3&2 “fix-.4.
«natum... rd! »
a .

run,
a?

.f
er;a

:3.
n3t.

.
.
. 2.134..

.5Mfl

IIV\'~

an»
‘zhflr:tfiu

. as“...
9. he”

 

.r. .
i- aux

 

{.1}... 1.4:.“ :1.

. A ‘ luXICriII

‘ . In. v.3)..9v. )J‘
_ ‘ . ‘ . . ..?§§
1-: r _ . Suinhwhsihurk... .
V P. .. ‘ {3)} ‘10..I1.~.In
1... 135 . , . z , 2...}; .

4‘.

 

 

 

 

.11.; 3. «angina. . “,3. ‘ . .239... . ”mam; 23.234.15. . 4 .mu i... . ififlb..§ 3.3,.
w P ‘ : . . . r 2.!

$091ng

”fun“ .J
ix)”; _/ u

3/18/1176,;

 

 

 

LIDDA RY

u.“ Ir'\

Michigan State
University

 

 

 

This is to certify that the
thesis entitled

Thermal Resistance of Sublethally Injured Salmonella

presented by

Alissa M. Wesche

has been accepted towards fulfillment
of the requirements for the

MS. degree in Food Science and Human
Nutrition

 

 

Major Professor’s’ Signature

G/I’c /o_3

Date

MSU is an Affimativa Action/Equal Opportunity Institution

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

U1

shvadjw

 

JAN 30 {2005f "

 

Ci?! E .12 Emmi

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIFlC/DateDuepSS—p. 1 5

THERMAL RESISTANCE OF SUBLETHALLY INJURED SALMONELLA
By

Alissa M. Wesche

A THESIS

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

MASTER OF SCIENCE
Department of Food Science and Human Nutrition

2003

ABSTRACT
THERMAL RESISTANCE OF SUBLETHALLY INJURED SALMONELLA
By

Alissa M. Wesche

Recovery of stressed cells is often limited by the choice of culture media and
environmental conditions, including oxygen exposure. The ability of reducing agents,
which can neutralize reactive oxygen species, to enhance recovery of sublethally injured
Salmonella was studied. Using Trypticase soy agar with 0.6% yeast extract (TSA-YE)
supplemented with Oxyrase®, 0.1% sodium pyruvate and 0.6% sodium pyruvate,
recovery of heat-shocked (30 min / 54°C), cold-shocked (2 h / 4°C) and starved cells (10
d in phosphate buffer at 4°C) ranged from 89.6 to 90.2%, 94.9 to 95.1% and 99.2 to
101.6%, respectively, and was not significantly different from unsupplemented TSA-YE.
Sublethally injured cells may also have heightened sensitivity to thermal processing, due
to activated stress response mechanisms. To investigate the effects of sublethal stress on
Salmonella inactivation kinetics, an 8-strain Salmonella cocktail was subjected to heat
shock, cold shock, and starvation stress, harvested by centrifugation and inoculated into
irradiated comminuted turkey. The DwoC-value for heat-shocked Salmonella on TSA-YE
was 0.64 min, which was significantly higher (P<0.05) than that for the unshocked
control (0.41 min), whereas those for cold shock (0.38 min) and starvation (0.41 min)
were not significantly different from the control. Additionally, more thermal inactivation
was observed using selective xylose lysine deoxycholatc (XLD) agar than using non-

selective TSA-YE, resulting in an underestimation of survivors on the selective medium.

To my Mom and Dad

iii

ACKNOWLEDGMENTS

I want to express my deepest gratitude to my major professor, Dr. Elliot Ryser.
The completion of this thesis would not have been possible without his immense support
and patient guidance in both research and academics. I also want to thank my committee
members, Dr. Bradley Marks and Dr. James Pestka. Their wonderful insight and
direction throughout this process have been greatly appreciated. I am very thankful to
Dr. George Garrity, without his mentoring I might not have pursued graduate school.

The Statistical Consulting Services of the CANR Biometry Group and the
Department of Statistics and Probability helped with the statistical analyses of my
research data. I especially want to thank Fernando Cardoso and Fang Li for teaching me
so much about statistics, including the intricacies of SAS®.

I want to acknowledge all of my labmates, past and present, for their
encouragement, fi'iendship and support during the good times and the bad. I am indebted
to my assistant and friend, Julie Rochowiak, without her help I would still be in the
middle of the thermal inactivation studies.

And finally, thank you to my family for all of your love and support.

iv

TABLE OF CONTENTS

LIST OF TABLES vii
LIST OF FIGURES x
KEY TO SYMBOLS OR ABBREVIATIONS xi

CHAPTER 1: Introduction 1
1.1 F oodbome Illness in the United States 1
1.2 The Meat Industry and Relevant Regulations 2
1.3 Justification 7
1.4 Objectives and Hypothesis 8

CHAPTER 2: Literature Review 10
2.1 Salmonella 10
2.2 Thermal Inactivation of Salmonella 17
2.3 Sublethal Injury and Stress 21
2.3.1 Definitions 21
2.3.2 Specific Stresses 25
2.3.3 Cellular Modification and Damage 30
2.3.4 Oxygen Sensitivity of Injured Cells 35
2.3.5 Repair and Stress Responses 40
2.3.6 Cross-Protection 48
2.3.7 Virulence 55
2.3.8 Recovery of Injured Microorganisms 59
CHAPTER 3: Efficacy of sodium pyruvate and Oxyrase® for recovering heat-, cold- and
starvation-injured Salmonella 64
3.1 Abstract 64
3.2 Introduction 64
3.3 Materials and Methods 67
3.3.1 Bacterial Strains 67
3.3.2 Culture Preparation 68
3.3.3 Heat Shock Treatment 68
3.3.4 Cold Shock Treatment 69
3.3.5 Starvation Treatment 69
3.3.6 Enumeration of Cells 70
3.3.7. Data Analysis 71
3.4 Results 71
3.4.1 Sublethal Injury 71
3.4.2 Recovery of Injured Cells 71
3.5 Discussion 75

3.6 Conclusion 80

CHAPTER 4: Thermal resistance of heat-, cold- and starvation-stressed Salmonella in

irradiated comminuted turkey 81
4.1 Abstract 81
4.2 Introduction 82
4.3 Materials and Methods 85

4.3.1 Bacterial Strains 85

4.3.2 Culture Preparation 85

4.3.3 Preparation of the Unshocked Salmonella Cocktail 86

4.3.4 Heat Shock Treatment 86

4.3.5 Cold Shock Treatment 87

4.3.6 Starvation Treatment 88

4.3.7 Comminuted Turkey 88

4.3.8 Inoculation of Comminuted Turkey 89

4.3.9 Thermal Inactivation 90

4.3.10 Enumeration 91

4.3.11 Data Analysis 92

4.4 Results 92
4.4.1 Composition of Ground Turkey 92

4.4.2 Sublethal Injury 92

4.4.3 Thermal Resistance of Salmonella Cocktails 93

4.5 Discussion 98
4.6 Conclusion 105
CHAPTER 5: Conclusions and Future Recommendations 106

APPENDIX A: Development of a novel gradient plate method for assessing sublethal

Injury 109
A.1 Abstract 109

A2 Introduction 109

A.3 Materials and Methods 111

A.3.1 Bacterial Strains 111

A.3.2 Culture Preparation 111

A33 Heat Shock Treatment 111

A34 Gradient Plate Production 112

A35 Recovery of Cells 115

A.3.6 Data Analysis 116

A4 Results 116

A.4.l Sublethal Injury 116

A42 XLD/TSA-YE Gradient Plates 116

A.4.3 BSA/TSA-YE Gradient Plates 117

A5 Discussion 121

A6 Conclusion 124
APPENDIX B: Supplementary Data and Statistics 125

REFERENCES 1 3 1

vi

LIST OF TABLES

 

 

Table Title/ Caption Page

No.

2.1 Biochemical characteristics of a typical Salmonella isolate (D’Aoust 11
1997, Madigan et al. 1997, Bell and Kyriakides 2002).

2.2 Species, subspecies and serovars of the genus Salmonella (adapted 11
from Brenner et al. 2000).

2.3 Salmonella nomenclature recommended by the WHO Collaborating 13
Centre for Reference and Research on Salmonella and used at the
CDC (adapted from Brenner et al. 2000).

2.4 Predominant clinical serovars of salrnonellae in the United States 14
between 1996 and 2000'.

2.5 Physical and chemical treatments (stresses) with the ability to 23
reversibly injure microorganisms (Ray 1989, Ray 2001, Wuytack et
al. 2003, Thanomsub et al. 2002).

2.6 Microorganisms known to experience reversible injury fiom sublethal 23
stresses that are of importance in food systems (Ray 1989, Ray 2001).

2.7 Types of cell damage resulting fiom different sublethal treatments 31
(adapted from Mossel and van Netten 1984, Ray 1986, Wuytack et al.
2003, Rowbury 2003).

2.8 Treatments known to enhance thermotolerance for microorganisms. 51

3.1 The effect of sodium pyruvate and Oxyrase® on recovery of heat- 73
injured Salmonella.

3.2 The effect of sodium pyruvate and Oxyrase® on recovery of cold- 73
stressed Salmonella.

3.3 The effect of sodium pyruvate and Oxyrase® on recovery of starved 74
Salmonella.

4.1 Dam-values and regression parameters on TSA-YE for unshocked 97
and heat-, cold- and starvation-stressed Sahnonella.

4.2 Dam-values and regression parameters on XLD for unshocked and 97

heat-, cold— and starvation-stressed Salmonella.

vii

 

 

Table Title/ Caption Page

No.

4.3 Injury indexa following thermal treatment (60°C) of Salmonella. 97

Al Growth of Salmonella on standard TSA-YE and XLD plates and in 117
selected columns of XLD/TSA-YE gradient plates.

A.2 Recovery of Salmonella on XLD/TSA-YE gradient plates 118

A.3 Growth of Salmonella on standard TSA-YE and BSA plates and in 118
selected columns of BSA/TSA-YE gradient plates.

A.4 Growth of Salmonella on standard TSA-YE and BSA plates and in 119
selected columns of BSA/TSA-YE gradient plates.

A.5 Recovery of Salmonella on BSA/TSA-YE gradient plates 119

8.1 Preliminary heat shock data comparing percent injury (on BSA) at 125
different temperatures.

32 Preliminary cold shock (4°C) data comparing percent injury (on BSA) 125
at different time intervals.

33 Preliminary starvation data comparing percent injury (on BSA) at 126
different time intervals.

B.4 Injury (on BSA) for 8 individual Salmonella strains following cold 126
storage in TSB-YE at 4°C for 10 (1.

B5 Analysis of variance (ANOVA) table for injury of 8 individual 127
Salmonella strains following heat shock (54°C/30 min), cold shock
(4°C/2 h) or starvation (4°C/10 d) (Proc Mixed, SAs® Version 8.02,
SAS Institute, Inc., Cary, NC).

B.6 AN OVA table for recovery of Salmonella following heat shock, cold 127
shock or starvation (Proc Mixed, SAS® Version 8.02, SAS Institute,
Inc., Cary, NC).

B.7 AN OVA table for the effects of treatment, medium, time and their 128

- interactions on the number of survivors following thermal (60°C)

treatment of Salmonella (Proc GLM, SAS® Version 8.02, SAS
Institute, Inc., Cary, NC).

 

viii

 

 

Table Tit le/ Caption Page
No.
B.8 Statistical analysis (Proc GLM, SAS® Version 8.02, SAS Institute, 128
Inc., Cary, NC) of the thermal inactivation of Salmonella, sorted by
medium (Proc Sort, SAS® Version 8.02), with unshocked as the
control on each medium.
B.9 Percent injury on three different selective media following heat shock. 129
B.10 Percent injury on three different selective media following cold shock. 129
B.11 Percent injury on three different selective media following starvation. 131

 

LIST OF FIGURES

 

 

Table No. Title/Caption Page

2.1 Specific types of damage to bacterial DNA (Russell 1984). 34

2.2 Certain stresses encountered by microorganisms in the 56
environment are also imposed by a host’s defense mechanisms
(Foster and Spector 1995).

3.1 Recovery of sublethally injured Salmonella on unsupplemented 74
and supplemented TSA-YE. Values are plotted as averages,
with standard deviations shown as error bars.

4.1 Thermal inactivation at 60°C of the unshocked Salmonella 94
cocktail. Values plotted are the arithmetic means of six trials,
with standard deviations shown as error bars.

4.2 Thermal inactivation at 60°C of the heat-shocked Salmonella 94
cocktail. Values plotted are the arithmetic means of four trials,
with standard deviations shown as error bars.

4.3 Thermal inactivation at 60°C of the cold-shocked Salmonella 95
cocktail. Values plotted are the arithmetic means of four trials,
with standard deviations shown as error bars.

4.4 Thermal inactivation at 60°C of the starved Salmonella cocktail. 95
Values plotted are the arithmetic means of four trials, with
standard deviations shown as error bars.

A.1 Integrid 100x 100x 15 mmsquare plates with6x6grid. 114

A2 Preparation of the bottom (selective) layer of gradient plates. 1 14
"‘ Angle measurement: ~ 4°

A.3 Completed gradient plate, with a selective (bottom) and non- 114
selective (top) layer.

A.4 Recovery on XLD/TSA-YE gradient plates following heat 120
shock.

A.5 Recovery on BSA/1‘ SA-YE gradient plates following heat 120

shock.

AMI
AOAC
APHA
ASP
ATP
ATR
BG
BPB
BSA
CDC
CFU
cm
CSP

d
D-value
DF
DNA
ERS

F SIS

g
GLM
HACCP
h

HSP
IFT

lb

LPS
min
mL
mm
MPa
PFGE
ROS
RTE
RNA

spv
SOD
SSOP
TCA
TGE
TSA

KEY TO SYMBOLS OR ABBREVIATIONS

American Meat Institute
Association of Official Analytical Chemists
American Public Health Association
acid shock protein

adenosine triphosphate

acid tolerance response

brilliant green

Butterfield’s phosphate buffer
bismuth sulfite agar

Centers for Disease Control

colony forming unit

centimeter(s)

cold shock protein

day(s)

decimal reduction time

degrees of freedom
deoxyribonucleic acid

Economic Research Service

Food Safety and Inspection Services
gram(s)

general linear model

Hazard Analysis and Critical Control Points
hour(s)

heat shock protein

Institute of Food Technologists
pound(s)

lipopolysaccharide layer

minute(s)

milliliter(s)

millimeter(s)

megapascal

pulsed-field gel electrophoresis
reactive oxygen species

ready-to-eat

ribonucleic acid

second(s)

Salmonella Plasmid Virulence
superoxide dismutase

Sanitation Standard Operating Procedure
tricarboxylic acid cycle

tryptone- glucose extract

Trypticase soy agar

xi

TSA-YE
TSB
TSB-YE

UHT
USDA
VRBA
WHO
XLD

Trypticase soy agar containing 0.6% yeast extract
Trypticase soy broth

Trypticase soy broth containing 0.6% yeast extract
units

ultra-high temperature

United States Department of Agriculture

violet red bile agar

World Health Organization

xylose lysine deoxycholate

xii

CHAPTER 1

INTRODUCTION

1.1 Foodbome Illness in the United States

Foodbome illness is a considerable public health problem in the United States.
An estimated 76 million cases of foodborne disease occur annually, with 325,000
hospitalizations and 5,200 deaths (Mead et al. 1999). Six to 33 million cases of
foodborne illness are thought to be bacterial in origin, with Salmonella and
Campylobacter accounting for almost 82% of these cases (CDC 2001a). Salmonella and
Listeria, along with T oxoplasma, are responsible for more than 75% of the 5,200 deaths
caused by known foodborne pathogens (Mead et al. 1999). The Economic Research
Service estimates that the medical costs, productivity losses and lost wages, and value of
premature death for the five major pathogens (Campylobacter, Salmonella, Escherichia
coli 0157:H7, E. coli non-0157zH7 STEC, and Listeria monocytogenes) total at least
$6.9 billion annually (Crutchfield 2001). Salmonella alone has an estimated annual cost
of $2.4 billion (Frenzen 2002). Many sectors of the economy are burdened with the costs
of foodborne illness. Those who become ill and their families, as well as other parties,
including employers, insurers and taxpayers, generally bear the brunt of these costs
(Buzby et al. 2001).

The cost estimate presented for foodborne illness does not include data on product
recalls due to bacterial contamination. For instance, from 1998 to 2001, the United States
Department of Agriculture Food Safety and Inspection Service (USDA-FSIS) recalled

over 3,250,000 pounds of meat and poultry due to Salmonella contamination (F SIS

2002a). The meat industry is particularly concerned with compromised product safety as
a result of L. monocytogenes, E. coli 01 S7:H7, Salmonella and Campylobacter, although
other microbial hazards (e. g., Clostridium perfi'ingens, Closlridium botulinum,
staphylococcal enterotoxins) are also important.
1.2 The Meat Industry and Relevant Regulations

Microbial contamination of meat and poultry products is paramount when
considering the staggering impact of the meat and poultry industry on the United States
economy. A US. Senate resolution designating the first annual National Meat Week in
1984 described the meat industry, with $70 billion in annual sales, as the “largest single
component of US. agriculture” (Romans et al. 2001). The impact of the meat and
poultry on the agricultural industry has remained virtually unchanged since 1984,
although the annual sales are now well over $90 billion (AMI 2000, ERS 2001, Jones
2001, Romans et al. 2001). Red meat and poultry production is continuing along a record
setting pace (Southard et al. 2000). Worldwide, the United States leads in beef and veal
production and total consumption, placing second only to Argentina in per capita
consumption. The poultry record is even more startling, with the United States a clear
leader in broiler chicken production, exports, and total and per capita consumption
(Romans et al. 2001). Red meat and poultry consumption per capita was 213 and 220 lb
in 2001 and 2002, respectively. Consumption of poultry meat alone has soared
drastically in the last two decades to 95 and 99 lb per person in 2001 and 2002,
respectively (ERS 2002).

