enzyme activity was low and averaged 5.69, 3.90, and 5.20 U/L (0.01 M,
pH 7 .2) in assay mixtures containing 2.5 x 1011 cell/mL of Lactobacillus plantarum,
Pediococcus pentosaceus, and Pediococcus acidilactici, respectively. Microbial TPI was
not detected in the three strains using the TPI sandwich ELISA, suggesting that TPI
antibodies did not cross react with the microbial TPI produced from the starter cultures

commonly used in fermented meat products.

5.4.6 Ground Beef Water Bath Model Study

TPI activities and concentrations decreased consistently (P < 0.0001) as cooking
temperatures of ground beef were increased when determined by enzyme assay and
ELISA, respectively (Table 5.3). However, the difference in TPI activity was more
apparent than the difference in TPI concentration among different endpoint temperatures.
For example, TPI activity was 537.53 U/kg meat at 48.9 0C and decreased to 186.96 U/kg
meat at 54.4 °C. TPI concentration was 86.30 pg TPI/g meat at 48.9 °C and was
decreased to 66.88 pg TPI/g meat at 54.4 °C. The differences were 350 U/kg meat and
19,000 ng TPI/g meat between 48.9 and 54.4 °C using enzyme assay and sandwich

ELISA, respectively. The coefficient of variations of TPI activity and concentration were

106

Table 5.3 Effect of cooking temperature on triose phosphate isomerase (TPI) activity

(U Ikg meat) and concentration (pg TPI/ g meat) of bovine meat cooked in a water bath

 

 

Internal temperature Muscle Activity Concentration
(° C) (U/kg meat) (pg TPI/g meat)

48.9 i 0.1 537.5 i 14.55 a ( 2.71) 86.3 1- 11.55 a (13.38)
54.4 i 0.1 187.0 i 21.88 (11.70) 66.9 i 6.64 ( 9.93)
60.0 i 0.1 72.1 i 9.03 (12.53) 48.3 i 6.67 (13.82)
65.6 i 0.1 38.0: 3.08 ( 8.12) 42.6 i 8.75 (20.56)
71.1 i 0.1 27.4 i- 3.98 (14.54) 36.8 i 7.17 (19.46)
76.7 i 0.1 19.4 i 1.01 ( 5.21) 34.1 :1: 6.29 (18.46)

 

a Values are expressed as mean i standard deviation of triplicate determinations and
values in parentheses represent coefficient of variation. Means in the same column

decreased linearly as temperature increased (p < 0.0001).

107

also used to compare method variability. In general, the coefficient of variation of TPI
activity was smaller (less than 15%) when compared to that of TPI concentration.

Both SDS-PAGE and Western blotting were also used to determine the effect of
heating temperature on TPI extracted from the ground beef model system. On SDS-
PAGE, the intensity of Coomassie Blue dye representing TPI bands of the cooked meat
extracts decreased as cooking temperatures increased when the same volume of meat
extract was loaded on each lane (Figure 5.6, lanes 3-8). TPI concentration was similar in
extracts of ground beef heated to 48.9 0C (120 °F) and 54.4 °C (130 °F) (lanes 3 and 4).
TPI concentration decreased in extracts of ground beef heated to 60.0 °C (140 °F) (lane
5), whereas TPI was not observed at 71.1 and 76.7 °C (170 oF) (lanes 7 and 8).

A western blot was also used to examine TPI concentration using TPI polyclonal
antibodies (Figure 5.7) from extracts of cooked ground beef. TPI concentration
decreased as heating temperature was increased in the ground beef model system when
the same volume of each meat extract was loaded in each lane, confirming the results
from SDS-PAGE (Figure 5.6). Some non-specific recognition was observed between TPI
polyclonal antibodies and raw extract proteins (lane 2). A higher amount of soluble
protein was present in raw bovine extracts when compared to the same volume of heated

extracts causing non-specific protein binding.

5.5 CONCLUSIONS
TPI could be a promising marker protein since the activity and concentration of

TPI decreased as cooking temperature was increased in the ground beef model system.

108

kDa
203
118

51.6

34.1

29

19.2

 

Figure 5.6 Representative sodium dodecyl sulfate-polyacrylamide gel electroetpgrarn of
muscle extracts of bovine semimembranosus meat heated to different end-point
temperatures. Proteins were stained with Coomassie Blue. (lane 1) molecular weight
standards; (lane 2) unheated bovine muscle extracts; (lane 3) 48.9 °C; (lane 4) 54.4 °C;
(lane 5) 60.0 °C; (lane 6) 65.6 °C; (lane 7) 71.1 °C; (lane 8) 76.7 °C; (lane 9) purified

bovine TPI.

109

 

Figure 5.7 Western blot of bovine muscle extract with anti-triose phosphate isomerase
polyclonal antibodies electrophoretically transferred from a sodium dodecyl sulfate-
polyacrylamide gel to a nitrocellulose membrane. Bovine semimembranosus muscle was
heated to different end-point temperature. (lane 1) bovine TPI; (lane 2) unheated bovine
muscle extract; (lane 3) 48.9 °C; (lane 4) 54.4 °C; (lane 5) 60.0 °C; (lane 6) 65.6 °C; (lane

7) 71.1 °C; (lane 8) 76.7 °C.

110

Further study is needed to investigate the applicability of using TPI enzyme assay and

sandwich ELISA to verify processing adequacy in beef roasts.

CHAPTER 6
VERIFICATION OF TRIOSE PHOSPHATE ISOMERASE ENZYME-LINKED
IMMUNOSORBENT ASSAY TO DETERMINE PROCESSING

ADEQUACY OF ROAST BEEF IN A PILOT STUDY

6.1 ABSTRACT

Triose phosphate isomerase was identified as a potential time-temperature integrator
(TTI) to verify processing adequacy of roast beef in previous ground beef model and
roast beef pilot studies. The objective of this study was to use a TPI sandwich ELISA
and enzyme assay to verify processing adequacy of beef roasts processed in a commercial
pilot plant. Adequate processing schedules were used to process roasts at low, medium
and high temperatures [54.4 0C (130 °F)/121 min, 58.3 0C (137 °F)/24 min, and 62.2 °C
(144 oF)/5 min] based on USDA schedules. Two inadequate processing schedules were
used at each temperature by reducing the holding time by 0.5 and 1.0 log from the
adequate processing schedules. Residual TPI was quantified in extracts from each roast
using a sandwich ELISA and enzyme assay. TPI concentration and activity did not differ
between adequately and inadequately processed roasts. It was difficult to consistently
control processing conditions in the pilot plant due to large variations caused by

smokehouse processing facilities, size and shape of roasts, types of muscles and locations

111

112

of roasts placed in the smokehouse. Future research is needed to determine if TPI can be
used as a TTT under a commercial processing conditions.