In this same timeframe, however, per capita consumption of red meat has actually

decreased. Although pork consumption has remained fairly constant, beef consumption

has not (Romans et al. 2001). The marketing of beef has not changed significantly over
the last twenty years, compounded by the poultry industry’s greater ability to cater to the
definitive trend toward convenience foods. It is estimated that three-quarters of all
women aged 25-54 in the United States are now in the work force, as opposed to only
about one-half twenty years ago. More households have two working adults, allowing
for less time to Spend preparing meals-a fact that has helped direct the shift toward
convenience foods (Carmel 2000).

Pork consumption has remained fairly constant, partly because so many processed
meats, including sausage and cold cuts, are made with pork. Likewise, the dramatic
increase in poultry consumption can in part be attributed to the use of more chicken and
turkey in fiIrther processed meat items (Romans et al. 2001). In fact, over sixty percent
of all turkey meat produced in the United States is used in the production of convenience
items (Romans et al. 2001). The trend towards convenience is perhaps best illustrated by
the growth of heat-and-eat meals, such as frozen dinners with meat as the major
contributor to the meal. Pre-cooked meat patties and luncheon meats, including bologna,
frankfurters and meat loaves, have also benefited from the increased use of convenience
foods (Pearson and Gillett 1996).

The United States Department of Agriculture (USDA) defines ready-to-eat (RTE)
meat and poultry products as those products that have been appropriately processed so as
to be safely consumed without further preparation by the consumer. Such products can
be either shelf-stable or non-shelf-stable (i.e., requiring refrigeration to prevent growth of
spoilage or pathogenic organisms). USDA specifically divides RTE meat and poultry

products into five categories: dried products (e.g., jerky), salt-cured products (e. g.,

prosciutto), fermented products (e.g., pepperoni), cooked and otherwise processed
products (e. g., ravioli, turkey franks), and thermally-processed, commercially sterile
products (e.g., canned soups with meat or poultry) (F818 2001). RTE meat and poultry
products are purchased because they offer flavor appeal, nutritional value, convenience
and variety to the diet at a price that is competitive with fresh meats and poultry
(Tompkin 1986). However, RTE and partially heat-treated meat and poultry products can
present a safety problem from a microbiological standpoint. Emphasis is particularly
placed on RTE products, as consumers are less likely to reheat these products to
sufficiently kill pathogenic microorganisms such as L. monocytogenes and Salmonella
spp. before cooking (Levine et al. 2001). Additionally, while pre-cooked meat patties
and other items are widely consumed by the general public, they are particularly popular
for school lunch programs, hospitals and other institutional settings where increased
numbers of susceptible individuals are present (Romans et al. 2001). However, USDA-
FSIS does have certain regulations in place to reduce and eliminate product
contamination.

The United States has been inspecting meat and poultry products since 1891,
beginning with certain salted pork and bacon items in response to European fears of
trichinosis. Federal meat and poultry inspection and regulations expanded throughout the
next century. Major milestones included the 1967 Wholesome Meat Act and the 1968
Wholesome Poultry Act requiring that all carcasses and meat products be inspected. By
the mid-1990’s, FSIS had over 7,400 inspectors in 6,200 slaughter and processing plants.
The system these inspectors were operating under, however, did not adequately target and

reduce microbial pathogens (Crutchfield et al. 1997). To help remedy this situation, in

1996 F SIS began phasing in sanitation standard operating procedures (SSOPs), Hazard
Analysis and Critical Control Points (HACCP) procedures, and microbial testing
(Crutchfield et al. 1997).

Most recently, FSIS has finalized new regulations for the production of certain
fully and partially cooked meat and poultry products (FSIS 1999). The agency has
likewise recently proposed similar regulations for RTE and all partially heat-treated meat
and poultry products (FSIS 2001). In both instances, the changes signal a move away
fiom the command-and-control nature of the previous requirements. Rather than dictate
step-by-step processing methods with prescribed endpoint temperatures, the regulations
are in the form of lethality, stabilization and handling performance standards. In
particular, these standards establish lethality levels for pathogen reduction as well as set
limits for pathogen growth. Lethality itself is defined as the required reduction in the
number of specific pathogenic microorganisms. A 6.5-log reduction of Salmonella in
cooked beef, roast beef and cooked corn beef and a 7 .0-log reduction of Salmonella in
certain fully and partially cooked poultry products are required. A comparable reduction
is proposed for remaining RTE meat and poultry products.

Alternately, processors may choose to achieve an “equivalent probability that no
viable Salmonella organisms remain in the finished product” (F SIS 1999). Regardless,
the new regulations allow processors a certain measure of flexibility accompanied by the
opportunity to implement new, cost-efl'ective technology and customized processing
procedures. At the same time, however, the importance of demonstrating and validating
the lethality of a given process becomes greater, particularly if the process meets

alternative lethalities. It is critical to document the relationship between lethality (and

lethality treatments) and product characteristics, in part because processing standards are
based largely on thermal inactivation studies performed in a laboratory and not with
actual meat products in a processing microenvironment. It also becomes critical to
document the relationship between lethality and microorganism characteristics, including
strain and sublethal injury. Of particular concern, several antibiotic resistant and heat
resistant microorgansirns have emerged. Also, several environmental conditions,
including starvation and cold shock, have either been proven or are suspected to alter
thermal inactivation kinetics of pathogens.

Current regulations have been subject to certain criticisms. Many feel that testing
food for pathogens will not necessarily make food safer. However, according to the
American Public Health Association (APHA), a group that represents over 50,000 public
health professionals in the United States and abroad, “people manage what they measure”
(Levinson 2000). Although microbial testing does not suffice as a control for HACCP,
microbial testing at appropriate Critical Control Points is imperative to verify the
effectiveness of preventative measures and critical limits (Levinson 2000). The APHA
and other groups particularly believe that requirements for microbial testing in the form
of performance standards accomplish the best balance between public and private
intervention and control in food safety. It is not the fundamental role of FSIS to dictate
how to exclude pathogens from meat products, but rather to test and affirm that these
products meet the established standards. At the same time, enforcing performance
standards allows technological innovation, creativity and advancement to stay in the

hands of the private sector, where most experts agree it belongs (Levinson 2000).

Proposed and finalized F SIS lethality standards reflect the destruction of specific

“reference” organisms whose elimination or reduction will ordinarily likewise indicate
the crucial elimination or reduction of other pathogenic organisms. Except for thermally-
processed, commercially sterile products, the lethality performance standards are based
on Salmonella. Salmonella was chosen as the target organism for a number of reasons. It
is associated with raw meat (poultry, beef and pork) and is responsible for a high
incidence of foodborne illness, which can often be severe. Also, the conditions required
for an appropriate reduction in Salmonella will simultaneously reduce most other
vegetative pathogenic microorganisms. In particular, if the given reduction of
Salmonella organisms is achieved, a safe reduction of T richinella spiralis,
Staphylococcus aureus and E. coli 0157:H7 should also be met, because these organisms
are typically less heat resistant than Salmonella. Although more heat resistant than
Salmonella, L monocytogenes, will most likely enter the product as a post-processing
contaminant, due to improper or inadequate sanitation, rather than as a result of
insufficient processing. To verify compliance with performance standards (e.g., via
challenge studies), F SIS suggests the use of a Salmonella cocktail comprised of relatively
heat resistant strains or serotypes as well as those implicated in outbreaks (FSIS 2001).
1.3. Justification

The status of the most recent regulation is in limbo. It is not finalized and FSIS is
requesting additional scientific information and data. The policy or belief of FSIS is that
the regulations must be based on “sound science and common sense measures that
involve significant public comment” (FSIS 2001). The research presented here aims to
serve as such, helping to answer some of the pressing questions from the regulation for

processed products, especially those related to bacterial injury and lethality.

Stressed cells behave differently fiom cells grown under optimal conditions in the
laboratory. Unfortunately, it is typically these latter cells that are used in food safety
studies. The possibility that preservation methods based on such studies may not be
sufficient to ensure the safety of processed food becomes a major cause for concern (Lou
and Yousef 1996).

An additional concern in the design of inactivation studies involving foodborne
pathogens is that most studies used various laboratory media rather than food menstra
(Farber and Brown 1990). Food components offer a selective advantage to
microorganisms for enhanced heat resistance (Van Schothorst and Maggie Duke 1984,
Mackey and Derrick 1987b, F arber and Brown 1990, Murphy et al. 1999). Sublethal
stress studies must also consider the importance of using actual foods. In fact, initial
studies by Mackey and Derrick (1987b) showed that the effect of heat shock was
accompanied by increased heat resistance conferred on Salmonella ser. Thompson by the
composition of the heating medium. Thus, studies designed to compare pathogen stress
and inactivation in real foods, specifically meat products, to that in liquid media (Murphy
et al. 1999) should be conducted to augment current data with information that may be
more directly applicable to real commercial processes.

1.4 Objectives and Hypothesis

After developing sublethal heat, cold and starvation treatments for an 8-strain
Salmonella cocktail, the objectives of this research were to (1) assess the effect of media
supplementation on the recovery of sublethally injured Salmonella and (2) investigate the
effects of pre-stress treatments on the thermal inactivation kinetics of Salmonella. Given

the physiological differences between injured and uninjured microorganisms, the

hypothesis is that a prior heat, cold or starvation stress will affect the recovery and

thermal resistance profiles of Salmonella.

CHAPTER 2

LITERATURE REVIEW

2.1 Salmonella

Salmonella spp. are facultatively anaerobic, small, Gram-negative, nonsporing
rods belonging to the family Enterobacteriaceae (Table 2.1). Currently, there are at least
2,463 serovars (serotypes) of Salmonella (Brenner et al. 2000). The major antigens for
serotyping are somatic (O lipopolysaccharides of the bacterial outer membrane), flagellar
(H antigens) and capsular (Vi antigens occurring in only three Salmonella serovars) (Le
Minor 1992, D’Aoust 1997). Antigenic structure has applications in diagnostics,
identification and confirmation, as well as classification and nomenclature.

Salmonella taxonomy and nomenclature are complex and have undergone several
significant changes in recent years. Most state and local public health agencies follow
the nomenclatural system used by the Centers for Disease Control and Prevention (CDC).
This system, based on recommendations from the World Health Organization (WHO)
Collaborating Centre for Reference and Research on Salmonella (Institut Pasteur, Paris),
divides the genus Salmonella into two species, S. enterica and S. bongori, each
containing multiple serotypes (see Table 2.2). S. enterica, the so-called type species, is
divided into six subspecies differentiated by biochemical reactions, by genomic
relatedness and by the Kauffman-White serotyping scheme. Although the subspecies are
differentiated, they do form a single DNA hybridization group. The majority of all
serotypes fall within S. enterica subsp. I (S. enterica subsp. enterica). Within this group,

CDC uses names for serotypes (e.g., Enteritidis, Typhimurium, Heidelberg) and follows

10

Table 2.1. Biochemical characteristics of a typical Salmonella isolate (D’Aoust 1997,
Madigan et al. 1997, Bell and Kyriakides 2002).

 

Biochemical Characteristic Reaction

 

HZS

Catalase

Citrate

Indole

Lactose

Lysine Decarboxylase
Ornithine Decarboxylase
Oxidase '
Urease '
+ = positive reaction; - = negative reaction

-+++

++-

 

Table 2.2. Species, subspecies and serovars of the genus Salmonella (adapted from
Brenner et al. 2000).

 

Species No. of Natural habitat
86mm

Salmonella enterica

 

 

I. subsp. enterica ................... 1454 warm-blooded animals
II. subsp. salamae ................... 489
Illa. subsp. arizonae .................. 94
IIIb. subsp. diarizonae .............. 324
IV. subsp. houtenae .................. 70 cold-blooded animals
VI. subsp. indica ...................... 12 and the environment3
Salmonella bongori ..................... 20
Total ........................... 2463

 

aIsolates from all subspecies have been recovered fi'om human sources.

11

the naming format of the WHO Collaborating Centre (see Table 2.3) (Brenner et al. 2000,
D’Aoust 1997, Le Minor 1992).

For purposes related to epidemiological investigation, salmonellae can be divided
into three groups (Jay 1998, Poppe 1999):

l. Serovars infecting humans only. These include S. ser. Typhi, the
causative agent of typhoid fever, and the paratyphoid salmonellae.
Typhoid and paratypho id fevers are the most severe diseases caused by
salmonellae (Jay 1998).

2. Host adapted serovars including S. ser. Abortusequi (horses), S. ser.
Abortusovis (sheep), S. ser. Choleraesuis (swine), S. ser. Dublin
(cattle) and S. ser. Gallinarum and S. ser. Pullorum (poultry). Several
organisms in this group are human pathogens and potentially can be
contracted fiom foods (Jay1998, Poppe 1999).

3. Unadapted serovars with no host preference. The largest group, these
serovars may be pathogenic for both humans and animals. This group
includes most foodborne salmonellae and is most commonly
responsible for foodborne infections in human adults (Jay 1998,

Le Minor 1992).

Many different serotypes are responsible for gastroenteritis. Table 2.4 gives the
serotypes most commonly isolated fiom human sources between 1996 and 2000. In
2000, the top three Salmonella serotypes (Typhimurium, Enteritidis and Newport)
represented 51% of all isolates. AS in other years, most isolates (25% in 2000) were from
children under 5 years of age (CDC 2001b). In 1999, 31% of human S. Typhimurium
isolates received by the National Antimicrobial Resistance Monitoring System (N ARMS)
for Enteric Bacteria were resistant to ampicillin, chloramphenicol, streptomycin,
sulfonamides, and tetracycline, the multi-drug resistance pattern associated with isolates
of S. ser. Typhimurium belonging to phage type DT104 (Ribot et al. 2002).

Symptoms resulting from infection with nontyphoid salmonellae generally appear

8 to 72 hours after exposure and include diarrhea, nausea, fever and abdominal cramps.

12

Table 2.3. Salmonella nomenclature recommended by the WHO Collaborating Centre for

Reference and Research on Salmonella and used at the CDC (adapted from Brenner et al.
2000).

 

 

 

 

 

 

 

 

 

 

Nomenclature
Genus (italicized) Salmonella
Species (italicized) 2 enterica
o bongori
Serotype (capitalized, not
italicized) (The serotype name should
be preceded by the word
“serotype” or “ser.” the first
time it is mentioned in the
text.)
Examples
S. entericaa subsp. enterica ser. Typhi S. ser. Typhi (S. ophi also
used)
S. entericaa subsp. enterica ser. Typhimurium S. ser. Typhimurium (S.
typhimurium also used)
S. entericaa subsp. enterica ser. Hadar S. ser. Hadar

 

aS. Choleraesuis or S. enteritidis may also be used

13

Table 2.4. Predominant clinical serovars of salmonellae in the United States between
1996 and 2000‘.

 

Year Total Rank Serotype (%)

 

1996 39,035 Enteritidis (24.5)
Typhimurium. (24.3)
3 Heidelberg (5.1)
Newport (5.1)
Montevideo (3.1)
Javiana (1.9)

NI—

kit-h

Typhimurium. (26.3)
Enteritidis (22.9)
Heidelberg (6.1)
Newport (4.6)
Agona (2.1)
Montevideo (2.1)

1997 34,608

{It-RUJNv—t

Typhimurium. (26.0)
Enteritidis (17.7)
Newport (6.7)
Heidelberg (5.6)
Javiana (3.4)

1998 33,971

LII-bbJNr-t

Typhimurium. (24.6)
Enteritidis (16.3)
Newport (8.0)
Heidelberg (5.5)
Muenchen (4.1)

1999 32,782

M-BUJNi—t

2000 32,022 Typhimurium. (22.1)
Enteritidis (19.4)
Newport (9.3)
Heidelberg (5.2)

Javiana (3.6)

Ut-hbJNH

 

‘Typhimurium includes var. Copenhagen
aData compiled by the Public Health Laboratory Information System (CDC 2001b, 2000,
1999, 1998, 1997a).

14

Although acute symptoms can last for weeks, the self-limiting condition is most typically
resolved within 4 to 7 days. Successful treatment of patients without complications may
only require fluid and electrolyte replacement. However, septicemia and other secondary
complications such as reactive arthritis have been reported, particularly among highly
susceptible individuals (D’Aoust 1997, F renzen et a1. 1999).

Salmonellosis consistently ranks at or near the top of the list of foodborne
diseases. The CDC estimates 1.4 million cases of salmonellosis, 95% of which are
foodborne, occur annually in the United States (Frenzen et al. 1999). The overall
incidence of diagnosed Salmonella infections per 100,000 people was 15.1 in 2001 (CDC
2002a) and 14.4 in 2000 (CDC 2001c). The Healthy People 2010 national health
objective for Salmonella is 6.8 cases per 100,000 people (CDC 2002a). The estimated
annual economic cost (medical costs, productivity losses, lost wages, and value of
premature death) of Salmonella is $2.4 billion (Frenzen 2002).

Outbreaks associated with Salmonella have been reported in a variety of foods,
including meat, poultry, cantaloupe, cheese, ice cream and chocolate (D’Aoust 1997).
Specific examples of salmonellosis outbreaks are numerous. For instance, S. ser. Hadar
was identified as the causative agent in a 1999 Georgia outbreak that sickened at least 58
people who ate barbecued pork at a local restaurant. Stool specimens and barbecue
samples from the restaurant and patient’s homes had identical pulsed-field gel
electrophoresis (PF GE) patterns (Toomey et al. 2001). Other examples include a 1998
outbreak of S. ser. Heidelberg in turkey that sickened at least 25 people at a New York
church luncheon, a 1997 outbreak of S. ser. Reading in corned beef that sickened 130

New Jersey restaurant-goers (CDC 1997b), three outbreaks between 2000 and 2002 of S.

15

ser. Poona in cataloupe that sickened 155 people in the United States and Canada (CDC
2002b) and 241 confirmed egg-associated S. ser. Enteritidis outbreaks between 1999 and
2001 (CDC 2003). Outbreaks caused by different Salmonella serotypes occur in a wide
variety of establishments, including private homes, hospitals, and restaurants.

The primary habitat of Salmonella is the intestinal tract of humans and animals
(Le Minor 1992). The ubiquity of Salmonella within the natural environment, along with
intensive husbandry practices employed in the meat, fish and shellfish industries has
helped foster the continued importance of this human pathogen in the global food chain
(D’Aoust 1997). Poultry and pigs are well established as major reservoirs for
sahnonellae, with cattle being a primary source of S. Typhimurium (Gilbert and
Humphrey 1998).

According to Gilbert and Humphrey (1998), meat may come in contact with
salmonellae through an infected animal, an animal slaughtered near an infected animal,
other meat on a contaminated work surface, or fi'om processors/employees with
salmonellosis. Since most Salmonella infections are asymptomatic in animals, one way
their prevalence may be reflected is in the frequency with which the micoorganism can be
detected in products of animal origin (Gilbert and Humphrey 1998).

The presence of salmonellae in the poultry, pork, beef and sheep industries stems
from the exposure of animals to environmental sources of contamination, contaminated
feeds, and parental transmission of infection. However, the fi'equency with which
salmonellae are detected in meat varies with location, season, and species of animal
(D’Aoust, 1997, Gilbert and Humphrey 1998). Of the various sectors within the meat

industry, poultry products remain the foremost reservoirs of salmonellae in many

16

countries, largely dominating other meat products as possible vehicles of infection
(D’Aoust 1997). In 2001, the prevalence of Salmonella in ground turkey was 26.2% and
19.5% in ground chicken, compared to 2.8% in ground beef (FSIS 2002b).

Salmonella spp. can be contracted from foods of animal origin and foods
contaminated with animal feces (Shallow et al. 2000). Although contaminated poultry
meat and poultry products (e.g., eggs) may dominate this group, any foodstuff can
conceivably be a vehicle for human salmonellosis. This is often a result of cross-

contamination, as has been documented with fruits and vegetables (D’Aoust 1997).

2.2 Thermal Inactivation of Salmonella

The decimal reduction time (D-value) is the time required to reduce a microbial
population by 90%. Historically, it has been used in the literature to assess the death rate
or heat resistance of microorganisms, with many intrinsic and extrinsic factors
influencing heat resistance.

Serotypes and strains of Salmonella are known to exhibit different degrees of heat
resistance. In egg products, for example, certain strains of S. Enteritidis are often more
heat resistant than S. Typhimurium (Doyle and Mazzotta 2000). A D-value of 4.09 min
for S. Enteritidis ME-14 in liquid whole egg at 567°C was significantly greater than a D-
value of 3. 13 min for S. Typhimurium DT104 (Brackett et al. 2001). However, while the
D-value of S. Enteritidis PT34 in liquid egg yolk at 56°C ranged from 5.14 to 7.39 min,
S. ser. Seltenberg had a D-value of 19.96 min (Chantarapanont et al. 2000).

S. Seftenberg is well known for its remarkable heat resistance in culture media

and various food products (Doyle and Mazzotta 2000). In ground chicken breast meat

17

heated to 675°C, a 1.10-log reduction in S. Seftenberg was observed as opposed to a >6—
log reduction in S. Heidelberg, S. ser. Mission, S. ser. Montevideo and S. ser. California
(Murphy et al. 1999). After heating for five minutes at 60°C, no viable cells of S.
Typhimurium Tm—l remained in ground chicken pectoral muscle initially inoculated at
108 CFU/g. However, an exposure of 10 to 15 min at 65°C was required to kill an equal
number of cells of S. Seftenberg 775W (Bayne et al. 1965). S. Seftenberg 775W is often
used as a test organism in validation studies. If a thermal process effectively reduces S.
Seitenberg 775W, it is expected to be as effective or more so against salmonellae that are
more commonly encountered in foods (Doyle and Mazzotta 2000).

Bacteria attached to meat tissue or surfaces are generally more heat resistant than
those dispersed throughout a food or broth or those suspended in liquid media (Murphy et
al. 2002, Doyle and Mazzotta 2000). Humphrey et al. (1997) reported that D-values at
58°C for S. Typhimurium DT104 in pork muscle tissue were >10 min for attached cells
versus only 2 min for free cells. Additionaly, growth and inactivation of microorganisms
in food and model systems may be quite different than in liquid media. For example,
Murphy et a1 (1999) found that a six strain Salmonella cocktail was more heat resistant in
ground chicken breast patties than in an agar-water solution. A 7-log (CFU/g) reduction
of Salmonella in agar-peptone occurred 19% faster than in chicken meat. Additionally,
Quintavalla et al. (2001) found that the heat resistance of eight strains of Salmonella (S.
Typhimurium strains ATCC 14028, 133 and 1116, S. ser. Derby B4373, S. ser. Potsdam
1133, S. ser. Menston 179. S. ser. Eppendorf 166, and S. Kingston 1124) was 1.5-4 times
higher in pork meat as opposed to Trypticase soy broth While published data comparing

the heat resistance of Salmonella in actual meat products versus liquid media is limited,

18

variations in food composition and culture media contribute to differences in thermal
inactivation

Certain food components such as fat offer protection against heat treatments (Jay
1998, Hansen and Riemann 1963, Line at al 1991, Ahmed et al. 1995). Maurer (2001)
found that increasing the level of fat in turkey increased the D-value of S. Seftenberg.
Juneja and Eblen (2000) observed significantly higher D-values at 58°C for an eight
strain S. Typhimurium DT104 cocktail in ground beef containing 7, 12, 18 and 24% fat
than in chicken broth containing 3% fat. They suggested that these differences could be
due to poor heat transfer through the heating matrix as a result of the high fat levels.
Thermal inactivation in ground beef was non-linear, and although the absolute D-values
of the S. Typhimurium DT104 cocktail at 58°C, 60°C, 625°C and 65°C in ground beef
decreased as the fat content increased, lag times increased substantially. Considering the
combination of lag time and D- value, 7.07 min at 65°C was required to reduce S.
Typhimurium DT104 by 7 logs in ground beef containing 7% fat, as opposed to 20.16
min in ground beef containing 24% fat (Juneja and Eblen 2000). The thermal protection
afforded by increased fat levels may be due to a localized absence of moisture (reduced
water activity) within bacterial cells (Hansen and Riemann 1963, Jay I998, Juneja and
Eblen 2000).

Water activity (aw) is defined as the ratio of the water vapor pressure of a food
substrate to the vapor pressure of pure water at the same temperature (Jay 1998). In
practice, it can be interpreted as the moisture available in foods for microbial growth. It
is believed that proteins are more stable in a drier state; therefore, a greater input of

energy is required to denature proteins when the moisture content is decreased (Hansen

19

and Riemann 1963). As a result, a decrease in water activity is generally associated with
a corresponding increase in the thermal resistance of microorganisms (Doyle and
Mazzotta 2000, Carlson 2002, Jay 1998, Hansen and Riemann 1963). Archer et al.
(1998) heated S. ser. Weltevreden in flour with hot air at temperatures ranging fi'om 57°C
to 77°C. As the initial aw increased over a range of values fi'om 0.2 to 0.6, D-values
decreased for all temperatures tested. The effects of aw on heat resistance may depend on
the solute used (Mattick et al. 2001, Moats et al. 1971). Mattick et al. (2000) found that
reducing the aw of Trypticase soy broth with glucose-fluctose more effectively increased
the thermal resistance of S. Typhimurium DT104 than did glycerol or NaCl.

Acid adaptation not withstanding, microorganisms tend to be most heat resistant
at their optimum pH for growth (Jay 1998). Casadei et al. (2001) found that a drop in pH
from 7 to 3 caused a 100-fold reduction in S. Typhimurium D-values at temperatures
between 48°C and 54°C, and Juneja and Eblen (1999) reported an increase in the heat
sensitivity of L. monocytogenes at 55°C and 60°C with increasing acidity. A theoretical
explanation for this effect of pH is that the logarithm of the rate of heat-destruction
increases linearly in the acidic and alkaline ranges but has a minimum at the optimum
growth pH (Juneja and Eblen 1999, Reichart 1994). However, Leyer and Johnson (1993)
found that acid-adapted (pH 5.8) cells of S. Typhimurium were more thermally tolerant
than nonadapted cells, exhibiting lO-fold more survivors after 20 min at 50°C, 10 min at
55°C and 2.5 min at 57.S°C. Acid adaptation includes two phases, pre—shock and acid
shock, that alter the expression of at least 70 proteins and are required for protection
against extreme acid stress (Leyer and Johnson 1993). Acid shock and adaptation may

fall under the phenomenon of sublethal injury and cross-protection.

20

2.3 Sublethal Injury and Stress
2.3.1 Definitions

The simplest definition of injury is the effect sublethal treatments have on
microorganisms (Hurst 1984). By extension, sublethal injury is a consequence of
exposure to a chemical or physical process that damages but does not kill a
microorganism (Russell 1984, Hurst 1977). Yousef and Courtney (2003) include damage
to cellular components in their description of injury and according to Gilbert (1984),
“Sublethal injury of microorganisms implies damage to structures within the cells, the
expression of which entails some loss of cell function that may be transient or
permanent.” Rather than produce only viable or dead cells, pathogen and spoilage
controls produce a continuum of effects, and a considerable proportion of
microorganisms in foods likely harbors sublethal lesions (IFT 2002, Mossel and van
Netten 1984, Zhao and Doyle 2001).

The term stress has also been used to illustrate the effect of sublethal treatments.
However, Hurst (1984) considers injury to be the preferred term as, by analogy with
higher organisms, its description evokes an image of “temporary and repairable physical
damage.” By similar analogy, the term stress apparently carries a more subtle meaning,
not necessarily causing physical damage but able to alter the behavior of an individual.
Current literature pertaining to microbial injury typically does not maintain this
distinction, and the terms are used interchangeably.

The term stress, however, is universally used in reference to the agents or
treatments causing injury. There is a tendency to perceive food as a fi'iendly environment

for bacteria. More than likely, the food itself is an unfi'iendly environment compromised

21

by competition, lack of moisture, acids and other by-products (Archer 1996), and almost
all microorganisms in the food have been subject to certain environmental and processing
stresses (Tables 2.5 and 2.6). The stresses organisms in a food environment can be
subjected to vary broadly fiom acid shock, oxidants, salt and starvation to heat shock,
freezing and thawing (Miller et al. 2000). In addition, certain emerging technologies
(e.g., high hydrostatic pressure) cause sublethal injury, although others (e.g., pulsed
electric field) do not (Wuytack et al. 2003, Yousef and Courtney 2003). Potential
stresses generally fit into three categories (physical, chemical or nutritional), can occur at
several stages of the food chain (pre-harvest, processing, distribution, storage) and cause
a wide range of damage to the bacterial cell.

Storz and Hengge-Aronis (2000) also include as a definition of stress any
departure from optimal conditions that has the potential to decrease growth rate. Any
situation that induces the expression of genes recognized to respond to specific
environmental conditions may also define a stressful situation and may be important in
the effect of bacterial stress responses on host-pathogen interactions.

Given that the perception of a stressful situation is coordinately linked to
characteristics of the individual cell, merging the above definitions of stress into one
universal definition is difficult. However, the working definition from the standpoint of
food microbiology is typically taken to be a physical or chemical (or nutritional)
condition insufficiently severe to kill, resulting in sublethally injured microbes (Hurst
1977, Murano and Pierson 1993). The physical and chemical conditions can be
processing treatments, but they may also be environmental conditions, such as limiting

nutrients in water.

22

Table 2.5. Physical and chemical treatments (stresses) with the ability to reversibly injure
microorganisms (Ray 1989, Ray 2001, Wuytack et al. 2003, Thanomsub et al. 2002).

 

Physical:

 

Drying-including air drying and freeze-drying
Heat-particularly sublethal heating processes

High hydrostatic pressure

Low temperatures-including both refrigeration and freezing
Pulsed white light

Radiation-gamma, ultraviolet and X-ray
Solids-concentrations of sugars, salts

 

Chemical:

 

Chemical sanitizers-chlorine, iodine, quaternary ammonium compounds
Oxidative treatments-including ozone

pH-alkali and acids (organic and inorganic)

Preservatives-sorbate, benzoate, nitrate, nisin, etc.

 

Table 2.6. Microorganisms known to experience reversible injury fi'om sublethal stresses
that are of importance in food systems (Ray 1989, Ray 2001).

 

 

Microorganism Main Importance in Food
Escherichia coli Indicator, Pathogenic
Enterobacter aerogenes Indicator

Klebsiella sp. Indicator, Pathogenic
Streptococcus faecal is Indicator

Salmonella sp. Pathogenic

Listeria monocytogenes Pathogenic

Shigella sp. Pathogenic

Vibrio parahaemolyticus Pathogenic

Yersinia enterocolitica Pathogenic
Campylobacterjejuni Pathogenic

Staphylococcus aureus Pathogenic

Clostridium perfi'ingens Pathogenic

Clostridium botulinum Pathogenic

Pseudomonas sp. Spoilage

Bacillus sp. Spoilage, Pathogenic
Lactococcus lactis Bioprocessing (Fermentation)
Lactobacillus bulgaricus Bioprocessing (Fermentation)
Lactobacillus acidophilus Dietary adjunct (Probiotic)
Saccharomyces cerevisiae Bioprocessing (Fermentation)
Candida sp. Spoilage

Aspergillusflavus Spoilage

 

23

Storz and Hengge-Aronis (2000) believe there are different levels of stress
severity, from minor to severe, extreme and eventually lethal. With minor stress,
bacterial cells adapt completely to the changed conditions, and growth rate is not
affected. Low levels of stress may cause a transient adaptation (adaptive response)
accompanied by a temporary change in physiology that often results in increased stress
tolerance (Yousef and Courtney 2003). At the other extreme, lethal stress can cause the
death of some but not necessarily all bacterial cells. The occurrence of death in only a
certain fraction of the population indicates that lethal stress may induce responses or even
adaptive mutations that may actually improve survival of the overall population (Storz
and Hengge-Aronis 2000, Archer 1996). Moderate stress may result in injury, and there
is also thought to be a continuum of injury, ranging from mild to severe along with
healthy and dead cells (Hurst et al. 1976, Hurst 1984, Mackey 2000, Stephens et al.
1997). The relationship between the different stress levels and degrees of injury, as well
as adaptation, is not well defined. For practical purposes, sublethal injury is essentially
anything less than death and, at some point, injured cells have likely undergone some
type of stress adaptation.

The expression of sublethal injury can be manifested in several ways in the
laboratory (Hurst 1977). Metabolic injury is described by the inability of bacterial cells
to grow on defined minimal media (Gilbert 1984). This inability is often temporary,
indicating repair of injury can occur (Hurst 1984). Structural injury is viewed as the
inability to proliferate or survive in media containing selective agents (e.g., bile salts,
tellurite, sodium desoxycholate, bismuth sulfite, sodium chloride, crystal violet, brilliant

green) with no obvious inhibitory effects on unstressed cells (Hurst 1977, Gilbert 1984,

24

Hurst 1984, Semanchek and Golden 1998, Ray 2001). Depending on other conditions
(e.g., enrichment, incubation conditions, etc.), the inability to grow on selective media is
also temporary, which once again indicates that repair of injury is imminent.

Some authors consider metabolic and structural to be different degrees of the
same injury (Hurst 1984, Kang and Siragusa 1999). Both are discemable by the inability
to form visible colonies under given conditions, making injury recognizable by the loss of
characteristic growth capabilities regardless of the method of evaluation (Hurst 197 7,
Busta 1976). However, it seems that while all injured cells suffer structural damage,
which upsets the permeability barrier in the cell wall and membrane, metabolic injury is
further damage to various functional components of the cell (Brashears et al. 2001).
Consequently, growth on selective media, or the lack thereof, is the primary method of
assessing sublethal injury (Hurst 1984, Gilbert 1984, Czechowicz et al. 1996, Semanchek
and Golden 1998, Brashears et a1. 2001). The difference in plate counts between
selective and non-selective media is used to quantify sublethal injury as a proportion or
percentage of the entire population (Kang and Siragusa 1999, Semanchek and Golden
1998, Baylis et al. 2000a, Dickson and Frank 1993, Restaino et al. 1980, Brashears et al.
2001).

2.3.2 Specific Stresses

Abrupt acid shock or gradual acid stress can occur in low pH conditions when H4r
ions cross the bacterial cell membrane and create an acidic intracellular pH. Likewise,
unprotonated organic acids can diffuse across the bacterial cell membrane and lower the
internal pH upon dissociation (Foster 2000, Abee and Wouters 1999). While acid injury

is thought to be the result of an easily reversible pH equilibrium shift, it is more complex

25

than simple neutralization of the acidic conditions (Przybylski and Witter 1979).

Acid stress can occur with fermentation (e. g., summer sausage) or the addition of
preservatives such as acetic and pr0pionic acids (Smith and Palumbo 1978, Abee and
Wouters 1999). Current recommendations for the use of acid washes to decontaminate
beef carcasses include concentrations of lactic acid at 1%-2.5% (Gerris 2000). Dickson
and Siragusa (1994) found that washing beef tissue with 1% lactic or acetic acid
sublethally injured S. Typhimurium. Also, an alkaline stress can occur under high pH
conditions. Many detergents and chemical sanitizers, such as caustic soda (NaOH) and
ammonium compounds, used to clean food processing facilities, including food contact
surfaces, are alkaline in nature (Taormina and Beuchat 2001).