KEYWORDS - ELISA, roast beef processing, TPI

6.2 INTRODUCTION

Adequate cooking is the easiest way to eliminate pathogenic bacteria and ensure
the safety of meat products. Sixteen time-temperature processing schedules are allowed
for cooked beef and rare roast beef (U SDA-F SIS, 1995). In May 1996, the USDA-FSIS
proposed several rules to regulate current federal meat and poultry inspection regulations
in order to establish “safety margins” for thermally processed meat and poultry products
(U SDA-FSIS, 1996b). A thermal process sufficient to cause a 7-D reduction in
Salmonella was proposed by the USDA-F SIS (1996b) as the lethality performance
standard for roast beef, cooked beef and cooked corned beef products. Salmonella was
selected as the indicator microorganism since it is more thermally resistant than E. coli
01 57:H7. The proposed regulation allows processors to use more sophisticated thermal
treatments and more flexible time-temperature processing conditions.

A time-temperature indicator (TTT) is “a small measuring device that shows a
time-temperature dependent, easily, accurately and precisely measurable irreversible
change that mimics the changes of a target attribute undergoing the same variable
temperature exposure” (Hendrickx et al., 1995). A TTT could be used by USDA and

processors to verify processing adequacy of meat products.

113

Triose phosphate isomerase (TPI) was identified as a potential endogenous time-
temperature integrator (TTT) in beef by Orta-Rarnirez et al. (1997). TPI has a similar 2
value to that of Salmonella suggesting similar thermal inactivation kinetics (Orta-
Ramirez et al., 1997). Hsu in a previous Chapter identified TPI as the best endogenous
TIT to verify compliance to the USDA processing schedules in both ground beef and
roast beef using medium (58.3 0C) to high (62.2 °C) temperature processes (chapter 4). In
a ground beef water bath model system, TPI activity averaged 2.6 U/kg in adequately
processed ground beef and increased to 4.9 and 13.3 U/kg when inadequately processed
by reducing the holding time 0.5 log and1.0 log, respectively. In a MSU smokehouse
pilot study, TPI activity in adequately processed roast beef 1.6 U/kg and increased to 3.7
and 7.8 U/kg when inadequately processed by reducing the holding time by 0.5 and 1.0
log at high temperature process (62.2 °C), respectively.

A sandwich ELI SA was developed to quantify TPI in cooked beef (Hsu, Chapter
5). We found that both TPI activity and concentration in ground beef decreased (P <
0.0001) as temperature was increased, suggesting that both the TPI sandwich ELISA and
enzyme assay were able to detect differences in residual TPI due to cooking temperature.

Two critical requirements must be met for TPI to function as a TTT: (1) TPI
activity or concentration within equivalent adequately or inadequately time-temperature
schedules should be similar, and (2) TPI activity or concentration should decrease as
holding time is increased at the same temperature. Therefore, a pilot study was needed to
further validate the applicability of the TPI sandwich ELISA for verifying processing

adequacy of roast beef. Our objectives were to validate the applicability of TPI as an

114

endogenous TTI for verifying processing adequacy of roast beef. TPI activity and
concentration of adequately and inadequately processed roast beef were measured by
enzyme assay and ELISA to validate compliance to the USDA roast beef processing

schedules.

6.3 MATERIALS AND METHODS
6.3.1 Processing Schedules

Choice top round roasts (semimembranosus, SM) donated by Bil-Mar Foods
(Zeeland, MI) were trimmed of external fat and connective tissue. Each roast was evenly
divided into two pieces (2.268 kg/each piece), cut into a rectangular shape (about 40 cm x
25 cm) and vacuumed packaged in Cryovac packaging bags (W.R. Grace Co., Duncan,
SC, 29681). Roasts were stored at 4 0C for less than 4 hr before processing. The pilot
study was conducted at Bil-Mar Foods (Zeeland, MI) using a H. Maurer and Sohne
smokehouse (Model # ASL-3611, Ranch und Warrnetechnik GmbH and Co. KG, D7752
Reichenau, Germany). Adequate processing schedules included low, medium and high
temperature processes [130 °F (54.4 oC)/121 min, 137 °F (58.3 °C)/24 nrin, 144 °F (62.2
oC)/5 min] (Table 6.1). Inadequate processing schedules were designed using a 0.5 log
and 1.0 log cycle reduction in holding time from that of the adequate processing
schedules. The wet and dry bulb processing conditions in the smokehouse are listed in
Tables 6.2, 6.3, and 6.4. Roasts were processed according to these schedules. After

processing, roasts were chilled in an ice-slush bath and stored at 4 °C. Roasts were

115

A32 .99“).de 838:8 9.6888 :8: ~82 «.me 05 :e 883 22> 83:88 x08 28:834. a

58 5:6 3: Op 4.:

ES ovkmo om: Uo Alum

EE Sac 2: Oc 4.4m

:8 gap 2: Op 2%

88 who—o RC 0o m.wm

EE asap 2: Op new

as Eu. 3.: Oc Se . e368 banana...
A88: 8 .8882 w£ We:

88 WQEO me: Do :0 888.88:
38: 8 82262 w£ o. 3

88 m. :Ce 9.3 Do Yam poxeeoLoncb

 

882: 238382 Be:

882: 238382 8882

888 2382.82 :wE

 

382 :89 8.6.8er 328388 pen >288?“ 232: 3 tom: 88:80:82 :8 88: @6885 me 2an

116

3:5 Hows“ v2.82 55 mm 33an 203 380% a

 

Co cm: 9 3% Go as 0.. 52 E. 3.: no 3% use 3 m
:0 mm: 0.. Sn as 23 0.. 3.: Go 2.: 0.. 3% as 8 v
E0 8: oo :8 £0 83 0.. 3: Co 2: 9. 3m 58 Q m
£0 em: 0.. v.8 £0 08 0.. 2S E. 8: 0.. 3: ea 8 N
or E: 00 35 CO o5 0° 3: Ba 8: 0.. in as n _
Gov 2380;an £3 83 6% 05389:“: £3 bQ 838388 3585 3:5 2:5. owfim

 

been “much wommoooa Cc o2 do $va 238382 32 832m 8 com: 2528 omsosoxofim N6 033.

117

.38: Hows“ 62.88 >2: mm “.3082 203 398% .w

 

 

Go 8: Do is Go $8 9. 3: EC 2: 0.. 3% “as mu m

£0 8 a 0.. 0.8 cc SS 00 Be QC 8: p. 33 as 8 N

go on: 0o 53. Ea 2.3 Do Qum— Eo co: 0.. wfim 58 mm _
Gav 2382.82 £3 63 Gov BEEKEB £3 b9 2382.82 BEBE €25 082‘ owfim

 

Moon ammo“ 3389a Go 52 ”Dc 0me 228382 8338 23th 3 com: 233:3 omsozoxofim m6 Bosh

118

.88: 638 3:88 >05 mm @3082 203 393% a

 

 

£0 em: 0,. v.8 Ea ONNV 0° 32 go 8: p. 98 «as 3 N

:0 no: 09 E Co mNNv 0.. NE. GO ON: 09 33 £6 8 N

Ce 8: 0.. N.NN Ea SNV 0° 3: £0 8: 9. SN 55 8 ~
Gov 238882 £3 83 CL 258388 £3 DD 23.2%an 12:85 3:5 25H owfim

 

moon ammo“ £8885 Go 3: moo NNB 8388—82 :3: 8805 3 new: 2.60:8 omsonoonm v.0 2an

119

transported at 4 °C to the the Department of Food Science and Human Nutrition,

Michigan State University within 2 days of processing.