Starvation stress can occur on animal carcasses, in food, on equipment surfaces,
on walls and floors, and in water (Lou and Yousef 1996, Dickson and Frank 1993).
Dickson and Frank (1993) define starvation stress as the survival of bacteria in
oligotrophic environments with low or no available nutrients to provide for growth and/or
reproduction. Typically, natural environments are places with limiting amounts of
nutrients and rapidly changing nutrient availability (Hengge-Aronis 1993, Givskov et al.
1994). Bacteria are present in different physiological states, including exponential
growth, slow growth or stationary phase and death. Of the stages or types of growth,
cryptic must also be considered. Cryptic growth is the phenomenon where dead
organisms provide nutrients for the multiplication of survivors (Postgate and Hunter
1963). Cryptic growth may contribute to the survival of populations well after grth
has ceased.

Salmonella has been reported to survive in cold storage at 5°C for up to 8 months

26

(Jeffreys et al. 1998, D’Aoust et al. 1985). This is likely due to cold shock and
subsequent cold adaptation. The cold shock phenomenon occurs when growing bacteria
are suddenly chilled, for instance fi'om 37°C to 15°C, 10°C or lower, and it has been
established that rather small temperature drops (e. g., from 37°C to room temperature) can
also generate cold shock in susceptible organisms (Mackey 1984, Jones et al. 1996). The
associated cold shock response is divided into stages of initial cessation of growth,
resumption of growth after an adaptive period and changes in protein synthesis (Miller et
al. 2000). Microorganisms inhabiting foods that must be refrigerated for pre and/or post-
processing storage are subject to cold shock. Additionally, injury due to cold shock may
occur if dilutions are placed in the refrigerator when subsequent laboratory tests cannot
be immediately completed (Van Schothorst and Maggie Duke 1984).

Sensitivity to cold varies broadly. No single property explains this sensitivity, but
population density, growth temperature, cooling rate and the temperature range over
which cooling occurs all impact the survival of cold-shocked cells (Postgate and Hunter
1963, Mackey 1984). The effect of food components on the degree of injury and survival
of cold-stressed microorganisms has not been studied in great detail. Preliminary work
has shown that water or diluents are more stressful environments than broth or food
(Mackey 1984). The time to decrease salmonellae levels 90% in water at 0 to 5°C was
between 2 and 16 d, while in vacuum-packaged beef in the same temperature range, only
a 50% reduction was observed after 28 d (Mackey 1984, Mitchell and Starzyk 1975,
Kennedy et al. 1980). However, chilling in oyster homogenate at 4°C for 24 h decreased
populations of Vibrio vulnificus nearly 7 logs (lethal cold stress), while cold storage in a

salt-based culture medium had no effect (sublethal cold stress) (Oliver 1981).

27

Although it is not known to be an effective method for destroying bacteria,
freezing can be an injurious treatment. Injury results fiom continued exposure to
concentrated solutes and physical damage caused by ice crystal formation. Many
constituents of food and culture media are protective against freeze damage. These
cryoprotectants include glycerol, sodium glutamate, certain sugars, peptides and proteins
(Mackey 1984). The presence of such protectants may be why Semanchek and Golden
(1998) found fieezing to have a minimal impact on the development of injury in E. coli
0157:H7.

Osmotic stress seems to be associated with both freezing and fi'eeze—drying.
Freeze-dried microorganisms are subject to freezing, drying, storage and rehydration
stresses (Mackey 1984). Rapid rehydration of foods and freeze-dried cultures can cause
osmotic injury (Van Schothorst and Maggie Duke 1984). Osmotic stress can occur when
shifts in external osmolarity cause water to flow either into or out of the bacterial cell. In
extreme conditions, this type of stress can cause physical damage to the cell. Less severe
osmotic changes affect water availability and can be caused by sodium or other salt
challenges (Bremer and Kramer 2000). Water availability is intimately linked to water
activity. In food processing, microorganisms respond to both a “storage” aw and a
“treatment” aw (Quesnel 1984).

In contrast to other stresses, heat Shock is perhaps the most studied and best
understood. Heat shock occurs when organisms are shifted for short periods of time fi'om
lower to higher temperatures within or above their normal growth range (Bunning et al.
1990, Mackey and Derrick 1986, Pagan et al. 1997, Farber and Brown 1990). The shifted

temperatures are not yet lethal, particularly given that the bacterial population is

28

heterogeneous with respect to growth phase and heat sensitivity. Even temperatures and
treatments meant to be lethal will not completely destroy all organisms, as a potentially
significant number of cells actually survive a given heat process and are merely injured
(Murano and Pierson 1993).

Both pre-processing and processing environments present conditions that mimic
or cause heat shock. Guidelines for the use of acid washes to decontaminate carcasses
include a spray temperature of 20°C to 60°C (Gerris 2000). Heat shock was reportedly
induced in Escherichia coli 0157:H7 at 42°C (Murano and Pierson 1993),
Campylobacterjejuni at 46°C (Palumbo 1984), S. Typhimurium at 48°C (Mackey and
Derrick 1986, Bunning et al. 1990), and Listeria monocytogenes in a fermented beef/pork
sausage homogenate at 48°C and 52°C (Farber and Brown 1990). Hot acid sprays may
elevate the superficial temperature of the carcass, altering the growth and resistance
profile of indigenous flora (Castillo et al. 2001).

Thermal processing can cause injury to microorgansirns. Food products requiring
long heating lag pluses, including products pasteurized at low temperatures for long
times and egg products, are subject to heat Shock conditions. (Mackey and Derrick
1987b, Murano and Pierson 1993, Bunning et al. 1990). The behavior of microorganisms
in slowly heated foods will presumably mimic the behavior of microorganisms subject to
isothermal heat shock (Mackey and Derrick 1987a). Thus, meat products and bulk
products that are slow cooked to a final internal temperature and sous vide foods (Pagan
et al. 1997) are also subject to heat shock conditions (Mackey and Derrick 1987a, Farber
and Brown 1990, Quintavalla and Campanini 1991). Microorganisms present in meat

products left on warming trays before receiving a final re-heating could also experience a

29

heat shock (Farber and Brown 1990).
2.3.3. Cellular Modification and Damage

Environmental stresses and stresses caused by processing are thought to cause
loss of some cell structure functions and/or protein denaturation (Lou and Yousef 1996).
Table 2.7 outlines the types of cellular damage caused by various treatments. Dead,
injured and otherwise normal cells likely differ in the degree of injury to structural and
fimctional components (Ray 1984).

Exposure to low pH environments may remove or sequester cations from key sites
in a microorganism (Gould 1984). Damage to ribonucleic acid (RNA) is linked to low
pH, and could be due to removal of magnesium (Mg2+) (Przybylski and Witter 1979,
Hurst 1984). Mg2+ is necessary for ribosomal integrity with acetic acid reportedly
interfering with the ribosome and/or protein synthesis (Hurst 1984, Mossel and van
Netten 1984). Shifts in pH can disrupt the proton motive force (Russell 1984, Rowbury
2003) and also alter the protein profile of the outer membrane in Gram-negative cells, the
cytoplasmic membrane in Gram-positive cells, or the protein coat of spores (Leyer and
Johnson 1993, Gould 1984). Low cytoplasmic pH can also damage deoxyribonucleic
acid (DNA) (Yousef and Courtney 2003, Rowbury 2003).

Starvation conditions can lead to an increase or decrease in exopolysaccharide
levels (Dickson and Frank 1993), as well as entry into stationary phase, a mode
characterized by drastic changes in the cell envelope, membrane composition, and DNA
structure (Hengge-Aronis 1993, Rowbury 2003). Cells under starvation experience
physiological changes that include decreased cell size and membrane fluidity and

increased protein turnover (Lou and Yousef 1996). Microorganisms in low nutrient

30

Table 2.7. Types of cell damage resulting fi'om different sublethal treatments (adapted
from Mossel and van Netten 1984, Ray 1986, Wuytack et al. 2003, Rowbury 2003).

 

 

 

damage to:
cell wall or membrane proteins RNA DNA
cell wall (leakage) (ribosomes) or
synthesis or RNA DNA
synthesis synthesis
observed in
cells treated
by:
freezing x x x
drying x x x x
fi'eeze-drying x x x x
heating x x x x
gamma-irradiation x x x ‘7 x
change in
oxygen potential x
osmotic shock x x
starvation ‘7 x
high hydrostatic
pressure x x
pulsed white light x x x
salt shock x

31

environments likely integrate cell density and starvation stress signals to allow for cell
surface modifications and utilization of alternate energy sources (Lazazzera 2000,
Postgate and Hunter 1963). Alterations in cellular morphology and modifications in cell
surface components enhance adherence and may contribute to biofilm formation.

Damage and modification of the cell membrane are associated with almost all
forms of physical stress (Mackey 2000). The membrane is a common site of injury by
chilling, fi'eezing and heating (Boziaris and Adams 2001). Damage to the outer
membrane of Gram-negative cells includes release of lipopolysaccharides (LPS), lipids,
phospholipids, divalent cations necessary for LPS stability and periplasmic enzymes, all
of which disrupt the permeability barrier (Boziaris and Adams 2001, Ray 2001, Hurst
1977). Permeability control is also lost at low temperatures because of decreased
membrane fluidity (Graumann and Marahiel 1996, Mackey 1984). Changes in
permeability have led to the suggestion that certain treatments (heating, freezing, drying,
radiation) create small pores in the outer membrane (Mackey 2000). Repair of sublethal
injury involves repair of these pores, but conditions for repair of the outer membrane
does not necessarily facilitate repair of the cytoplasmic membrane (Hurst 1977).

While Gram-positive organisms do not have an outer membrane, they do have a
surface protein layer outside the cell wall. Both the surface protein layer and cell wall are
potential sites of injury. Freeze injury to Lactobacillus bulgaricus has been shown to
damage the protein layer (Mackey 2000, Wright and Klaenhammer 1981). Mg2+ and D-
alanine were lost from cell wall teichoic acid polymers of Staphylococcus aureus heated
in phosphate buffer (Hurst and Hughes 1981).

VanBogelen and Neidhardt (1990) theorize that ribosomes may function as

32

sensors of conditions or temperatures that elicit a heat or cold shock. Ribosome and
ribosomal RNA (rRNA) degradation are often observed as heat-induced lesions
(Genthner and Martin 1990, Tolker-Nielsen and Molin 1996). Due to destruction of 16S
rRNA, the 308 ribosomal subunit is damaged or destroyed with thermal stress in S.
aureus, S. Typhimurium B and E. coli (Hurst 1977). Destruction of the 308 subunit is
prohibited when Mg2+ (as MgClz) is added to the heating menstruum. Loss of ribosomes
also occurs with starvation, particularly when conditions lead to intracellular Mg2+
depletion (Mackey 2000). As Mg2+ is necessary for ribosomal integrity and inhibition of
ribonuclease, it appears ribosomal damage is largely a consequence of Mg2+ loss (Hurst
1977).

While DNA damage (Figure 2.1) has been observed in heated, freeze-thawed,
dried, and acid-treated cells, it may be an indirect result of these treatments (Mackey
2000, Russell 1984). For instance, strand breakage during or after heat treatment is likely
due to the stimulation of endonuclease activity (Russell 1984, Hurst 1977). Death in
cold-shocked E. coli is connected to the loss of Mg”, which is required by DNA ligase
for DNA synthesis and repair (Sato and Takahashi 1970).

Certain treatments, particularly heating, can denature proteins, leading to the
inactivation of enzymes and disruption of the active transport of cations, sugars and
amino acids (Hurst 1977). Dehydrogenases are especially heat sensitive and glycosylases
and endonucleases, enzymes involved in DNA repair, are heat-labile at mild temperatures
of 40°C-50°C (Pellon and Sinskey 1984). Sublethally heat-injured S. aureus have
reduced catabolic capabilities, and glucose-transport is disrupted in heat damaged S.

Typhimurium (Busta 1976). Heating has been shown to inactivate catalase and

33

 

o Single-strand breaks D Freezing (and thawing)
Ionizing radiation

 

o Double-strand breaks A} Freezing (and thawing)

0 Strand breakage during

 

 

or after treatment 5
(as a result of heat Moist heat.
stimulated endonuclease
activity)
0 Depurination ’ Dry heat

Figure 2.1. Specific types of damage to bacterial DNA (Russell 1984).

34

superoxide dismutase in several microorganisms, including L. monocytogenes and S.
aureus (Andrews and Martin 1979, Dallrnier and Martin 1988). The loss of this activity
may contribute to the oxygen sensitivity of injured microorganisms.

2.3.4 Oxygen Sensitivity of Injured Cells

Oxygen species lethal to injured microorganisms may be rapidly formed in both
selective and non-selective culture media. Reactive oxygen species (ROS) include the
superoxide radical (02"), hydrogen peroxide (H202), the hydroxyl radical (OH), singlet
oxygen, hypochlorous acid, lipid peroxides, protein hydroperoxides and ozone (Valentine
et al. 1995, Simpson et al. 1992). Autooxidation of reducing sugars in the presence of
phosphate during autoclaving has been implicated in the generation of ROS in culture
media (Stephens et al. 2000, Baylis et al. 20003). Media containing manganese and/or
citrate can also autooxidize to form peroxides. If peroxides accumulate in media in the
presence of superoxide radicals, hydroxyl radicals will form. Hydroxyl radicals can also
form when H202 reacts with Fe2+ (McDonald et al. 1983, Sub and Knabe12000). ROS
can also be formed endogenously as a consequence of the normal metabolic reduction of
molecular oxygen to water. Reactive intermediates include 02”, H202 and 'OH (Stephens
et al. 2000).

Oxidative stress caused by elevated levels of O2", H202 or “OH can lead to
damage of cellular components (Storz and Zheng 2000). Although the superoxide radical
is believed to be only moderately reactive, it is still capable of exerting detrimental
effects, including oxidation of membrane phospholipids, proteins and DNA, inhibition of
catalase and peroxidases, and reduction of metal ions (Stephens et aL 2000, Schellhom

and Hassan 1988, Takeda and Avila 1986). Metal ions in a reduced state are unstable and

35

have the potential to reduce H202, producing the extremely toxic hydroxyl radical. H202
is more reactive than the superoxide radical, and H202 produced in the media can readily
diffuse across the cell membrane and disrupt membrane transport systems, as well as
enzyme and nucleic acid synthesis. Of the reduced forms of oxygen, the hydroxyl radical
is the most reactive, being able to immediately oxidize cellular components, including
proteins, nucleic acids and lipids (Stephens et al. 2000).

Two antioxidant enzymes, superoxide dismutase (SOD) and catalase, work
together to protect bacteria from the harmful effects of oxygen radicals produced during
the course of normal respiration. SOD converts superoxides to 02 or H202 and catalase
converts H202 to H20 and 02 (Schellhorn and Hassan 1988, Takeda and Avila 1986).

As early as 1925, it was suggested that survival of heated microorganisms is
affected by the presence or absence of air, and that this factor should be included in
studies of thermal death time (Jackson and Woodbine 1963, Oerskov 1925). D623oc-
values were six- fold higher when L. monocytogenes was enumerated with strictly
anaerobic techniques than with aerobic recovery (Knabel et al. 1990). Equivalent heat
treatments at 55°C for 100 min, 59°C for 5 min and 61°C for l min effected a 6-log
reduction in E. coli 0157:H7 (Bromberg et al. 1998). Only half of this 6-log reduction
was due to thermal inactivation (as determined by anaerobic enumeration), with the
remainder due to the inability of sublethally injured cells to grow in the presence of
oxygen. Gért and lmlay (1998) suggest E. coli produce only enough SOD to protect
intracellular components from endogenously generated superoxide. Hence, even
unstressed microorganisms face the challenge of neutralizing ROS produced in culture

media. The challenge is even greater for stressed cells, which typically exhibit

36

heightened sensitivity to peroxides and superoxide radicals due to diminshed SOD and
catalase activity and the inability of injured organisms to synthesize new enzymes (Suh
and Knabe12000, Patel et al. 1995, McDonald et al.1983, Czechowicz et al. 1996).

Research on the stability of catalase and SOD in a wide variety of food-related
microorganisms to different environmental stresses is limited. Heating S. aureus in
phosphate buffer at 52°C for 20 min reduced catalase activity by nearly 40%, with
catalase loss increasing to about 90% during the first 2-4 h of repair in Trypticase soy
broth (T SB) (Andrews and Martin 1979). Amin and Olson (1968) saw a 10- to 20-fold
reduction in staphylococcal catalase activity at 54.4°C compared to 378°C. Dallmier and
Martin (1988) and Zemser and Martin (1998) reported a sharp decrease in catalase
activity when L. monocytogenes was heated between 55°C and 60°C. Shifting the
temperature of E. coli from 30°C to 45°C inhibited catalase (Smirnova et al. 2001a), and
Mackey and Seymour (1987) attributed the increased peroxide sensitivity of heat-injured
E. coli K12 to at least partial inactivation of inducible catalase/peroxidase. Murano and
Pierson (1993) found that heat shock at 42°C lowered the activity of catalase and SOD in
E. coli 0157:H7, and that even without the heat shock, heating at 55°C could eliminate
the activity of both enzymes. While Vasconcelos and Deneer (1994) saw no change in
SOD levels in L. monocytogenes subjected to a heat shock at 42°C, Dallmier and Martin
(1988) demonstrated a decrease in SOD activity after heating at 45°C.

Smirnova et al. (2001b) showed that the responses of E. coli to cold shock (from
37°C to 20°C) and long-term cultivation at 20°C were similar to the typical oxidative
stress response, with an enhanced expression of SOD and catalase. Stead and Park

(2000) identified superoxide radicals, and H202 in the event that these radicals are not

37

detoxified, as the toxic species produced during freeze-thaw treatment of Campylobacter
coli. By showing that SOD-deficient mutants are sensitive to freeze-thaw, their work
demonstrated that SOD is important in the resistance of C. coli to this stress. Catalase-
deficient mutants also had an increased sensitivity to fi'eezing and thawing. However, the
increased sensitivity to H202 of C. jejuni injured by freezing and mild heating observed
by Humphrey (1988) was not associated with changes in total catalase activity. However,
these cells required the addition of catalase for growth on nutrient agar.