6.3.2 Extraction of TPI

A 2.54 cm slice was removed from the center portion of each roast using a meat
slicer (Model 512, Hobart Mfg. Co. Troy, OH 45 374). A 5 cm diameter meat core,
centered around the thermocouple, was removed from this slice. Meat was minced into
small pieces. Minced meat (50 g) was extracted in 3 volumes (150 ml) of 0.1 M NaCl,
0.05 Na phosphate buffer, pH 7.2, by blending twice (45 sec on, 1 min off) in a Waring
Blenderm. The meat homogenate was stirred on a magnetic plate for 30 min at 4 °C and
centrifuged at 4,500 x g for 10 min at 4 °C to obtain the supernatant fraction containing

TPI.

6.3.3 TPI Enzyme Activity and Concentration

Triose phosphate isomerase activity was assayed as described by Hsu (chapter 4).
The initial velocity was read at 340 nm for l min at 25°C. Samples were diluted in PBS
(pH 7.2) so that absorbance changed less than 0.2 units per minute (Norton et al., 1970).

TPI in the extracts was quantified using a sandwich ELISA (Hsu, Chapter 5).

6.3.4 Statistical Analysis
Roasts were smokehouse processed using three different processes (low, medium

and high temperature processes). Each process was conducted in triplicate for a total of 3

120

separate smokehouse processing runs. Data were analyzed using two way ANOVA
(analysis of variance) (treatment x replication). The linear or curvilinear change in
enzyme activity or concentration with temperature was determined using the polynomial

contrast test by SAS software (SAS Institute Inc., Version 6.1, 1995, Cary, NC 27513).

6.4 RESULTS AND DISCUSSION

TPI activity (U/kg meat) was similar when roasts were adequately cooked at all
temperatures (Table 6.5). TPI activity was similar at 54.4 and 62.2 °C when roasts were
inadequately processed by reducing the holding time by 0.5 log, but different when roast
beef was processed at medium temperature (58.3 °C). TPI activity was also similar when
roasts were under processed by reducing the holding time by 1.0 log from the USDA
schedules at all three temperatures. In general, no change in activity (p < 0.05) was found
as processing time of roasts was increased at low, medium and high temperatures. These
results suggested that changes in TPI activity could not be used to detect adequacy of
thermal processing of roast beef.

TPI concentration (pg TPI/g meat) by ELISA was similar when roast beef was
adequately cooked at medium (58.3 °C) and high (62.2 0C) temperature schedules (Table
6.6). TPI concentration was similar when roast beef was under processed at 0.5 log or 1.0
log reduction in holding time from the USDA schedules at medium and high
temperatures. TPI concentrations did not change when the holding time was increased at
medium (58.3 °C) and high (62.2 °C) temperature schedules. TPI concentration decreased

(p > 0.0121) when the processing holding time was increased at the low processing

12]

Table 6.5 Triose phosphate isomerase activity (U/kg meat) of roast beef processed using
three USDA approved schedules for roast beef and inadequately processed by using the

processing time by 0.5 and 1.0 log cycle in commercially processed roasts

 

Low temperature a Medium temperature a High temperature

(54.4 °C; 130 °F) (58.3 °C; 137 °F) (62.2 °C; 144 °F)

 

Under-cooked 4.4 i 2.77 A 7.3 i 4.30 A 17.8 i- 9.81 A
(1.0 log reduction in time)
Under-cooked 7.2 i 3.61 A 23.5 i 19.36 B 5.3 i 1.74 A
(0.5 log reduction in time)

Adequately cooked 3.4 i 2.10 A 9.7 a 5.26 A 9.1 :t 6.25 A

 

' Means (i standard deviation) in the same column showed a linear response to

temperature; 54.4 °C (p < 0.8856), 58.3 °C (p < 0.7151), 62.2 °C (p < 0.2042).

A 3 Means in the same row followed by the same letter are not different (p > 0.05).

Table 6.6 Triose phosphate isomerase concentration (ug/g meat) of roast beef processed
using three USDA approved schedules for roast beef and inadequately processed by

decreasing the processing time by 0.5 and 1.0 log cycle in commercially processed roasts

 

Low temperature a Medium temperature a High temperature a

(54.4 °C; 130 0F) (58.3 °C; 137 °F) (62.2 °C; 144°F)

 

Under-cooked 0.91 a: 0.177 A 0.33 3.- 0.245 B 0.38 i- 0.308 B
(1.0 log reduction in time)
Under-cooked 0.64 3: 0.424 A 0.13 i 0.065 B 0.11 i 0.038 B
(0.5 log reduction in time)

Adequately cooked 0.41 3: 0.222 A 0.13 i 0.025 B 0.11 i 0.052 A B

 

3‘ Means (a: standard deviation) in the same column showed a linear response to

temperature; 54.4 °C (p < 0.0121), 58.3 °C (p < 0.2852), 62.2 °C (p < 0.1404).

A 3 Means in the same row followed by the same letter are not different (p > 0.05).

123

temperature. Again, these results indicated that changes in TPI concentration could not
be used to detect processing adequacy of commercially processed roast beef.

Thus, although, TPI was identified as a potential TTI to detect the thermal adequacy of
roast beef processing in our previous pilot studies (Hsu, Chapter 4), TPI could not be
used as a TTI to differentiate the thermal adequacy in roast beef prepared in a commercial
pilot plant. Processing variables may have contributed to the lack of differences in TPI
concentration and activity detected between adequately and inadeaquately processed
roasts. For example, temperature variations within the smokehouse or differences in the
size and shape of roasts may have caused uneven heating processes and differential heat
transfer. Some factors were different when roasts processed in the smokehouse at the
MSU pilot plant (Chapter 4) were compared to roasts processed in a commercial
smokehouse at Bil-Mar Foods. For instance, processing conditions between the
smokehouse were different. A 55.5 °C (100 °F) difference was found between the
internal temperature (wet bulb), of the roasts and the smokehouse temperature (dry bulb)
in the commercial pilot plant, whereas only a 2.8 °C (5 °F) difference was observed
between the internal temperature (wet bulb) of roasts and the smokehouse temperature
(dry bulb) in the MSU smokehouse. Roasts processed at MSU using a slower heating
rate might undergo a more homogenous heat transfer process, allowing us to detect
differences in TPI activity or concentration between adequately and inadequately

processed roasts.

l24

6.5 CONCLUSIONS

TPI concentration and activity could not be used to detect processing adequacy of
beef roasts processed in a commercial processing plant using the USDA time-temperature
processing schedules. Variations generated by smokehouse facilities, and product heating

rate may have caused inconsistent results.

CHAPTER 7

CONCLUSIONS

Proper cooking is critical to eliminate foodbome pathogens from roast beef
products. In this study, an endogenous TTI was identified and a rapid immunoassay was
designed to detect this TTI for verifying that beef roasts received adequate thermal
treatment. TPI was selected as an endogenous TTI based on its 2 value of 5.71 °C as
determined in a previous study. This value was found to be similar to that of Salmonella
(5.65 °C) which was used by the USDA to establish the roast beef process regulations.

A ground beef water bath model system and a pilot smokehouse study at MSU
were conducted to further investigate the use of a TPI enzyme assay to determine the
thermal processing adequacy of roast beef. TPI was identified as the best marker protein
in beef cooked using medium and high temperature processes when compared to lactate
dehydrogenase, peroxidase and acid phosphatase. In the ground beef water bath model
system, TPI activity was lower in adequately processed ground beef compared to
inadequately processed ground beef. TPI activity was similar within both medium and
high temperature processes when adequately processed and TPI activity decreased as
processing time was increased at these temperatures. In the pilot smokehouse study at

MSU, TPI activity in adequately processed roast beef was lower compared to

125

126

inadequately processed roast beef. TPI activity remained constant within adequate
cooking treatments and increased as cooking time was decreased.

This doctoral research was the first study to develop a sandwich ELISA to detect
an endogenous TTI, triose phosphate isomerase (TPI), in bovine semimembranosus
muscle to ensure compliance with the USDA roast beef processing schedules. In a time-
independent ground beef system, samples were cooked from 48.9 to 76.7 °C at 5.5 °C
increments and TPI activity and concentration decreased as temperature increased. This
was confirmed on SDS-PAGE gels and Western blots.

TPI sandwich ELISA was not a suitable assay for verifying the thermal adequacy
of commercially processed roast beef. TPI sandwich ELISA could not indicate the
thermal adequacy of processed roast beef due to many factors, which may include the
nature of assay, variations within the smokehouse facilities, and size and shape of roasts.

Therefore, I conclude that the TPI enzyme assay was able to determine processing
adequacy for medium (58.3 °C) and high (62.2 °C) temperature USDA processing
schedules in the ground beef water bath model system and in the roasts processed in the
pilot plant at MSU. The TPI sandwich ELISA was notable to consistently assess thermal

adequacy in the ground beef model system and at a commercial facility.

CHAPTER 8

FUTURE RESEARCH

This study was the first to use an endogenous protein as a time-temperature
indicator (TTI) of the thermal adequacy of roast beef processing. Further investigations
as described in the following sections should be considered.

The D and 2 values of TPI and acid phosphatase were determined using
semitendinosus muscle in a previous study (Orta-Ramirez et al., 1997). D and 2 values
should also be determined in semimembranosus muscle, the primarily muscle used in
roast beef products to verify that the D and 2 values of enzymes and microorganisms in
semimembranosus muscle are similar to those of semitendinosus muscle for the same
temperature range.

Based on this study, TPI was identified as a TTI for medium (58.3 °C) and high
(62.2 °C) temperature schedules used in roast beef processes. It would be beneficial to
screen more endogenous enzymes to identify TTIs able to assess thermal adequacy of
roast beef products for all USDA process schedules.

Additional research is needed to further investigate factors which may affect TPI
activity and concentration in roast beef before TPI can be used as an intrinsic TTI on a
practical scale. The effect of heating rate, size and shape of roasts, type of smoke house,

location of roasts within the smoke house, and sampling techniques on the TPI

127

128

concentration of processed roast beef must be studied. The effect of different ingredients
(e.g., salts and spices) and processing techniques (e. g., rubbing, tumbling, injection) on
TPI activity and concentration in roast beef should be evaluated in both model and pilot
studies.

The TPI sandwich ELISA used to assess the thermal adequacy of processed roast
beef in this study utilized polyclonal antibodies and the sensitivity was low.
Development of a monoclonal antibody to incorporate in a sandwich ELISA for TPI may
yield a more sensitive and highly specific assay.

In addition to an intrinsic protein as a TTI, exogenous proteins or chemical
markers may be useful to detect processing adequacy of roast beef. A project will be
needed to identify and study how to utilize an extrinsic TTI to monitor the thermal

adequacy in an actual roast beef production system.

 

ll llll'.||'l"‘l|l|lltnlllll

 

 

REFERENCES

Abouzied, M.M., Wang, CC, Pestka, 1.1. and Smith, D.M. 1993. Lactate dehydrogenase
as safe endpoint cooking indicator in poultry breast rolls: development of
monoclonal antibodies and application to sandwich enzyme-linked
immunosorbent assay (ELISA). J. Food Protec. 562120-124, 129.

Ahmed, N.M., Conner, DE. and Human, D.L. 1995. Heat resistance of Escherichia
coli OlS7:H7 in meat and poultry as affected by product composition. J. Food
Sci. 60:606-610.

AOAC. 1990. Official Methods of Analysis, 15th ed. Association of Omcial Analytical
Chemists, Washington, DC.

Bacus, J. 1984. Meat Cultures. Ch. 3, In Utilization of [Microorganisms in Meat
Processing: A Hamflmok for Meat Plant Operators. pp. 57-84. lst ed. Research
Studies Press Ltd., Letchworth, England.

Bean, N. H. and Grimn, P. M. 1990. Foodborne disease outbreaks in the United States,
1973-1987: Pathogens, vehicles and trends. J. Food Protec. 53: 804-817.

Beisenherz, G. 1955. Triosephosphate isomerase fiom calf muscle. Ch. 57 inMethocis~ in
Enzymolog, S.P. Colowick and NO. Kaplan, Vol 1, pp. 387-391. Academic
Press, Inc., New York, NY.

Bergmeyer, H.U. 1983. Lyases, Isomerases, and Ligases. Ch. 8, inMethods of
Enzymatic Analysis, Vol. 3, 3rd ed. Verlag Chemie International, Deerfield Beach,
FL.

Bergmeyer, H.U. 1974. Enzymes as biochemical reagents. Ch. 2, inMethods of
Enzymatic Analysis, Vol. 1, 2rd ed. pp. 515, Verlag Chemie International,
Deerfield Beach, FL.

Bingham, E. W. and Zittle, C. A 1963. Purification and properties of acid phosphatase
in bovine milk. Arch. Biochem. Biophys. 101:471-477.

Bingham, E. W. and Garver, K. 1990. Purification and properties of an acid phosphatase
fiom lactating bovine mammary gland. J. Dairy Sci. 73:964-969.

129

130

Bogin. E., Israeli, B.M., and Klinger, I. 1992. Evaluation of heat treatment of turkey
breast meat by biochemical methods. J. Food Protec. 55:787-791.

Bradford, M.M. 1976. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Bioch.
72:248-254.

Caldironi, HA. and Bazan, N. G. 1980. Quantitative determination of low-salt soluble
proteins patterns of bovine muscles cooked at difi‘erent temperature. J. Food Sci.
45 :90 1-904.

CDC. 1996. http://www.cdc.gov/ncidod/diseases/foodbom/e_coli.htrn
Preventing Foodbome illness: Escherichia coli 0 157:H7. August 9, 1996.

Chaimovich, H. and Nome, F. 1970. Purification and properties of an acid phosphatase
from bovine brain. Arch. Biochem. Biophys. 139:9-16.

Cohen, EH. 1969. Determination of acid phosphatase activity in canned hams as an
indicator of temperature attained during cooking. Food Technol. 23 (7): 961-964.