Efforts have been made to increase the recovery of injured microorganisms by
eliminating oxidative stress by incorporating reducing agents or other agents that act to
scavenge 02 or degrade H202 in recovery media (Suh and Knabe12000, Patel et al. 1995,
McDonald et al. 1983). Overall effectiveness of the various supplements is dependent on
the organism and variations in the type, degree and site of injury.

When sodium pyruvate or a—ketoglutaric acid, both nonenzymatic H202-
degrading compounds, were added to Luria-Bertani agar, recovery of starvation
(sterilized distilled water) and low-temperature (4°C) stressed E. coli 0157:H-strain
E32511/HSC increased from less than 0.1 CFU/mL to 104-105 CFU/mL within 48 h
(Mizunoe et al. 1999). Calabrese and Bissonnette (1990) observed a significant increase
in the recovery of chlorine-stressed coliforms when standard recovery media was
supplemented with catalase, pyruvate or a combination of the two. Rayman et al. (1978)
found that the addition of pyruvate or catalase to Trypticase soy agar (TSA) stimulated
the growth of heat-injured S. Seftenberg but not of Streptococcus faecium, an organism
not possessing catalase. TSA supplemented with 0.1% sodium pyruvate was shown to

recover 10 times as many heat-damaged E. coli 0157:H7 as unsupplemented TSA

38

(Czechowicz et al. 1996). Pyruvate has also been shown to enhance the recovery of heat-
stressed L. monocytogenes (Busch and Donnelly 1992, Patel et al. 1995), Shigellaflexneri
(Smith and Dell 1990), S. aureus (Hurst et al. 1976, Martin et al. 1976), and heat— and
fi'eeze-injured E. coli (McDonald et al. 1983).

The addition of catalase to media has also been shown to increase the recovery of
injured cells, up to “00-fold in the case of heated and freeze-dried S. aureus, as well as
S. aureus subjected to low aw (Flowers et al. 1977). Martin et al. (1976) reported that the
addition of catalase to TSA containing 7.0% NaCl increased the enumeration of heat-
injured S. aureus 104-fold. Although Stephens et al. (2000) determined catalase did not
have a significant effect on the recovery of heat-injured S. Typhimurium, Rayman et al.
(1978) reported a lSl-fold increase in heat-injured S. Seftenberg counts when catalase
was added to TSA. Catalase has also been shown to increase the recovery of acid-injured
E. coli (Martin et al. 1976) and heat-injured L. monocytogenes (Patel et al. 1995) and E.
coli (Mackey and Seymour 1987).

The metabolic reduction of 02 to H20 is achieved by the action of electron
transport enzymes located in the cytoplasmic membrane. Oxygen-reducing membrane
fractions have been derived fi'om several bacterial species (Adler 1990). 0xyrase®, the
partially purified, oxygen-reducing membrane fragment of E. coli, can remove 02 fi'om
culture media and destroy oxygen radicals. The latter is probably due to the presence of
catalase. Oxyrase® has been shown to enhance recovery of L. monocytogenes following
heat injury (Yu and F ung 1991, Sub and Knabe12000). Additionally, Patel and Beuchat
(1995) and Patel et al. (1995) showed that Oxyrase® increased recovery L.

monocytogenes fiom heat-pasteurized whole milk and filtrates of homogenized beef, but

39

not fi'om skim milk or cabbage filtrates. Media containing Oxyrase® recovered up to 1.5-
log CFU/mL more heat-stressed S. Typhimurium and up to 2.0-log CFU/mL more
acid/salt-stressed S. Typhimurium DT104 than media without Oxyrase® (Stephens et al.
2000). Oxyrase® also can reportedly enhance recovery of radiation-damaged E. coli
(Adler et al. 1981), as well as heat-injured E. coli 0157:H7 and Yersinia enterocolitica
(Thippareddi et al. 1995).

2.3.5 Repair and Stress Responses

Several key factors define the concept of injured cells. The injured population
may contain bacterial cells that are slightly to severely injured, with viable nonculturable
and dead cells also likely to be present. If appropriate intrinsic (nutrients, aw) and
extrinsic (temperature, relative humidity) parameters exist during the farm-to-fork
continuum, most injured cells will repair and regain the characteristics of normal cells
(Ray 1986, Busta 1994), as well as the additional characteristic of increased stress
tolerance associated with stress adaptation. The repair process is reflected by delayed
germination of spores, a prolonged lag phase for vegetative organisms and the inability to
multiply until repair has taken place (Busta 1994). Following repair, resistance to
selective agents and the ability to proliferate in their presence is regained (Brashears et al.
2001).

Repair times can be estimated from the length of the extended lag phase or by the
differential plating method. Using non-selective media, the estimated lag phase is the
time to repair injury and begin multiplication. The differential plating method uses a
selective medium containing a substance inhibitory to injured microorganisms (i.e., due

to membrane damage) and a non-selective medium with no obvious inhibitory effects on

40

unstressed cells. The non-selective (reference) medium enumerates the entire population
and the selective medium only the healthy fiaction. With differential plating, repair time
is the time necessary to regain resistance to the selective medium. The time needed for
repair depends on the bacterial species and strain, the type and severity of injury and the
inhibitory nature of the selective media (Mackey 2000).

Measurements with differential plating suggest that, under optimal conditions,
injury caused by treatment with acid, chilling, freezing, freeze-drying, gamma-radiation
or mild heating can be repaired within 4-5 h, with severe heat injury sometimes requiring
much longer repair times (Mackey 2000). Repair of C. jejuni subject to heat shock in
potassium phosphate buffer at 46°C for 45 min was complete within 4 h of incubation at
37°C. No repair occurred at 5°C (Palumbo 1984). Damage to E. coli caused by sublethal
acidfication (sodium acetate buffer, pH 4.2, 60 min) was completely repaired after 1-2 h
at 32°C in TSB, with 95% repair reported after 120 min in potassium phosphate buffer
(pH 8.0) (Przybylski and Witter 1979). Freeze- and alkali-injured E. coli cells can also
recover when incubated in phosphate buffer (Musarrat and Ahmad 1988).

Repair times determined by estimating lag time are not always in agreement with
those determined by differential plating. Mackey and Derrick (1982) determined that
gamma-irradiated and dried S. Typhimurium sometimes recovered full resistance to the
selective medium before the lag phase was complete. In some instances, the time
individual cells of heat-injured S. Typhimurium required to regain tolerance to salt (up to
14 h) was longer than the lag time (9 h). Mathew and Ryser (2002) reported growth of
heat-injured L. monocytogenes within the first two hours of incubation, although percent

injury was not fully repaired when assessed by the differential plating method. Meyer

41

and Donnelly (1992) also reported growth of heat-injured L. monocytogenes before
resistance to the selective medium was fully regained.

Lag and repair times for injured bacteria reported in the literature must be viewed
with caution. Cell—to-cell variability (including growth phase and cell-cycle stage) during
stress exposure typically leads to variations in injury within a population (Stephens et al.
1997), and repair time as determined by differential plating or estimation of lag phase can
be biased due to this heterogeneity. In general, less severely injured cells repair quicker
and have a shorter lag period than more severely injured cells. When the cells with the
shorter lag period begin multiplying, they can mask the presence, on non-selective and
selective media, of cells requiring longer repair times. Consequently, a single or average
repair time does not reflect the distribution within a population and may well be an
underestimation (Mackey and Derrick 1982, Mackey 2000). This may help explain the
growth observed by Mackey and Derrick (1982), Meyer and Donnelly (1992) and
Mathew and Ryser (2002) before all cells regained resistance to selective agents.

Stephens et al. (1997) examined the distribution of lag times of heat-injured S.
Typhimurium by serially diluting the culture, inoculating the dilutions across many
microtitre plates, monitoring growth by turbidity and calculating lag times fi'om a model
that extrapolated the growth curve back to initial inoculum levels. While the lag times
for individual healthy cells were narrowly distributed, the lag times for injured cells were
widely distributed, from <12 h to >20 h, with some >30 I). These results highlight the
heterogeneous nature of bacterial populations and have apparent implications for
traditional and rapid methodology involving pre-enrichment and repair steps, particularly

when the overgrth of competing bacteria is considered.

42

Repair requires that specific biochemical events occur, the nature of which differ
with the type of stress. Metabolic processes that occur during repair can include the
synthesis of adenosine triphosphate (ATP), DNA, RNA, proteins and the reorganization
of existing macromolecules, including LPS in Gram-negative organisms and teichoic acid
in Gram-positive organisms (Ray 2001, Hurst 1984, Ray 1986). Repair of the cell
membrane through lipid synthesis must occur relatively early so that cells can fiilly repair
other stress-induced lesions (Beuchat 1978).

Normal cellular functions may be re-established by the transient synthesis of
general or specific stress proteins (Lou and Yousef 1996). Since the confirmation of heat
shock proteins in Drosophila sp. in 1974, virtually every organism subsequently studied
has been shown to respond to a moderate temperature shock with the increased
production of specific proteins (Mackey and Derrick 1986, Bunning et al. 1990, Farber
and Brown 1990, Watson 1990, Snyder and Champness 1997). Bacterial responses to
stress can be general or specific. In some instances, a component of a given stress
response may be part of both general and specific response pathways (IFT 2002).

Characteristics of stress proteins include increased production under conditions
that repress the synthesis of most other cellular proteins (Lindquist 1992), a functional
role in adaption of bacteria to growth- and survival-limiting conditions (Vlilker et al.
1992), and the overall mediation of recovery of stress-induced damage (Bunning et al.
1990). Stress proteins in prokaryotic and eukaryotic organisms are among the most
evolutionarily conserved proteins (Watson 1990, Snyder and Champness 1997), and the
occurrence of the heat shock response in prokaryotes, eukaryotes and archaebacteria

indicate it must have developed in the earliest common form of cellular life on Earth,

43

possibly 2 to 2.5 billion years ago (Neidhardt and VanBogelen 1987).

The general stress response in most Gram-negative bacteria, including the enteric
bacteria E. coli, Shigella flexneri and S. Typhimurium, is regulated by RpoS, the
alternative sigma-subunit of RNA polymerase (68) (Abee and Wouters 1999). In E. coli,
RpoS controls the expression of more than 50 genes involved in the general stress
response (Loewen et al. 1998). RpoS can be induced by a number of stresses, including
nutrient starvation, osmotic shock, high and low temperatures, pH stress and oxidative
stress (Abee and Wouters 1999, Hengge-Aronis 2000, Rowbury 2003). Dodd and
Aldsworth (2002) demonstrated induction of RpoS in Salmonella with changes in aw
produced using different humectants and food preservatives. Bacteria defective in the
gene for RpoS experience greater sensitivity to food processing conditions, such as heat
shock, starvation, acid and ethanol (Abee and Wouters 1999, Rees et al. 1995).

In L. monocytogenes, S. aureus, Bacillus subtilis and other Gram-positive
bacteria, the alternative sigma-subunit O'B regulates the general stress response. In B.
subtilis, 0'8 controls the expression of over 40 genes involved in the general stress
response (Abee and Wouters 1999). Like RpoS, 63 can be induced by a number of
environmental stresses, including ethanol, nutrient starvation, oxygen limitation, low and
high temperatures, high salt concentrations and hydrogen peroxide (Abee and Wouters
1999, Volker et al. 1992). Disruption of 0° in B. subtilis can increase sensitivity to
oxidative stress, and disruption in L. monocytogenes reportedly decreases resistance to
acid and osmotic stress (Abee and Wouters 1999).

Several specific stress responses have been identified, and like the general stress

response, these are adaptive responses that allow bacteria to survive and, in some cases,

44

multiply under stressful conditions (Abee and Wouters 1999, Ray 2001). The heat shock
response and associated heat shock proteins (HSPS) are the most studied. Several HSPS
act as molecular chaperones or chaperonins. These chaperones help cells survive by re-
folding or targeting for destruction proteins that are improperly assembled or denatured
by the abrupt rise in temperature (Snyder and Champness 1997, Jones et al. 1996,
Graurnann et al. 1996, Lindquist 1992). Some HSPS possess proteases or peptidases
activity for degrading irreparably denatured proteins (Snyder and Champness 1997, Jones
et al. 1996, Lindquist 1992, Yousef and Courtney 2003). HSPS may also play a role in
DNA repair and replication, cell division and modification of cellular morphology (E.
coli exhibits a transient tendency to elongate), and the accumulation of osmolytes, which
may maintain or enhance protein stability (Abee and Wouters 1999, Tsuchido et al. 1986,
Neidhardt and VanBogelen 1987).

One theory of the heat shock response is that it is responsible for the synthesis of
proteins that replace thermolabile components of macromolecule-synthesizing systems,
such as ribosomes. Presence of an inducible rather than constitutive heat shock system in
thermophilic bacteria suggests involvement as a mediator of temperature shifts rather
than high temperatures (Neidhardt and VanBogelen 1987). HSPS have been identified
following several other stresses, including starvation and anaerobiosis, as well as
exposure to ethanol, other organic solvents, oxidative agents and high salt concentration
(Jenkins et al. 1988, Watson 1990, Lindquist 1992, Rowbury 2003, Girgis et al. 2003).
Accumulation of damaged or denatured proteins is thought to trigger the synthesis of
HSPS following environmental stress (Watson 1990).

Many bacteria respond to abrupt decreases in temperature by transiently

45

synthesizing a number of protective proteins (Graumann and Marahiel 1996). Cold
shock proteins (CSPs) have not been studied in prokaryotes and eukaryotes to the same
degree as HSPS, but they have been demonstrated in Bacillus cereus (Mayr et al. 1996),
B. subtilis (Lottering and Streips 1995, Graumann et al. 1996), E. coli (Jones et al. 1987,
Goldstein et al. 1990), Lactococcus lactis (Wouters et al. 1998), L. monocytogenes
(Bayles et al. 1996), V. vulnificus (McGovern and Oliver 1995), S. Enteritidis (Jeffreys et
al. 1998) and S. Typhimurium (Craig et al. 1998). These CSPs are thought to play a
critical role in various cellular and physiological functions, including DNA
recombination, transcription, translation, messenger RNA and protein folding, sugar
uptake, chemotaxis and general metabolism efficiency (Jones et al. 1996, Graumann and
Marahiel 1996, Wouters et al. 2000). The maintenance of membrane fluidity at low
temperatures is necessary for cold adaptation and involves at least one constitutively
expressed enzyme that is active only at low temperatures (Abee and Wouters 1999).

S. Typhimurium can withstand potentially lethal acid shock conditions (pH<4) if
preceded by adaptation to milder conditions (Bearson et al. 1996, Foster 1995). This acid
adaptation, stemming from a gradual decrease in pH to non-lethal levels (pH>4), induces
the acid tolerance response (ATR) (Leyer and Johnson 1993, Abee and Wouters 1999,
Bang et al. 2002). Acid responses, also described for L. monocytogenes, E. coli and S.
flexneri (Abee and Wouters 1999), can be induced by organic acids commonly used in
the food industry (O’Driscoll et al. 1996). The more effective acid resistance systems for
E. coli and S. flexneri allow survival at pH 2, while the ATR of S. Typhimurium is a pH 3
tolerance system (Audia et al. 2001, Abee and Wouters 1999, Foster and Moreno 1999).

During acid adaptation and shock, S. Typhimurium induces the expression of over 50

46

acid shock proteins (ASPS) in the exponential phase and 15 ASPs in the stationary phase
(Foster 1991, Foster and Moreno 1999, Foster 1995). These proteins are under the
control of multiple, overlapping regulatory systems and protect cells against acid and
perhaps other environmental stresses (Audia et al. 2001).

The oxidative response typically involves proteins for preventing (e. g., catalase
and superoxide dismutase) and repairing (e. g., exonucleases and glycosylases) oxidative
damage (Farr and Kogoma 1991). Starvation proteins also have been demonstrated,
independent of the nutrient for which a cell is starved. Some of these proteins are unique
to starvation while others are common to other stresses (Hengge-Aronis 1993, Matin
1991, Jenkins et a]. 1988, Vtilker et al. 1992). Starvation and stationary phase proteins
are likely responsible for maintaining long-term survival and perhaps enhancing general
cellular resistance by stabilizing the ribosome against degradation (Tolker-Nielsen and
Molin1996), morphing cells into a more spherical shape (Givskov et al. 1994) and
improving metabolic potential through the utilization of alternative growth substrates
(Matin 1991).

Their overlapping nature and the wide variety of stresses capable of inducing their
expression complicates the classification of stress proteins as general or specific (Watson
1990). The universal induction of many of the same stress proteins after exposure to
different stresses has been reported in B. subtilis, E. coli, E. faecalis and L. lactis (Girgis
et al. 2003). Jenkins et al. (1988) studied starvation followed by heat and hydrogen
peroxide stress in E. coli. While each stress produced its own distinct protein pattern,
eleven heat shock and six oxidative stress proteins were common to starvation proteins,

with three proteins common to all three stresses. Vc'ilker et al. (1992) determined that the

47

proteins induced by B. subtilis following heat stress were both general stress proteins and
heat shock specific proteins, indicating that the heat shock and general stress responses
are related. Some genes associated with the general stress response have clear functions
for managing specific stresses, such as oxidative and osmotic stress (Yousef and
Courntey 2003) while others play a role in general protection under multiple stress
conditions (IFT 2002).
2.3.6 Cross-Protection

Sykes (1963) attributed survival in adverse environments to sublethal treatments
insufficient to cause death and/or to protection from the unfavorable conditions. It was
inferred from the latter that bacterial cells could adapt or acquire resistance to different
conditions by modifying metabolic activities, adjusting nutrient utilization or making use
of enzymes previously present in a recessive role. Although the exact mode of action is
not fiilly understood, a role in stress protection has been proposed for stress proteins,
especially HSPs and starvation proteins (Pagan et al. 1997, Hengge-Aronis 1993, Abee
and Wouters 1999, Hecker at al. 1996). The long-term survival of E. coli is dependent on
the synthesis of starvation proteins (Matin 1991), and Givskov et al. (1994) concluded
that protein synthesis induced by starvation was necessary for Pseudomonas putida to
develop a general stress-resistant state. Jenkins et al. (1988) correlated the induction of
starvation proteins with protection of E. coli against heat and H202.