Collins, S.S., Keeton, J .T. and Smith, SE 1991a. Lactate dehydrogenase activity in
bovine muscle as a potential heating endpoint indicator. J. Agric. Food Chem.
39:1291-1293.

Collins, S.S., Keeton, IT. and Smith, SE 1991b. Lactate dehydrogenase enzyme
activity in raw, cured and heated porcine muscle. J. Agric. Food Chem. 39: 1294-
1297.

Dabrowska, A, Kamrowska, I. and Baranowski, T. 1978. Purification, crystallization
and properties of triosephosphate isomerase from human skeletal muscle. Acta
Biochimica Polonica 25 :247-256.

Darnall, D.M. and Klotz, I.M. 1975. Glycolytic enzymes activity in porcine muscles.
Acta. Biochem. Biophy. 166:646-655.

Davey, CL. and Gilbert, K.V. 1974. Temperature-dependent cooking toughness in beef.
, J. Sci. Food Agric. 25:931-938.

Davis, CE. and Townsend, W.E. 1994. Rapid fluormetric analysis of acid phosphatase
activity in cooked poultry meat. J. Food Protec. 57:1094-1097.

Davis, GB, Searcy, G.K., Blankenship, LC. and Townsend, W.E. 1988. Pyruvate
kinase activity as an indicator of temperature attained during cooking of cured
pork. J. Food Protec. 51:773-777.

I31

Debruyne, I. 1983. Hen’s egg yolk acid phosphatase: purification by hydrophobic
chromatography; general characterization and kinetic properties. Int. J.
Biochem. 15:417-425.

Deshpande, SS. 1996. Assay development. evaluation, and validation. Ch. 9, In Enzyne
Immunoassays, from Concept to Product Development. pp. 27 5-3 59. Chapman
and Hall, New York, NY.

Desrocher, L.D. 1994. Enzyme-linked immunosorbent assay for rapid endpoint
processing temperature determination in turkey ham; and efi‘ects of delayed
bleeding on residual serum proteins in turkey muscles. MS. Thesis. Michigan
State University, East Lansing, MI.

Dorn, C R. 1993. Review of foodbome outbreak of Escherichia coli 0157: H7 infection
in the western United States. JAVMA 203: 1583- 1587.

Doyle, MP. and Schoeni, IL. 1984. Survival and growth characteristics of Escherichia
coli associated with hemorrhagic colitis. Appl. Environ. Microbiol. 48:855-
856.

Eber, SW. and Krietsch, W.K.G. 1980. The isolation and characterization of the
multiple forms of human skeletal muscle triosephosphate isomerase. Biochim.
Biophys. Acta 614: 173-184.

Eye, JG. 1982. A rapid procedure for the detection of underprocessing of roast beef.
Annual Meeting of the Food Research Institute. Madison, WI. May 25, 1982.
Cited in Townsend, W.E. and Blankenship, LC. 1989. Methods for detecting
processing temperatures of previously cooked meat and poultry products - A
review. J. Food Protec. 52:128-135.

Fain, Jr. AR, Line, J .E., Moran, A.B., Martin, L.M., Lechowich, R.V., Carosella, J .M.
and 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 Protec. 54:
756-761.

Farrell, Jr., H. M., Bingham, E. W., and Behe, M. J. 1988. Purification and properties of
an acid phosphoprotein phosphatase fiom lactating bovine mammary gland with
activity toward phosphotyrosine. J. Dairy Sci. 71:316-323.

Fukal, L. 1991. Modern immunoassays in meat-product analysis. Die Nahrung 35:
43 1-448.

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

132

Harlow, E. and Lane, D. 1988. Antibodies: A Laboratory Manual. Cold Spring Harbor
Laboratory, New York, NY.

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

Holbrook, J.J., Liljas, A, Steindal, SJ. and Rossmann, MG. 1975. Lactate
dehydrogenase. Ch. 4, in The Enzymes, P.D. Boyer (Ed) , Vol. 11, pp. 191-292.
Academic Press, Inc., New York, NY.

Hsu, Y.C. 1993. Characterization of enzymes suitable as endpoint temperature indicators
in turkey muscle. MS Thesis, Texas A&M University, College Station.

Jay, J. M. 1992. High-temperature food preservation and characteristics of thermophilic
microorganisms. Ch. 14, inModem FoodMicrobiolog, pp. 335-355. 4th ed.
Chapman and Hall, New York, NY.

Kang’ethe, EK. 1990. Use of immunoassays in monitoring meat protein additives.
Ch. 5, In Development and Application of Immunoassay for Food
Analysis. J.H. Rittenburg (Ed), pp. 127-139. Elsevier Applied Science, New
York, NY.

Kay, H.D. and W.R. Graham. 1993. Phosphorous compounds of milk: the efi‘ect of
heat on milk phosphatase. J. Dairy Res. 5:65-74.

Kormendy, L. and Gantner, Gy. 1960. Uber die saure Phosphatase des Flesisches. Z.
Lebensm.-Unters. Forsch. 113:13-14. Cited in Cohen, EH. 1969. Determination
of acid phosphatase activity in canned hams as an indicator of temperature attained
during cooking. Food Technol. 23(7): 961-964.

Kormendy, L. and Gantner, Gy. 1967. Neuere angaben uber die saure
phosphomonoesterase des fleisches. Z. Lebensm.-Unters. Forsch. 134: 141-145.
Cited in Cohen, EH. 1969. Determination of acid phosphatase activity in canned
hams as an indicator of temperature attained during cooking. Food Technol.
23(7): 961-964.

Kormendy, L., Rehasi, E. and Fetter, I. 1987. Determination of the extent of heat
treatment in canned hams by use of the phosphatase test. Meat Sci. 19:77-79.

Kuby, J. 1994. Antigens. Ch. 4, inlmmunology, pp. 85-108. 2nd ed. W. H.
Freeman and Company. New Yorlg NY.

Lee, E.W., Barriso, J .A, Pepe, M. and Snyder, R. 1971. Purification and properties of
liver triose phosphate isomerase. Biochim. Biophys. Acta 242: 261-267.

133

Levieux, D., Levieux, A, and Venien, A. 1995. Irnmunochemical quantification of heat
denaturation of bovine meat soluble proteins. J. Food Sci. 60:678-684.

Line, J .E., Fain, Jr., A.R., Moran, AB., Martin, L.M., Lechowich, R.V., Carosella, J .M.
and Brown, W.L. 1991. Lethality of heat to Escherichia coli 01572H7: D-value
and Z-value determinations in ground beef. J. Food Protec. 54:762-766.

Lund, DB. 1975. Heat Processing. Ch. 3, In Principles of Food Science, Part 11,
Physical Principles of Food Preservation, M. Karel, O.R F ennema, and DB.
Lund (Ed), pp. 31-92. Marcel Dekker, New York, NY.

Rittenburg, 1H. and Grothans, GD. 1992. Introduction to immunoassay techniques:
Immuoassays : Formats and Applications. Cited in In Food Safety and Quality
Assurance: Applications of Immunoassay Systems. Morgan, M.R.A, Smith, C].
and Williams, P.A. (Ed) pp. 3-10. Elsevier Applied Science, New York, NY.