That the induction of certain stress proteins is coincident with the development of
acquired resistance to heat (acquired thermotolerance) and other stresses (cross-
protection) does not establish a direct cause-and-effect relationship (Jenkins et al. 1988,

Juneja and Novak 2003). However, the heat shock and general stress responses may

48

induce changes tint protect cells from the effects of heat and other stress conditions (IFT
2002). The phenomenon that one stress offers protection when subsequently stressed at
higher levels (especially for heat) or by other types of stresses is widely documented.
Lou and Yousef (1996, 1997) describe this as “stress-hardening”, suggesting “if the first
encountered stress is not lethal to a microorganism, the adaptation of the microorganism
to this sublethal stress may ‘harden’ the microorganism and allow its survival at
subsequent stresses.”

Watson (1990) believes that acquired thermotolerance is the most characteristic
physiological response microorganisms have to mild temperature shock. Novak et al.
(2001) reported that vegetative cells of Clostridium perjringens incubated at 46°C for 60
min had significantly higher Dam-values than those incubated at 28°C. Exponential
phase cells of Vibrio parahaemolyticus heat shocked at 42°C for 30 min had a Dunc-
value of 3.33 min and were significantly more heat resistant than unshocked cells having
a Dunc-value of just 2.03 min (Wong et al. 2002). Heat shocking E. coli 0157:H7 at
42°C for 5 min before thermal inactivation at 55°C also increased the D-value >2-fold
(Murano and Pierson 1992, 1993). Juneja et al. (1998) sublethally heated samples of beef
gravy and ground beef previously inoculated with a four-strain cocktail of E. coli
0157:H7 at 46°C for 15 to 30 min. The heat-shocked cells were more therrnotolerant
than unshocked cells, with a 1.56- and 1.50-fold increase in the time to achieve a 10,000-
fold reduction at 60°C in beef gravy and ground beef, respectively. When treated at
62°C, cells of L. monocytogenes heat-shocked in Trypticase soy broth containing 0.6%
yeast extract (TSB-YE) at 45°C for 180 min were six-fold more heat resistant than

unshocked cells (Pagan et al. 1999), and a heat shock in TSB-YE at 48°C for 10 min

49

increased the DssoC—value of L. monocytogenes more than two-fold (Linton et al. 1992).
Mackey and Derrick (1987b) determined that S. Thompson inoculated in TSB, liquid
whole egg, and 10% or 40% reconstituted dried milk was more thermal tolerant at 54°C
or 60°C if held at 48°C for 30 min prior to treatment.

The response of microorganisms in foods heated slowly is thought to be the same
as the response to an instantaneous heat shock, namely acquired thermotolerance. There
is evidence to indicate that heating rate should be included when determining inactivation
kinetics. Quintavalla and Campanini (1991) found that L. monocytogenes was more heat
resistant at 60°C, 63°C and 66°C when heated slowly (0.5°C/min) in a pork emulsion
than if heated virtually instantaneously to these temperatures. Mackey and Derrick
(1987a) examined the heat resistance of S. Typhimurium as affected by rising
temperatures and concluded that slowly increasing the temperature (2°C/min, 1°C/min
and 0.6°C/min) from 20°C up to 55°C, versus instantaneous heating at 55°C, increased
thermal resistance. The resistance of cells increased with decreasing heating rates, with
the most resistance afforded by a heat shocking at 48°C for 30 min before heating. The
observation of increasing thermal resistance with decreasing heating rates may be a
general phenomenon, as it was also reported by Stephens et al. (1994) for L.
monocytogenes.

Thermotolerance can also be induced by stresses other than heat. Table 2.8
summarizes some of the available data for different microorganisms. In addition, many
sublethal treatments can cross-protect against stresses other than heat. Jenkins et al.
(1988) reported that starvation or adaptive treatments with heat, H202 or ethanol protects

E. coli against further oxidative (H202) challenges. According to look et al. (2001),

50

 

as a. he 0.2 a 8.2 m5
Son “Eczema 98 «5888. 06mm .8 Doom €05 Sufi—moi 88an E sausages Sammgnueaefi 4

00R
82 88:2 e5 23 ea 8 s as he 0.8 a m 5285 a :3 me ceases? 29a

065m 8 a 3 new 882w $86

82 as: e5 eeaazaeafi 00?. £5 8288 Ease a a Esteem eaaeaafl a
can a e 4 5
SE a. a Seas can sees names a: a sewage 88:6
0.2 a e M:
58 ease Ba 8583 as on he ooem he map a 8.33 me 8:938 23,
0.2 a e =8 .m
sea agate e5 8:284 as on he poem we as Ge me Caz sad 5 eeeatsm Uanaeseeaeez
0.2 a e M:
88 ease eea 8583 as on he ooem ae mme a at? me Susana. 22
us: a e
88 ease as 8353 as 8 ae poem 5. he Ge me Q2 :23 5 Susan
0.2

88 E as seem as 8 e a: he ooem a e em he as; Bane a cashew 2:20 :8 .m

 

3:93qu omen-.30 :3: Eon—28:. 393m EfluawaeEom—z

 

.mEmewSSoE no.“ 85838.53“ 8:25 3 9592 3:08:35. .m.m 03:.

51

 

0353 no oafimmona dowobm: €3.80

 

$2 a 5 >396 003 .8 853.3 a moon a Suntan seam m
n _ not“ 0535qu“
82 eeewaa as Sea 00% see 3 SEE a a. :3 case 22
e a ca 0.2
82 case, eea =3 ooem a a... me Enema 5 Swansea 29..
n
82 as; es... 2: ooem A he 0.? a No“: 8% 8m 9 eaaeaxm
a me_ e e. as 0.3 a a
82 sneer e5 =3 0.8 me has neaaaeea 23 a Susana
e A Le 0.2
33 gonzo.» use so.“ Doom 6 3550 co>>o>v e\ew.¢ 2 230qu S§MSA8§E .4
353.3% cane—=30 :8: 3253:. $9.5 Ema—33232

 

azeee a." sea.

52

treatment with a commercial sanitizer (0.1% peroxyacetic acid) also increased oxidative
tolerance of E. coli 0157:H7. Starvation of P. putida resulted in enhanced resistance to
ethanol, heat shock and osmolarity (Givskov et al. 1994). Bang and Drake (2002)
observed that starvation increased fieeze-thaw resistance of V. vulnificus, while Leenanon
and Drake (2001) observed an increase in freeze-thaw resistance of E. coli 0157 :H7
following both starvation and acid adaptation. Bollman et al. (2001) observed that cold
shocking E. coli 0157:H7 in milk, whole egg or sausage, but not beef or pork, enhanced
survival during frozen storage. Adaptation of S. Typhimurium to pH 5.8 reportedly
increases tolerance to osmotic (salt) stress, crystal violet and an activated lactoperoxidase
system (Leyer and Johnson 1993). Lou and Yousef (1997) found that adaptation of L.
monocytogenes to 5% (vol/vol) ethanol significantly increased resistance to lethal doses
of acid, ethanol and NaCl. Similarly, adaptation to acid (pH 4.5-5), ethanol, 500 ppm
H202, 7% (wt/vol) NaCl and heat (1 h at 45°C) increased resistance to H202.

The response to a given stress differs among bacterial Species. While Lou and
Yousef (1997) showed that both acid and ethanol adaptation increased the resistance of L.
monocytogenes to lethal doses of acid, adaptation of S. Typhimurium to stresses other
than acid did not protect against lethal acid treatment (Lee et al. 1995). Bunning et a1.
(1990) found that a 48°C heat shock significantly increased thermotolerance at 578°C for
S. Typhimurium but not for L. monocytogenes. Mazzotta (2001) determined that while
acid adaptation (pH 5, 18-24h) increased the heat resistance at 56°C, 58°C and 60°C for
L. monocytogenes, E. coli 0157:H7 and two Salmonella composites (S. Gaminara,
Rubislaw and Hartford, and S. Typhimurium and Enteritidis) in apple, orange and white

grape juices adjusted to pH 3.9, the increase in heat resistance was greater for L.

53

monocytogenes and E. coli 0157:H7 than for Salmonella.

The effect of adaptive pre-treatments can also vary between different bacterial
strains. Cheng et al. (2002) found that acid adaptation at pH 5 for 4 h increased the
thermal tolerance at 52°C of two strains of E. coli 0157:H7, but not of a third strain.
Similarly, starvation was shown to increase the heat resistance of only two of three E.
coli 0157:H7 strains (Rowe and Kirk 2000). These observed differences for species and
strains may be due to inherent degrees of heat or other stress resistance.

The magnitude and nature of a protective stress response also varies with the
treatment. Volker et al. (1992) found that treatment of B. subtilis with low salt was less
effective at inducing thermotolerance than a preceding heat shock. Jenkins et al. (1988)
reported that the oxidative and thermal resistance of starved E. coli cells was greater than
for heat-adapted cells, with resistance to H202 and heat increasing as the starvation period
increased. Looking at L. monocytogenes, Lou and Yousef(1996) also found that the
magnitude of the increase in thermotolerance provided by starvation, acid, ethanol or
H202 depended on the duration of starvation (i.e., the longer the starvation period, the
greater the increase in heat resistance), the pH level and the concentration of ethanol or
H202. The type of acid can also affect the outcome, as Farber and Pagotto (1992)
observed that, while hydrochloric was effective at inducing thermotolerance in L.
monocytogenes, acetic and lactic acids were not. For heat-shocked S. Typhimurium, the
degree of thermal resistance and rapidity of its onset was shown to increase with
increasing heat shock temperatures (Mackey and Derrick 1986). While cells of L.
monocytogenes incubated at pH 12 for 45 min were significantly more heat resistant at

56°C than those incubated at pH 7.3, cells incubated at pH 9, 10 or 11 did not exhibit this

54

enhanced thermotolerance (Taormina and Beuchat 2001). The strength of the agent used

to adjust the pH tol2 also had an effect on thermotolerance.

2.3.7 Virulence

The presence of injured microorganisms in food gives rise to significant public
health concerns. Injury may initially leave cells undetected during quality control checks
and at critical control points during manufacture. However, subsequent repair in the food
may allow for growth and its consequences, including spoilage and the production of
toxins and other virulence factors (Johnson and Busta 1984). AS an example, three
virulence factors of E. coli 0157:H7, verotoxin (VT)1, VT2 and the attaching and
effacing gene, eae, were shown to be retained following starvation and heat stress (Rowe
and Kirk 2000). According to Singh and McFeters (1987), virulence of Y. enterocolitica
in orally inoculated mice was also unaffected by chlorine stress. A bacterium’s
pathogenicity or virulence may be considered the ultimate expression of its ability to
repair injury (Dillon and Bezanson 1984).

Mekalanos (1992) defines virulence determinants as those factors contributing to
infection and disease, but not to general “housekeeping” firnctions. There is not always a
clear line of distinction between the two, but virulence genes, to some degree, are part of
an adaptive response to stresses encountered in a host (IFT 2002). Many of the stresses
that are intrinsically part of a host’s defense system are similar to those encountered in
the natural environment (see Figure 2.2). Exposure to stresses in both natural
environments and food processing facilities may be seen by pathogenic microorganisms

as a signal for the expression of virulence factors (Lou and Yousef 1997). A strain of S.

55

 

Macrophage

 

 

 

-D Low pH
H Starvation
Lt Heat

'9 Cationic Peptides

l" Oxidative Stress

 

Figure 2.2. Certain stresses encountered by microorganisms in the environment are also

 

 

Gut

 

 

-D Low pH
H Starvation
-) Heat

-D Cationic Peptides

'9 Osmolarity

 

 

 

Epithelial Cells

 

H Starvation

-D Heat

 

imposed by a host’s defense mechanisms (Foster and Spector 1995).

56

 

Enteritidis possessing enhanced acid and heat tolerance was shown to be more virulent
for mice and more invasive for chickens than a non-resistant reference strain (Humphrey
et al. 1996).

Expression of many virulence factors is dependent on environmental cues
(Mekalanos 1992, Knachel and Gould 1995). Several environmental conditions have
been identified that induce expression of Spv (Salmonella plasmid virulence) proteins,
including glucose starvation, low pH, elevated temperature and iron limitation (D’Aoust
1997, Valone et al. 1993, Foster and Spector 1995). The spv genes are thought to
facilitate rapid multiplication in host cells and systemic spread and infection of
extraintestinal tissues by several serovars of Salmonella (Typhimurium, Dublin,
Enteritidis) (D’Aoust 1997). An invasion gene of S. Typhimurium, invA, is reportedly
induced by high osmolarity (Mekalanos 1992, Archer 1996) and expression of
listeriolysin, a major virulence factor of L. monocytogenes, by heat shock, oxidative
stress and transition to the stationary phase (Sokolovic et al. 1990, Mekalanos 1992,
O’Driscoll et al. 1996, Datta and Kothary 1993, Sokolovic et al. 1993). Production of
thermostable direct hemolysin, a major virulence factor of V. parahaemolyticus, is
enhanced by heat shock at 42°C (Wong et aL 2002).

Temperature-regulated virulence factors have been identified in enteroinvasive E.
coli (Berlutti et al. 1998), S. flexneri (Maurelli 1989, Mekalanos 1992, Dorman et
al.1990), L. monocytogenes (Gahan and Hill 1999, Leimeister-Wachter et al. 1992) and Y.
enterocolitica (Skumik and Toivanen 1992), and heat shock has been linked to virulence
in L. monocytogenes (Archer 1996, Mekalanos 1992), S. Typhimurium (Archer 1996,

Foster and Spector 1995) and Shigella species (Suni12000). As pathogens traverse from

57

the natural environment, through contaminated food, water or insect vectors, into
mammalian hosts, the sudden increase in body temperature triggers a strong heat shock-
like response, which is intensified when host defense mechanisms (including fever) are
encountered (Lindquist 1992).

The facultative intracellular pathogen Y. enterocolitica experiences a global stress
response to the hostile environment of macrophages, including the induction of heat
shock proteins (Yamamoto et al. 1994). While no set of specific stess-induced proteins is
induced in its entirety within the macrophage environment, S. Typhimurium also
expresses a number of stress proteins, including HSPS (Foster and Spector 1995, Abshire
and Neidhardt 1993, Buchmeier and Heffron 1990). To a certain degree, HSP synthesis
appears to protect pathogens from host defense mechanisms and HSPS themselves may
well be ubiquitous evasion factors of pathogens (Kaufmann 1989). Due perhaps to their
wide distribution and homology among different species, HSPS are often dominant
antigens of the immune response (Zugel and Kaufmann 1999, Buchmeier and Heffron
1990, Lindquist 1992). Host cells can also synthesize HSPS, and an extended assault on
the immune system with HSP antigens that are similar in the host and infecting pathogens
may promote autoimmune disease (Zugel and Kaufmann 1999, Kaufmann 1989).

Acid tolerance is thought to enfmnce virulence in one of two ways. Resistance to
strong acid conditions facilitates survival in the stomach, thereby decreasing a pathogen’s
necessary infective dose. Resistance to moderate acid conditions improves pathogen
survival in foods dependent on low pH for microbial inactivation (Leenanon and Drake
2001). Acid tolerance of E. coli 0157:H7 likely contributes to its low infective dose.

Acid sensitive strains of S. Typhimurium exhibit reduced virulence (Foster 1995),

58

whereas acid tolerant mutants of L. monocytogenes exhibit increased virulence in the
mouse model (O’Driscoll et al. 1996). For many pathogens, acid tolerance seems to
enhance survival in the host macrophage (Archer 1996, Gahan and Hill 1999).

Due to the limited oxygen availability in the small intestine, anaerobic stress is
hypothesized to enhance virulence of pathogens invading the gastrointestinal tract
(Kapoor et al. 2002). Anaerobiosis can reportedly induce the invasion phenotype in S.
Typhimurium, confirmed by the restricted invasion of aerobically grown cells
(Mekalanos 1992, Foster and Spector 1995). Exposure of S. Typhi to anaerobic
conditions also enhances virulence (Kapoor et al. 2002). Osmotic stress may prepare
microorganisms for survival within a host, as the expression of some virulence factors
may be enhanced in the osmolarity range of host tissues. Osmoregulation of virulence is
noted for several microorganisms, including S. Typhimurium, S. flexneri, and Vibrio
cholerae (Mekalanos I992, Sleator and Hill 2002).

Preceding examples indicate alterations in cellular physiology, including stress
protein synthesis, in response to environmental stresses may strongly impact virulence.
A bacterium’s capability of successfully handling environmental stresses in part defines
its virulence, as the response to such stresses often includes the expression and control of
various virulence factors (Archer 1996). The consequences of this has led Archer (1996)
to question if a “reduction in preservation might not in fact lead to a reduction in the
immediate virulence of certain pathogens, and, additionally, to a lowering of the rate of
emergence of new or better host-adapted pathogens.”

2.3.8 Recovery of Injured Microorganisms

Recovery is the “getting back” of injured or otherwise treated microorganisms

59

fi'om their environment (Harris 1963). Some of these organisms will be repaired and
others will not. Damage to microorganisms cannot be understood and characterized
unless the injured population has been recovered. The return of colony-forming ability,
particularly on selective media, is recognized as recovery fiom initial injury and indicates
repair of the structural and metabolic functions damaged by stress (Przybylski and Witter
1979). It is very important that appropriate techniques and media are used to
quantitatively recover all cells, regardless of injury status, following exposure to stressfirl
treatments, since any of these cells could potentially induce illness (Brashears et al.
2001). Historically, the term “resuscitation” has described treatments or processes used
for complete recovery (Mossel and van Netten 1984, Allen et al. 1952).

Resuscitation requires incubation under optimal conditions in order to facilitate
the repair of injured bacteria (Hurst 1984). Any generalizations regarding the recovery of
injured microorganisms must be made with caution, as optimal conditions vary widely
between different species, strains and stress treatments. Incubation at temperatures well
above or below the optimum can significantly reduce recovery of injured cells (Mackey
et al. 1994). For example, better recovery of heat-injured L. monocytogenes occurred at
temperatures between 20°C and 25°C than at 5°C (Mackey et al. 1994) or >30°C (Teo
and Knabe12000). Although in many instances the optimum repair temperature has been
reported to be below the optimum growth temperature (Mackey 2000), the effect of
incubation temperature varies with the organism and the type of stress (Mossel and van
Netten 1984). Although the recovery of certain strains of irradiated E. coli was improved
at temperatures below 37°C, temperatures above 37°C are also reportedly favorable for

recovery of irradiated E. coli (Harris 1963). For microorganisms for which cold

60

enrichment has been recommended (e.g., L. monocytogenes, Y. enterocolitica), it has
been suggested that incubation at 25°C for 2 to 3 h before incubation at 4°C could
produce a favorable environment for the repair and recovery of injured cells (Restaino et
al. 1980).

Stressed cells are affected by the levels of several nutrients, including Ca”, Mg”,
K+, carbohydrates and amino acids (Mossel and van Netten 1984). Mg2+ is important in
stabilization of the ribosome, which can be compromised by many environmental
stresses. Although injured cells typically repair faster in rich as opposed to minimal
media (Mackey 2000), the effect of nutrients in the recovery medium is likely dependent
on a number of additional factors, including the phase of growth when the cells are
stressed and the medium used to grow the cells before stressing (Mossel and van Netten
1984, Harris 1963, Mackey 2000). As a result, the effect of minimal medium recovery
must not be overlooked. In some instances, recovery of stressed cells is dramatically
increased on minimal salts media. Gomez and Sinskey (1975) determined that minimal
medium recovery of heated S. Typhimurium depended on the presence of air, and several
authors (Mackey and Derrick 1986, Mackey 2000, Stephens et al. 2000) suggested that
minimal medium recovery is an efl'ect of oxidative stress, including sensitivity to
peroxides in rich media.

By eliminating the challenges of oxidative stress, some authors have improved the
recovery of sublethally injured microorganisms with anaerobic incubation. Bromberg et
al. (1998) recovered 103 CF U/mL when heat-treated cells of E. coli 0157:H7 were
recovered aerobically, but 10° CFU/mL when they were recovered anaerobically.

Anaerobically held cells gradually regained their ability to grow in the presence of

61

oxygen, perhaps due to the synthesis of enzymes that detoxify oxygen radicals. Xavier
and Ingham (1993) reported that anaerobic incubation recovered significantly more heat-
treated cells of S. Enteritidis than did aerobic incubation, and strictly anaerobic conditions
recovered significantly more heat-injured cells of L. monocytogenes compared to
aerobically incubated controls (Knabel et al. 1990).

Other factors must also be considered when optimizing the recovery of injured
microorganisms, including pH, presence of reducing agents or H202-degrading
compounds and salt concentration, among others. Although often included in culture
media formulations, sodium chloride can impair repair and recovery of injured
microorganisms (Harris 1963). However, inhibition fi'om the recovery medium may be
partially overcome by the protective effect from the heating medium (Matias et al. 2001).
Other salts, such as ferric chloride, can enhance recovery of phenol-treated E. coli (Harris
1963). The method of inoculating media must also be considered, because if not
carefully done, pour plating can result in additional thermal injury and a reduction in the
recovery of stressed cells (Mossel and van Netten 1984, Czechowicz et al. 1996).

Even if optimum repair/recovery conditions are used, the extended lag period
characteristic of injury will still be observed. The length of the required repair or lag
time depends on the bacterial species, strain and the type/severity of injury. For this
reason, any attempt to shorten or eliminate pre-enrichment periods, which are designed to
allow for resuscitation of injured cells or to enable target organisms to reach a detectable
level, are often met with false-negatives (Andrews 1986, Palumbo 1984, Flowers et al.
1992, Mackey 2000). Rapid methods employing a non-selective enrichment step might

be affected by this, particularly in food matrices where the overgrowth of competing

62

bacteria must be considered; however, more research is needed, since studies on the
impact of injury on incubation conditions required to reach detectable microbial levels
are limited (Mackey 2000). Direct selective enrichment of highly contaminated products
is not uncommon. This method seems to be inappropriate for detecting injured
microorganisms, particularly in the presence of large numbers of competitive flora, and
the incorporation of a non-selective pre-enrichment can significantly increase recovery

(D’Aoust 1981, Blackburn and McCarthy 2000).

63

CHAPTER 3
EFFICACY OF SODIUM PYRUVATE AND OXYRASE® FOR

RECOVERING HEAT-, COLD-, AND STARVATION-INJURED SALMONELLA

3.1 Abstract

Recovery of heat-injured, cold-shocked or otherwise stressed cells is often limited
by choice of media and environmental conditions, including oxygen exposure. Reliable
enumeration of these injured and potentially more resistant cells is particularly important
in thermal death-time studies so as to not underestimate D-values, which can lead to a
false sense of security in cooked products. This study assessed the efficacy of adding two
reducing agents, Oxyrase® and sodium pyruvate, to a non-selective culture medium for
recovery of heat-, cold- and starvation- stressed Salmonella. Using TSA-YE
supplemented with Oxyrase®, 0.1% sodium pyruvate and 0.6% sodium pyruvate,
recovery of heat-shocked, cold-shocked and starved cells ranged from 89.6 to 90.2%,
94.9 to 95.1% and 99.2 to 101.6%, respectively, and was not significantly different from
unsupplemented TSA-YE. From these findings, it is apparent that, under the specified
conditions, supplementation of TSA-YE with Oxyrase® or sodium pyruvate is not

required to enhance the recovery of heat-shocked, cold-shocked or starved salmonellae.

3.2 Introduction
Salmonella is a major cause of gastroenteritis in humans when contaminated food
is undercooked or eaten raw. In the United States, an estimated 1.4 million cases of

salmonellosis occur each year (Frenzen et al. 1999) at a cost of $2.4 billion (Frenzen

64

2002). Outbreaks associated with Salmonella have been reported in a wide range of
foods, including meat, poultry, eggs, fi'uits, vegetables, dairy products and chocolate
(D’Aoust 1997).

Physical and chemical treatments intended to inactivate Salmonella can lead to
injury, with such cells harboring various sublethal lesions that adversely affect the
permeability barrier and other physiological properties. As a result, injured cells
typically are unable to undergo repair and proliferation on solid media containing
selective components (Hurst 1977, Gilbert 1984, Hurst 1984, Semanchek and Golden
1998, Ray 2001). Repair and subsequent growth is also inhibited by selective agents in
enrichment broth (Patel and Beuchat 1995). Consequently, the potential for obtaining
false-negative results or underestimation of injured target cells is increased (Blackburn
and McCarthy 2000). Even if undetected in food products, injured cells are still virulent
if ingested by susceptible individuals (Reissbrodt et al. 2002).

Methods developed to allow injured organisms to repair before exposure to a
selective medium do not always take into account potential inhibition from the non-
selective medium (McDonald et al. 1983). Recovery of injured organisms is often highly
sensitive to factors that do not affect, to the same degree, the recovery of uninjured
organisms, with non-selective media differing in their ability to resuscitate injured cells
(Stephens et al. 2000). Enhanced recovery in simple media, a phenomenon known as
minimal medium recovery, was suggested to result from sensitivity to peroxides present
in rich media (Stephens et aL 2000, Mackey and Derrick 1986). Autooxidation of
reducing sugars in the presence of phosphate during autoclaving can result in the

production of toxic oxygen species, particularly hydrogen peroxide (H202) (Stephens et

65

al. 2000, Baylis et al. 2000a). Media containing manganese or citrate can also
autooxidize to form peroxides. If peroxides accumulate in media in the presence of
superoxide radicals (02"), hydroxyl radicals ('OH) will form. Alternatively, hydroxyl
radicals can form when H202 reacts with Fe” (McDonald et al. 1983, Sub and Knabel
2000). Reactive oxygen species (ROS), including O2", H202 and ’OH, can be produced
endogenously as a consequence of the normal metabolic reduction of molecular oxygen
to water (Stephens et al. 2000).

Oxidative stress caused by elevated levels of ROS can damage cellular
components (Storz and Zheng 2000) and possibly inhibit repair and recovery of injured
microorganisms in culture media. Injured cells, particularly those that are heat-injured,
often have diminished activity of superoxide dismutase (SOD) and catalase, the enzymes
instrumental in neutralizing the harmful effects of oxygen radicals (Suh and Knabel 2000,
McDonald et al.1983). Research on the stability, in response to different environmental
stresses, of catalase and SOD in Salmonella is limited. However, a number of studies
have attempted to enhance recovery of sublethally injured microorganisms by
supplementing various media with reducing agents, 02 scavengers and/or H202-
degrading agents (Suh and Knabe12000, Patel et al. 1995, McDonald et al. 1983).

Incorporating sodium pyruvate in culture media reportedly improved recovery of
heat-injured Escherichia coli 0157:H7 (Czechowicz et al. 1996), Listeria monocytogenes
(Patel et al. 1995, Busch and Donnelly 1992), Salmonella Seftenberg (Rayman et al.
1978), Shigellaflexneri (Smith and Dell 1990) and Staphylococcus aureus (Hurst et al.
1976, Martin et al. 1976), as well as heat- and freeze-injured E. coli (McDonald et al.

1983). Addition of catalase also enhanced recovery of heat-injured and freeze-dried S.

66

aureus (Martin et al. 1976, Flowers et al. 1977), acid-injured E. coli (Martin et al. 1976),
and heat-injured S. Seftenberg (Rayman et al. 1978), L. monocytogenes (Patel et al. 1995)
and E. coli (Mackey and Seymour 1987). The benefits of other agents, including o.-
ketoglutaric acid (degrades H202) (Mizunoe et al. 1999), sodium thioglycolate (blocks
formation of H202) (McDonald et al. 1983, Xavier and Ingham 1993, Yu and Fung
1991), L-cysteine (reducing agent) (Knabel and Thielen 1995) and Oxyrase® (oxygen—
reducing membrane fraction from E. coli) (Stephens et aL 2000, Yu and F ung 1991,
Reissbrodt et al. 2002, Patel and Beuchat 1995, Patel et al. 1995) have also been reported.
Research on the elimination of oxidative stress for the enhanced recovery of
injured Salmonella is limited, particularly when non-thermal injury is considered. The
aim of the present study was to assess the impact of sodium pyruvate and Oxyrase® on

the recovery of heat-, cold- and starvation-stressed Salmonella.

3.3 Materials and Methods
3.3.1 Bacterial Strains

The following eight strains of Salmonella were used: S. Thompson FSIS 120
(chicken isolate), S. Enteritidis H3527 (clinical isolate phage type 13A), S. Enteritidis
H3502 (clinical isolate phage type 4), S. Typhimurium H3380 (DT104 clinical isolate), S.
Hadar MF60404 (turkey isolate), S. Copenhagen 8457 (pork isolate) and S. Heidelberg
F503 8BG1 (clinical isolate), all obtained from V.K. Juneja (Agricultural Research
Service, Eastern Regional Research Center, USDA-ARS, Philadelphia, PA), and S.
Typhimurium 420, a DT104 strain isolated from a pork processing facility. All strains

were maintained at -70°C in Trypticase soy broth (TSB) (Difco Laboratories, Detroit,

67

MI) containing 10% (v/v) glycerol (J.T. Baker, Mallinckbrodt Baker, Inc., Phillipsburg,
NJ).
3.3.2 Culture Preparation

Each culture was propagated by transferring one loopfirl of the frozen cell
suspension into 9 mL of TSB containing 0.6% yeast extract (TSB-YE) (Difco). Cultures
were subjected to a minimum of two consecutive overnight transfers (3 7°C, 18-24 h) in
TSB-YE before use.
3.3.3 Heat Shock Treatment

An 8-strain Salmonella cocktail was prepared by combining 3 mL of the
individual overnight cultures in a sterile 50 mL centrifirge tube (Clear Propylene, Plug
Seal Cap, Corning Inc., Corning, NY), centrifuging at 10,000 rpm for 15 min at 4°C
(Super T21, Sorvall® Products, L.P., Newtown, CT) and re-suspending the pellet in 3 mL
0. 1% peptone (Difco) to obtain a culture containing approximately 1010 CFU/mL. Heat-
injured cells were obtained using the methods of Mathew and Ryser (2002) and Busch
and Donnelly (1992), with minor modifications. In this procedure, 200 mL of TSB-YE
(in a 2800 mL wide-mouth Fernbach flask) was tempered to 54°C in a shaking (30 rpm)
water bath (Reciprocal Shaking Bath, Precision Scientific, Winchester, VA), inoculated
with 2 mL (~10‘° CFU/mL) ofthe Salmonella cocktail and heated at 54°C for 30 min.
After 30 min, heat-shocked cells were immediately cooled in an ice-water bath for 3 to 5
min.

The initial temperature for heat shock was 48°C, which was also used by Mackey
and Derrick (1986, 1987a, 1987b) and Bunning et al. (1990) for Salmonella. However,

for the individual Salmonella strains in the cocktail, the extent of injury observed at 48°C

68

(6-54%), as well as at 52°C (23-68%), was decided to be insufficient for the current study
(see Appendix B, Table 3.1). At 56°C, a 30 min heat shock was lethal for 4 of the 8
strains. As 54°C resulted in approximately 90% injury in preliminary studies, it was this
temperature that was incorporated into the heat shock procedure.

3.3.4 Cold Shock Treatment

The 8-strain Salmonella cocktail was prepared as previously described for heat
shock, except the pellet was re-suspended in 4 mL 0.1% peptone to obtain a culture
containing approximately 1010 CFU/mL. Cold-stressed cells were obtained by
inoculating 200 mL TSB-YE (in a 500 mL Erlenmeyer flask) previously tempered to 4°C
with 3 mL (~10lo CFU/mL) of the Salmonella cocktail. This cell suspension was then
gently agitated for 1 min and held quiescently at 4°C for 2 h.

The 2 h time period for the cold shock treatment was chosen because preliminary
work showed, that by 4 h, injury had decreased for the individual Salmonella strains in
the cocktail (see Appendix B, Table B.2).

3.3.5 Starvation Treatment

The method for starvation was similar to that of Dickson and Frank (1993). The
8-strain Salmonella cocktail was prepared as previously described for heat shock, except
the pellet was re-suspended in 24 mL Butterfield’s phosphate buffer (BPB), pH 7.18
(prepared according to the Bacteriological Analytical Manual (FDA 1998)). To prevent
carry-over of nutrients, the culture was centrifirged again at 10,000 rpm for 15 min at 4°C
and the pellet re-suspended in 3 mL BPB to obtain a cell suspension containing

approximately 1010 CFU/mL. Cells were starved by transferring 1 mL of the Salmonella

69

cocktail into 99 mL BPB in a 160-mL screw-capped glass bottle, gently agitating the cell
suspension for l min and holding it quiescently at 4°C for 10 d.

Preliminary work showed that 5 d of starvation resulted in insufficient injury for
the individual Salmonella strains in the cocktail and by 14 d, injury had begun to decrease
from those levels observed at 10 d (see Appendix B, Table 8.3). AS a result, the 10 (1
period was incorporated into the starvation procedure. To verify that the injury observed
at 10 d was due to starvation in BPB and not to the effects of low temperature, injury was
also determined after 10 d at 4°C for cultures of the Salmonella strains grown and held in
TSB-YE (see Appendix B, Table 3.4).

3.3.6 Enumeration of Cells

Four different non-selective recovery media were used: Trypticase soy agar
(Difco) containing 0.6% yeast extract (TSA-YE), TSA-YE containing 0.1% (w/v) sodium
pyruvate (Sigma-Aldrich Co., St. Louis, MO) (TSA-YE + 0.1% NaPyr), TSA-YE
containing 0.6% (w/v) sodium pyruvate (TSA-YE + 0.6% NaPyr) and TSA-YE
containing Oxyrase® (Oxyrase, Inc., Mansfield, OH). Oxyrase® was added after
sterilizing TSA-YE at a final concentration of 10% (v/v) as recommended by the
manufacturer, whereas sodium pyruvate was added to TSA-YE before autoclaving.
Bismuth sulfite agar (Difco) (BSA) was used as the selective medium to assess sublethal
injury. Before and after stressing, the cells were enumerated by serial dilution in 0.1%
peptone followed by spiral plating (Autoplate® 4000, Spiral Biotech, Inc., Norwood, MA)
in duplicate on the five different recovery media. All plates were counted after 18 to 24 h

of incubation at 37°C.

70

Percent injury following the individual treatments was determined according to
the formula: % injury = [(count on non-selective agar — count on selective agar)/count on
non-selective agar] x 100. The unsupplemented, non-selective medium, TSA-YE, and
the selective BSA were used to calculate injury. Percent injury was calculated for each
replication of each treatment, with the results reported as averages.

3.3.7 Data Analysis

All experiments with Oxyrase® were repeated three times, whereas experiments
with TSA-YE + 0.1% NaPyr and TSA-YE + 0.6% NaPyr were repeated nine times for
heat shock and six times for cold shock and starvation. Results are reported as averages.
The significance of bacterial recovery was assessed with a mixed model analysis of

variance (ANOVA) (Proc Mixed, SAS® Version 8.02, SAS Institute, Inc., Cary, NC).