National Canners Association. 1968. Laboratory manual for food canners and
processors. Vol. 1. AVI Publishing Co., Westport, CT.

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

Norton, I.L., Pfuderer, P., Stringer, CD, and Hartman, EC. 1970. Isolation and
characterization of rabbit muscle triose phosphate isomerase. Biochem. 924952-
4958.

Nusirnovich, A.D., Celrni RA. and Pagliaro AF. 1979. Color determination of beef
juices as an indicator of beef cooking temperatures. Meat Sci. 19277-79.

Oreshkin, E.F., Borisova, M.A., Tchubarova, GS. and Gorbatov, V.M. 1968.
Conformational changes in the muscle proteins of cured beef during heating.
Meat Sci. 16:297-305.

Orta-Ramirez, A, Wang, C.H., Abouzied, M.M., Veeramuthu, G.J., Price, J.F., Pestka,
J .J . and Smith, D.M. 1996. Lactate dehydrogenase monoclonal antibody sandwich
ELISA to determine cooking temperature of ground beef. J. Agric. Food Chem.
44:4048-4051.

Orta-Ramirez, A. , Price, J.F., Hsu, Y.-C., Veeramuthu, G.J., Cherry-Merritt, IS. and
Smith D.M. 1997. Thermal inactivation of Escherichia coli 01 57:H7, Salmonella
and enzymes with potential as endpoint temperature indicators in hamburger
patties. J. Food Protec. In press.

Osborne, B.G. and Fearn, T. 1986. Applications of near infi'ared spectroscopy in food

134

and beverage analysis. Ch. 8, In Near Infiared Spectroscopy in Food Analysis.
pp. 145-199. Longman Scientific and Technical, New York, NY.

Pestka, II 1988. Enhanced surveillance of foodbome mycotoxins by immunochemical
assay. J. Assoc. Ofi‘. Anal. Chem. 71:1075-1081.

Pestka, 1]., Lee, Y.K. and Chu, F .S. 1982. Reactivity of aflatoxin 32a antibody with
aflatoxin B1 modified DNA and related metabolites. Appl. Environ. Microbiol.
44: 1159-1165.

Petell, J.K., Killion, RB. and Lebherz, KG. 1982. Isolation of several abundant muscle
enzymes. Ch. 80 inMethods in Enzymology, W. A. Wood (Ed), Vol. 90, pp.
490-497. Academic Press, Inc., New York, NY.

Quinn, J.R., Raymond, DP. and Harwalker, V.R. 1980. Difi‘erential scanning
calorimetry of meat proteins as affected by processing treatment. J. Food Sci.
45: 1 146-1 149.

Reiss, NA and Schwartz, RI. 1987 . High performance purification of glycolytic
enzymes and creatine kinase from chicken breast muscle and preparation of their
specific immunological probes. Preparative Biochemistry 17:157-172.

Samarajeewa, U., Wei, C.I., Huang, TS. and Marshall, MR. 1991. Application of
immunoassay in the food industry. Crit. Rev. Food Sci. Nutr. 29:403-434.

SAS Institute, Inc. 1995. SAS User’s Guide: Statistical Analysis System, version 6.1 ed.
SAS Institute Inc. Gary, NC.

Sawyer, T.H., Tilley BE, and Gracy R.W. 1972. Studies on human triosephosphate
isomerase. Nature of the electrophoretic multiplicity in erythrocytes. J. Biol.
Chem. 247: 6499-6505.

Scopes, R. K. 1973. Studies with a reconstituted muscle glycolytic system, the rate and
extent of creatine phosphorylation by anaerobic glycolysis. Biochem. J. 134:197-
208.

Scopes, R.T. and Stoter, A. 1982. Purification of all glycolytical enzymes fiom one
muscle extract. Ch. 79, InMethods in Enzymology, W. A. Wood, Vol. 90, pp.
479-490. Academic Press, Inc., New York, NY.

Smith, D.M. and Desrocher, LB. 1996. Immunoassays for determination of endpoint
processing temperatures in poultry and beef products. J. Muscle Foods 7:33 5-344.

Smith, D.M., Desrocher, L.D., Booren, A.M., Wang, C.H., Abouzied, M.M., Pestka, 1.].
and Veeramuthu, GJ. 1996. Cooking temperature of turkey ham affects lactate

135

dehydrogenase, serum albumin and Irnmunoglobulin G content as determined by
ELISA. J. Food Sci. 61:209-212, 234.

Smith, G.L. 1991. Utilization of enzymes to provide heating endpoint markers and modify
endogenous cholesterol in muscle foods. Ph.D. Disseration, Texas A&M
University, College Station, TX.

Stalder, J .W., Smith, G.L., Keeton, IT. and Smith, SE. 1991. Lactate dehydrogenase
activity in bovine mucscle as a means of determining heating point. J. Food Sci.
56: 895-898.

Todd, E.C.D. 1994. Cost of foodbome illnesses caused by bacteria and seafoods toxins.
Paper No. 27-1, presented at the 54th Annual Meeting of Inst. of Food
Technologists, Atlanta, GA, June 25-29.

Townsend, WE. and Blankenship, L.C. 1987a. Assessment of previous heat treatment of
laboratory heat-processed meat and poultry using a commercial enzyme system. J.
Food Sci. 52:1445-1448.

Townsend, W.R. and Blankenship, L.C. 1987b. Enzyme profile of raw and heat-
processed beef, pork, and turkey using the APIZYME system. J. Food Sci.
52:5 1 1-512.

Townsend, W.E. and Blankenship, LC. 1989. Methods for detecting processing
temperatures of previously cooked meat and poultry products - A review. J. Food
Protec. 52: 128-135.

Townsend, W.E., Thomson, J .E. and Hutchins, J .R. 1984. Coagulation test for cooked
meat temperature; efi‘ect of variations in filtration. J. Food Sci. 49: 853-858.

Townsend, W.E. and Davis, CE. 1992. Transaminase (AST/GOT and ALT/GPT)
activity in ground beef as a mean of determining end-point temperature. J. Food
Sci. 57:555-557.

USDA-F SIS. 1986a. Determination of internal cooking temperature (Coagulation).
Revised basic chemistry laboratory guidebook. No. 3.019: 3-55. Science
Chemistry Division. Food Safety and Inspection Service, Washington, DC.

USDA-F SIS. 1986b. Determination of internal cooking temperature (Acid phosphatase
activity). Revised basic chemistry laboratory guidebook. (Revised March 1986)
No. 3.018: 3-49. Science Chemistry Division. Food Safety and Inspection Service,
Washington, DC.

USDA-FSIS. 1989. Performing the catalase enzyme test. A self instruction guide.
Technical Services Training Division. Food Safety and Inspection Service,

136

Washington, DC.

USDA-FSIS. 1995. Requirements for the production of cooked beef, roast beef and
cooked corned beef. Meat and Poultry Inspection Regulations. Animal and
Animal Products. Oflice of the Federal Register, National Achieves and Records,
GSA Washington, DC. Title 9 of Code of Federal Regulations, Title 9, Part
218.] 7.

USDA-FSIS. 1996a. http://www.usda.gov/agency/fsis/finalnrl.htrn
The Final Rule on Pathogen Reduction and Hazard Analysis and Critical Control

Point (HACCP) system. August 9, 1996.

USDA-F SIS. 1996b. Performance standard for the production of certain meat and poultry
products. Federal Register. 61(86), 19564. US. Department of Agriculture,
Food Safety Inspection Service, Washington, DC. May 2, 1996.

Van Loey, A, Hendrickx, M., De Cordt, S. De., Haentjens, T. and Tobback, P. 1996.
Quantitative evaluation of thermal processes using time-temperature integrators.
Trends Food Sci. Tech. 7:1-11.

Wang, C.H., Abouzied, M.M., Pestka, J .J . and Smith, D.M. 1992. Antibody development
and enzyme-linked immunosorbent assay for the protein marker lactate
dehydrogenase to determine safe cooking end-point temperatures of turkey rolls. J.
Agric. Food Chem. 40:1671-1676.

Wang, C.H., Booren A.M., Abouzied, M.M., Pestka, J .J . and Smith, D.M. 1993. ELISA
determination of turkey roll endpoint temperature: efi‘ects of formulation, storage,
and processing. J. Food Sci. 58:1258-1261, 1264.

Wang, C.H., Pestka, J.J., Booren, AM. and Smith, D.M. 1994. Lactate dehydrogenase,
serum protein, and immunoglobulin G content of uncured turkey thigh rolls as
influenced by endpoint cooking temperature. J. Agric. Food Chem. 42: 1829-1833.

Wang, C.H., Abouzied, M.M., Pestka, U. and Smith, D.M. 1995. Lactate dehydrogenase
polyclonal antibody sandwich ELISA for determination of endpoint heating
temperatures of ground beef. J. Food Protec. 59:51-55.

Wang, S.F., Abouzied, M.M. and Smith, D.M. 1996. Proteins as potential endpoint
temperature indicators for ground beef patties. J. Food Sci. 61 :5-7.

Willardson, R.R., Busta, FF. and Allen, CE. 1977. Dialysis technique for containment
of microbial populations inoculated into food systems. Appl. Environ. Microbiol.
34:240-241.

Wilson, CM. 1983. Staining of proteins on gels: comparsions of dyes and procedures.
Ch. 18, in Methods in Enzymology, C.H..W. Hirs and SN. Timasheff (Ed), Vol.
91, pp. 236-247. Academic Press, Inc., New York, NY.

Wright, J .D. and Wilding, P. 1984. Differential scanning calorimetric study of muscle
and its proteins: myosin and its subfragrnents. J. Sci. Food Agric. 35:357-372.

APPENDIX

PRELIMINARY EXPERIMENTS TO DETERMINE

EXTRACTION CONDITIONS

Pure TPI Model System
Materials and Methods

Lyophilized TPI was diluted in double distilled water to achieve 0.1 mg TPI/mL
MES buffer (since TPI was suspended in 30 mM MES buffer, pH 6.5, before
lyophilized). The TPI solution was pipetted (0.6 mL) into 2 mL polypropylene
microcentrifuge tubes (Fisher Scientific, Pittsburgh, PA 15219). Tubes were heated in a
Polystat circulator water bath (Model 1268-52, Cole-Parmer Instrument Co., Chicago, IL,
60714) at 82.2 °C (180 °F). Tubes were unheated or heated to 48.9 °C (120 °F), 60.0 °C
(140 °F) and 76.7 °C (170 °F), vortexed, and centrifuged at 4,500 x g for 10 min. TPI
activity and concentration in the supernatant was measured using enzyme assay and

ELISA, respectively.

Results and Discussion
A pure TPI model system was used to understand how the denaturation and
insolubilization of TPI were affected by heating and centrifugation. The effect of
centrifugation on activity and concentration of TPI heated at 48.9 °C (120 °F), 60.0 °C
(140 °F), and 76.7 °C (170 °F) were investigated in preliminary experiments (Table 1).
The effect of centrifugation on TPI activity and concentration in unheated and

heated extracts was studied. TPI activity and concentration decreased after centrifugation

139

of unheated TPI extracts, suggesting that some native TPI was precipitated by
centrifugation. When heated, TPI activities in centrifuged and non-centrifuged extracts
were similar. Total protein in centrifuged and non-centrifuged extracts did not differ in
unheated extracts. When heated, total protein decreased in centrifirged extracts as heating
temperature increased. Without centrifugation, TPI activity and concentration decreased
as heating temperature was increased except TPI activity at 60 °C. When centrifuged,

TPI concentration or activity decreased as heating temperature was increased.

140

Table 1. Preliminary experiments evaluating the effect of heating and centrifugation on

TPI activity and concentration in 30 mM MES buffer a, pH 6.5

 

 

 

Extraction TPI activity TPI concentration Total protein
Method (U/L extract) (pg/mL extract) (g/L extract)
Not centrifuged
Unheated 582.2 3: 567.55 b (3) c 60.6 i 72.35 (3) 0.4 i 0.03 (2)
48.9 °C 353.3 (1) 20.4 (1) ** A
60.0 °C 385.2 (1) 9.3 (l) **
76.7 °C 29.4 (1) 4.7 (1) **
Centrifuged
Unheated 242.6 i 105.22 (3) 12.2 3: 4.25 (3) 0.4 (1)
48.9 °C 391.7 i 100.99 (3) 45.0 i 63.12 (3) 0.4 i 0.001 (2)
60.0 °C 345.6 i 38.03 (3) 17.2 i 22.20 (3) 0.4 :t 0.01 (2)
76.7 °C 87.4 :1: 122.85 (3) 7.6 i 9.26 (3) 0.3 i 0.01 (2)

 

' 30 mM MES buffer: pH 6.5, including 30 mM (2-[N-Morpholino]ethanesulfonic acid),

10 mM KOH, 0.5 mM magnesium acetate, 0.1 mM EDTA.

b Results were expressed as mean i standard deviation.

° value in parentheses indicates number of replicates.

d **: data not available.

141

Selection of Extraction Buffers

Phosphates may inhibit TPI activity (Beisenherz, 1995), therefore, a preliminary
study was used to find a suitable buffer system for extracting TPI from cooked beef.
Bovine TPI was extracted from ground beef with a series of phosphate buffers (PBS) of
different molarities (.01, .05, 0.1 M) containing 0.1 M NaCl, pH 7.2, or a MOPS buffer
(0.1 M NaCl, .01 M MOPS, pH 7.2). In general, no difference in TPI activity was found
when different molarities of PBS, ranging from .01 to .1 M, were used to extract TPI
from ground beef (Table 2). TPI activities were higher when meat was extracted with
phosphate buffers of different molarities than with MOPS buffer.

Water and PBS (0.15 M NaCl, 0.01 M Na phosphate, pH 7.2) were used as
diluents to examine their effect on TPI activities in meat extracts. When diluted in water,
after extraction, but prior to determination of enzyme activity, TPI activities were about
50% lower in raw meat extracted with MOPS than with phosphate buffer; TPI activities
were similar in phosphate buffer systems. When diluted in PBS, TPI activities were
about 70% lower in raw meat extracted with MOPS than with phosphate bufl'ers. Results
suggested that phosphate ions increased the extractability of TPI from ground beef.
Phosphates are known to cause swelling and destruction of myofibrillar fibers by
enhancing electrostatic repulsion between myofilaments (Offer and Trinick', 1983);
therefore, phosphates facilitate the release of sacroplasmic enzymes fiom the muscle
fibers and enhanced the extractability of TPI. In PBS and MOPS extraction systems, TPI
activities were similar or slightly higher in meat extracts diluted in water compared to

dilution in PBS.

I42

Table 2. Effect of extraction buffers and diluents on TPI activity (U/kg meat) extracted

fi'om semimembranosus muscle at pH 7.2

 

0.1 NaCl 0.1 M NaCl 0.1 M NaCl 0.1 M NaCl

0.01 Na phosphate 0.05 M Na phosphate 0.] M Na phosphate 0.01 M MOPS

 

diluted 3425.12 3680.79 3741.76 1809.48
in water
diluted 3445.24 3536.31 3342.47 1037.56
in PBS 3

 

a 0.15 M NaCl, 0.01 M Na phosphate (pH 7 .2) buffer

n=l

Methods of Meat Extract Preparation

The purpose of this study was to establish a suitable method to extract TPI from
cooked ground beef. Three sample preparation methods were examined: ( l) supernatant
collected after centrifugation of meat extracts (treatment A), (2) filtrate collected after
filtering the meat extract through # 4 Whatrnan filter paper (treatment B), and (3) filtrate
collected after filtering the meat extract through 0.22 pm filter (19952, Millipore
Products Division, Bedford, MA 01730) (treatment C).

The filtrate was collected and held at 4 °C until used. TPI activity and
concentration of extracts were determined within 24 hr. The filtrate was diluted in either
water or 0.15 M NaCl, 0.01 M Na phosphate (pH 7.2) buffer prior to the enzyme assay.
TPI activity and concentration were compared to establish a more consistent and reliable
procedure for extracting TPI from a cooked ground beef model system.

In treatments A, B, and C, TPI activities of ground beef decreased (P < 0.0001) as
temperature was increased from 48.9 °C (120 °F) to 76.7 0C (170 °F) (Table 3).
However, in treatment B, TPI activity was lower in the 60 °C extract than in the 65.6 °C
extract. The coefficient of variations of treatments A, B, and C were also used to
compare method variability. In general, the coefficient of variation of TPI activity was
smaller (less than 15%) in treatment A when compared to treatments B and C, suggesting
that the centrifugation method (treatment A) of meat extract preparation provided more
consistent results.

In treatments A, B, and C, TPI concentrations by ELISA decreased in ground

meat as temperature was increased from 48.9 °C (120 °F) to 76.7 0C (170 °F) (Table 4).

144

Table 3. Effect of cooking temperature on triose phosphate isomerase activity (U/kg
meat) of bovine meat cooked in a water bath and extracted using centrifugation (treatment

A), centrifuging and filtering (treatment B), and centrifuging and filtering through

 

0.22um filter (treatment C) a
Internal Treatment
temperature (°C) A B C

 

48.9:0.1 5375:1455a ( 2.71) 409.6: 13.14( 3.21) 5495:2058 ( 3.75)

54.4i0.l 187.0i21.88 (11.70) 109.6: 5.37( 4.90) 179.1: 6.11 (3.41)

60.0:0.1 72.1: 9.03 (12.53) 38.2: 10.65 (27.91) 47.0: 11.01 (23.45)
65.6:0.1 38.0: 3.08 (8.12) 66.0:23.63(35.78) 21.3: 2.14 (10.05)
71.1:0.1 27.4: 3.98 (14.54) 31.9: 2.94( 9.22) 15.4: 3.09 (20.12)
76.7:0.1 19.4: 1.01 ( 5.21) 28.1: 1.06(3.77) 12.3: 0.64 ( 5.22)

 

‘ Values are expressed as mean i standard deviation of triplicate determination and
values in parentheses represent coefficient of variation. Means in the same column

decreased linearly as temperature increased (P < 0.0001).

145

Table 4. Effect of cooking temperature on triose phosphate isomerase content of bovine

meat cooked in a water bath and extracted using centrifugation (treatment A),

centrifuging and filtering (treatment B), and centrifuging and filtering through 0.22pm

filter (treatment C)

 

 

Internal Treatment

temperature (°C) A B C

48.9: 0.1 86.3 : 11.55‘(13.38) 68.9: 27.17 (39.45) 59.2 : 30.41 (51.40)
54.4:0.1 66.9: 6.64 ( 9.93) 44.8: 10.80 (24.12) 32.8: 7.07 (21.59)
60.0:0.1 48.3: 6.67 (13.82) 26.7: 15.33 (57.50) 26.6: 9.56 (35.94)
65.6:0.l 42.6: 8.75 (20.56) 23.6: 11.56 (48.98) 19.7: 8.31 (42.20)
71.1:0.1 36.8: 7.17 (19.46) 28.7: 17.10 (59.60) 24.9: 10.13 (40.75)
76.7:0.1 34.1: 6.29 (18.46) 20.7: 9.82 (47.39) 29.4: 19.53 (66.41)

 

a Values are expressed as mean : standard deviation of triplicate and values in

parentheses represent coefficient of variance. Means in the same column decreased

linearly as temperature increased, Treatment A (P < 0.0001), Treatment B (P < 0.0031)

and Treatment C (P < 0.0432).

146

However, in treatment B, TPI concentration was higher in meat cooked at 71.1 °C than at
60 and 65.6 °C. In treatment C, TPI concentration was lower in 65 .6 °C than at 71.1 °C;
TPI concentration was greater in meat cooked to 76.7 °C than at 71.1 °C. Greater
coefficients of variation were also found in treatments B and C, ranging from 21 to 66%,
compared to treatment A (the coefficient of variations were all less than 20%). The
centrifugation method was the best method to prepare meat extracts as it produced the

most consistent experimental results when TPI was measured by enzyme assay and

ELISA.

 

1 Offer, G. and Trinick, J. 1983. On the mechanism of water holding in meat: the

swelling and shrinking of myofibrils. Meat Sci. 8:245-281.

I47

PROXIMATE COMPOSITION OF BIL-MAR ROAST BEEF

Beef roasts (semimembranosus) fiom Bil-Mar was analyzed for protein, fat and
moisture contents using AOAC2 (1990) standard methods 981.10, 960.39 and 950.46B,
respectively. The pH was determined by homogenizing 10 g ground beef with 90 mL of
double distilled water using a WaringTM blender for 30 5.

Protein, fat and moisture contents were 27.05 : 0.95 %, 1.19 : 0.11 % and 74.19 : 4.05

%, respectively. The pH was 5.8.

 

2 AOAC. 1990. Ofi‘icial Methods ofAnalysis,15th ed. Association of Official

Analytical Chemists, Washington, DC.

148

MICHIGAN srarE UNIV. LIBRARIES
llHI”WWWlllllllllllllllllull“lllllllllllllllllllll
31293015650207