3.4 Results
3.4.1 Sublethal Injury

The individual stress treatments produced Salmonella cocktails containing 86.83
:t 8.09% (mean injury d: stande deviation) heat-injured, 68.25 i 8.94% cold-injured,
and 85.58 :t 8.23% starvation-injured cells, respectively.
3.4.2 Recovery of Injured Cells

Following heat shock, TSA-YE recovered 89.63 :I: 2.25% of the original cells
present, while TSA-YE + 0.1% NaPyr, TSA-YE + 0.6% NaPyr and TSA-YE + Oxyrase®
recovered 90.18 i 1.67%, 90.01 i 2.32% and 89.56 3: 0.41%, respectively (Table 3.1).
The application of heat significantly decreased (P<0.05) the populations on each

individual medium, indicating that recovery was affected by the presence of dead and/or

71

injured cells. After heat shock, differences between the populations recovered on the
different non-selective media were not statistically significant (P>0.05). The same was
also true before heat shock. The ANOVA results for heat shock, as well as cold shock
and starvation, can be found in Appendix B (Table 8.6).

After cold shock, TSA-YE recovered 94.94 d: 1.85% of the original population,
while TSA-YE + 0.1% NaPyr, TSA-YE + 0.6% NaPyr and TSA-YE + Oxyrase®
recovered 95.10 :l: 3.10, 94.93 i 2.39% and 95.06 i 2.39%, respectively (Table 3.2).
Cold shock significantly decreased (P<0.05) the numbers recovered on each medium
tested; however, before and after cold shock, the differences between the populations
recovered on the four non-selective media were not statistically significant (P>0.05).

Following starvation, TSA-YE recovered 101.64 3: 5.92% of the original
Salmonella population, while TSA-YE + 0.1% NaPyr, TSA-YE + 0.6% NaPyr and TSA-
YE + Oxyrase® recovered 99.22 a 5.68%, 100.00 a. 8.20% and 100.25 r 8.03%,
respectively (Table 3.3). There were no significant differences (P>0.05) in the initial and
post-starvation populations, indicating no treatment effect. As with heat shock and cold
shock, the differences between the populations recovered on the four non-selective media
were not statistically Significant (P>0.05). Recovery following starvation, as well as heat

and cold shock, is shown in Figure 3.1.

72

Table 3.1 The effect of sodium pyruvate and Oxyrase® on recovery of heat-injured
Salmonella.

 

 

 

Initial Post-Heat Shock
Medium Log (CFU/mL) Lg (CFU/mL) % Recovery”
TSA-YE 10.40 i 0.31 9.32 d: 0.34 89.63 :t 2.25
TSA-YE 10.29 :I: 0.32 9.38 :I: 0.28 90.18 i 1.67
+ 0.1% NaPyr
TSA-YE 10.37 i 0.33 9.36 at: 0.29 90.01 i 2.32
+ 0.6% NaPyr
TSA-YE 10.02 a: 0.02 8.97 :h 0.03 89.56 :1: 0.41
+ Oxyrase®

 

Arithmetic Means d: Standard Deviations
° Percent recovery was calculated with initial counts on TSA-YE as 100%.

Table 3.2 The effect of sodium pyruvate and Oxyrase® on recovery of cold-stressed

 

 

 

Salmonella.

Initial Post-Cold Shock
Medium Log (CF U/mL) Log (CFU/mL) % Recovery”
TSA-YE 10.66 i: 0.44 10.11 :1: 0.38 94.94 i 1.85
TSA-YE 10.6021:O.41 10.13 $0.25 95.10zt3.10
+ 0.1% NaPyr
TSA-YE 10.66 i: 0.45 10.11 :1: 0.29 94.93 :1: 2.39
+ 0.6% NaPyr
TSA-YE 10.22 :1: 0.21 9.78 :1: 0.01 95.06 :E 2.39
+ Oxyrase®

 

Arithmetic Means i Standard Deviations
a Percent recovery was calculated with initial counts on TSA-YE as 100%.

73

Table 3.3 The effect of sodium pyruvate and Oxyrase® on recovery of starved

 

 

 

Salmonella.

Initial Post-Starvation
Medium Log (CFU/mL) Log (CFU/mL) % RecoveryI
TSA-YE 10.59 i 0.04 10.76 i 0.62 101.64 :1: 5.92
TSA-YE 10.57 i 0.03 10.50 :t 0.56 99.22 :I: 5.68
+ 0.1% NaPyr
TSA-YE 10.56 i 0.03 10.58 :t 0.86 100.00 :h 8.20
+ 0.6% NaPyr
TSA-YE 10.55 5: 0.02 10.60 i 0.85 100.25 :1: 8.03
+ Oxyrase®

 

Arithmetic Means :1: Standard Deviations
‘ Percent recovery was calculated with initial counts on TSA-YE as 100%.

 

110—

 

8

.—

1: 105*
O

I:

'3 100~
is

E. 95—
E

I:

s 90-
"6

3° 854

 

Recovery of Sublethally Injured Salmonella on
Supplemented and Unsupplemented Medla

 

Heat

Cold
Stress Treatment

 

 

 

I TSA-YE

I TSA-YE +
Oxyrase®

DTSA-YE + 0.1%
sodium pymvate

lTSA-YE + 0.6%
sodium pyruvate

 

 

Starvation

 

 

Figure 3.1 Recovery of sublethally injured Salmonella on unsupplemented and
supplemented TSA-YE. Values are plotted as averages, with standard deviations shown

as error bars.

74

3.5 Discussion

While it is common for authors to describe how they calculated the proportion of
injured cells in a culture following a stress treatment, many fail to report actual numerical
values. Although this limits the comparisons of our data to earlier findings, several
observations can be made. In the current work, between 68% and 87% of the Salmonella
cocktail was injured by heat shock, cold shock or starvation. With the exception of S.
Montevideo FSIS 051 in place of the DT104 laboratory strain S. Typhimru'ium 420,
Smith (2000) used the same 8-strain Salmonella cocktail and found that 90% of the
population was injured following heat shock in TSB at 56°C for 30 min. While this is in
close agreement with our findings, Baylis et al. (2000a) observed a much greater range of
injury (59% and 98%) in S. Typhimurium CRA 8378 heat-treated in nutrient broth for 25
min at 515°C. For the same heat shock conditions and strain of S. Typhimurium, Baylis
et al. (2000b) calculated 86% and 95% injury in two cultures, but observed little or no
injury in a third culture. Using other test organisms, >99% of Listeria monocytogenes
was injured by heating in tryptose phosphate broth at 56°C for 30 min (Mathew and
Ryser 2002), 90% by heating in TSB at 54°C for 30 min (Pascual et al. 2001) and 73%
by heating in tryptose phosphate broth at 52°C for 18 min (Patel et al. 1995). Percent
injury following starvation in BPB for 5 d at 10°C was 2.1% for L. monocytogenes,
3.55% for Escherichia coli, and non-detectable for S. Typhimurium (Dickson and Frank
1993). This latter finding may differ from the high percent injury by starvation reported
in the present study due to the shorter starvation period used (see Appendix B, Table

8.3). Also, for S. Enteritidis, Boziaris and Adams (2001) observed a lesser degree of

75

injury (<50%) following cold shock in nutrient broth or phosphate-buffered saline for 10
min at 05°C.

The extent of injury following exposure to sublethal stress is dictated by
differences between bacterial species and strains, treatment or stress conditions, the type,
degree and site of injury, the medium in which the test organisms is grown and stressed,
the selective and non-selective media used to assess injury and the growth phase of the
culture (Ray 1989, Ray 2001). Heat, for instance, simultaneously affects the cell
membrane, proteins and nucleic acids (Wuytack et al. 2003), and the degree of membrane
damage and protein denaturation associated with heat could be more extensive than that
associated with other treatments. Also, exponentially growing cells are more susceptible
to injury than are stationary phase cells (Pascual et al. 2001, McMahon et al. 2000), and
at least in the case of heat, injury may be difficult to manipulate and control in the
stationary phase (Stephens et al. 1997). In addition, injury of planktonic cells (i.e., free,
single, grown in broth, etc.) is probably quite different from injury of cells grown on an
agar plate. Growth on an agar plate is perhaps the simplest expression of a biofilm, and
bacterial cells in biofilms express enhanced resistance to heat, chemicals and sanitizers
(Montville 1997, Costerton et al. 1995).

Standard deviations associated with average injury for each treatment were
relatively large, indicating that some degree of variability exists. Since preliminary work
(see Appendix B, Table B.5) indicated that all 8 strains in the cocktail were equally
susceptible to injury by the three treatments, these variations are more likely due to the
inherent heterogeneity present in every bacterial population and the choice of selective

media.

76

All three supplemented media (TSA-YE containing 0.1% sodium pyruvate, 0.6%
sodium pyruvate and Oxyrase®) were as equally effective as unsupplemented TSA-YE
for recovering heat-, cold- and starve-injured Salmonella. McDonald et al. (1983)
reported that tryptone-glucose extract (TGE) agar and violet red bile agar (VRBA)
supplemented with 0.33% or 0.66% sodium pyruvate recovered significantly more
freeze- and heat-stressed E. coli than did their unsupplemented counterparts. TGE and
VRBA with sodium pyruvate at a concentration of 1.2% recovered more fieeze-stressed
cells, but 1.2% sodium pyruvate was only effective in TGE for heat-injured cells. In their
study, McDonald et al. (1983) used pour plating, a method with the potential to impart an
additional thermal stress on the test organism.

Czechowicz et al. (1996) found that recovery of heat-stressed E. coli 0157:H7
increased 10- to 1000-fold using Trypticase soy agar (TSA) and plate count agar (PCA)
containing 1% sodium pyruvate, as compared to tmsupplemented TSA. Since the
supplemented media were plated by the spread plate method and unsupplemented TSA
by the pour plate method, part of the increased recovery with sodium pyruvate was
perhaps due to the lack of exposure to additional thermal stress during spread plating.
Czechowicz et al. (1996) also noted that spread plating on TSA containing 1% sodium
pyruvate did not enhance recovery of heat-stressed cells over that seen using
unsupplemented TSA. In the current study, the impact of the different supplements could
have been outweighed by the impact of surface inoculation of media by spiral plating.

The increased recovery reported by Czechowicz et al. (1996) on PCA with 1%
sodium pyruvate was not due to spread plating alone. The authors suggested that the

presence of glucose and yeast extract in PCA may be as important as the addition of

77

sodium pyruvate. Growth factors available in yeast extract could contribute to the repair
of injured organisms. Although the potential exists for pyruvate to be utilized as an
energy-generating compound by injured organisms, most authors believe its main
function is to degrade H202, and it is unlikely to be used as a nutrient (Mizunoe et al.
1999, Baird-Parker and Davenport 1965). However, if the use of pyruvate (and other
TCA cycle intermediates) were to promote growth, the benefits of the supplement could
be masked by the presence of yeast extract in our non-selective media.

The concentration of the chosen supplement may also affect the recovery of
stressed cells. Nebra et al. (2002) found that sodium pyruvate used at a concentration of
0.66% in selective media enhanced recovery of metabolically injured (3 days of storage
at 11°C) Bifidobacterium cells, while a concentration of 0.33% was not effective.
Mizunoe et al. (1999) tested different concentrations of sodium pyruvate, catalase and a-
ketoglucaric acid in Luria-Bertani agar for improving the recovery of starvation- and low-
temperature-stressed E. coli 0157 :H-strain E32511/HSC. Each supplement had a
particular concentration that afforded the highest recovery. In the case of sodium
pyruvate, 0.1% recovered more cells than did 0.01%, 0.02%, 0.05%, 0.5%, 1%, 1.5% or
2%. Sub and Knabel (2000) also observed differences in recovery of heat-injured
(628°C, 7.5 min) L. monocytogenes with different concentrations of Oxyrase® in
Pennsylvania State University (PSU) broth. Oxyrase® was measured in units (U), with 1
U equivalent to the amount that decreased 1% of the dissolved oxygen per second per mL
of broth as instructed by the manufacturer. Following incubation for 72 h, 5, 10, 20, 40

and 50 U/mL detected 26.7%, 50%, 43.7%, 23.3% and 13.3% of heat-injured cells,

78

respectively. The conclusions of the current work could be further examined using a
wider range of concentrations for both sodium pyruvate and Oxyrase®.

Suh and Knabel (2000) also observed that Oxyrase® at any activity (or
concentration) did not enhance the detection of L. monocytogenes heated at 628°C for 10
min. Results pertaining to Oxyrase® in the literature are mixed, with reports for both
increased recovery (Stephens et al. 2000, Yu and Fung 1991, Baylis et al. 2000b) and no
effect (Tran 1995, Sub and Knabel 2000) with Oxyrase®, the latter of which was also
seen in our study. Oxyrase® is the oxygen-reducing membrane fraction fiom E. coli. The
membrane electron transport chain that it contains could transfer electrons to residual
oxygen in the media, contributing to the generation of O2" (Suh and Knabe12000).
Guidot et al. (1993) determined that the electron transport chain is responsible for
Significant production of 02“, which could hinder recovery of injured cells. Injured
microorganisms, particularly those that are heat-injured, are often deficient in enzymes
required to neutralize toxic oxygen species. This may be one reason why healthy
microorganisms that exhibit superior growth in Oxyrase®-supplemented media have not
been recovered as well in the same media when injured.

The results presented in Tables 3.1, 3.2 and 3.3 include another dimension,
namely a treatment effect. Heat shock significantly decreased the Salmonella populations
on all media tested. More work is needed to determine how much of this effect is due to
the inability to recover injured cells and how much is due to the potential lethality of the
heat shock treatment.

Cold shock also decreased the recovery of Salmonella, with this decrease

attributed to the inability to recover injured cells. Since TSA-YE + 0.1% NaPyr, TSA-

79

YE + 0.6% NaPyr and TSA-YE + Oxyrase® did not recover more heat- or cold-stressed
cells than unsupplemented TSA-YE, the effectiveness of other concentrations or other
reducing agents or oxygen-scavengers warrants further investigation. For starvation,
neither the choice of non-selective recovery media nor starvation itself was significant,
indicating, at the very least, that work involving the given starvation treatment is not

limited by recovery methods.

3.6 Conclusion

OveralL TSA-YE with 0.1% or 0.6% sodium pyruvate or Oxyrase® was no more
effective than unsupplemented TSA-YE at recovering heat-, cold- or starvation-stressed
Salmonella. However, the effects of toxic oxygen species and neutralizing supplements
on the recovery of injured microorganisms are complex. Differences can be observed
with different bacterial species, strains, inoculum levels, stress treatments, the type,
degree and site of injury, type and concentration of supplement, media ingredients and
media preparation and storage conditions (including exposure to light, oxygen and high
temperatures). Great care should be taken to include as many of these factors as possible
when attempting to recover injured microorganisms, because the safety of thermally and
otherwise processed foods is based on the detection of all pathogenic organisms,

including those sublethally injured cells that retain their pathogenicity.

80

CHAPTER 4
THERMAL RESISTANCE OF HEAT-, COLD- AND STARVATION-STRESSED

SALMONELLA IN IRRADIATED COMMINUTED TURKEY

4.1 Abstract

To investigate the effects of sublethal stress on Salmonella thermal inactivation
kinetics, an eight-strain Salmonella cocktail was subjected to heat shock (30 min / 54°C),
cold shock (2 h / 4°C), and starvation stress (10 d in phosphate buffer at 4°C), harvested
by centrifugation and inoculated into irradiated comminuted turkey. Immediately after
stress treatment, the Salmonella cocktails contained 89.02% heat-shocked, 44.71% cold-
shocked, and 67.72% starved cells, respectively. Dance-values for the heat-shocked
cocktail (0.64 min on Trypticase soy agar with 0.6 % yeast extract (TSA-YE), 0.35 min
on xylose lysine deoxycholate (XLD) agar) were significantly higher (P<0.05) than for
the unshocked control (0.41 min on TSA-YE, 0.17 min on XLD), whereas those for the
cold-shocked cocktail (0.38 min on TSA-YE, 0.17 min on XLD) were not significantly
different from the control. Starved cells had the same Dsooc-value on TSA-YE as the
unshocked cocktail, but the Dam-value on XLD was significantly lower (0.14 min).
Although starvation and cold shock were unable to provide a protective effect, heat shock
increased thermal resistance, making it important for the product history and
physiological state of Salmonella to be considered when developing and validating
thermal processes. In addition, Dam-values observed on selective media were
significantly lower than those observed on non-selective media for all stress treatments as

well as the control. As a result, non-selective culture media should be used to assess the

81

response of microorganisms to a thermal challenge, as selective media can underestimate

the number of cells remaining in a population.

4.2 Introduction

In 1999, the United States Department of Agriculture Food Safety and Inspection
Service (U SDA—F SIS) finalized regulations for the production of certain fully and
partially cooked meat and poultry products (FSIS 1999). Rather than dictate step-by-step
processing methods with specific endpoint temperatures, the regulations establish
lethality levels for pathogen reduction as well as set limits for pathogen growth based on
performance standards for lethality, stabilization and product handling. Salmonella
reductions of 6.5 logs in cooked beef, roast beef and cooked corn beef and 7.0 logs in
certain firlly and partially cooked poultry products are now required (FSIS 1999), with a
comparable reduction proposed for all other ready-to-eat (RTE) meat and poultry
products (FSIS 2001).

FSIS lethality standards reflect the destruction of “reference” organisms, whose
elimination or reduction will indicate the elimination or reduction of other pathogenic
organisms. Except for thermally-processed, commercially sterile products, the lethality
performance standards are based on Salmonella. An estimated 1.4 million cases of
salmonellosis occur each year in the United States (Frenzen et al. 1999) at a cost of $2.4
billion (Frenzen 2002). Outbreaks associated with Salmonella have been reported in a
wide range of foods, including meat and poultry. The conditions required for an
appropriate reduction of Salmonella in raw meat and poultry will simultaneously reduce

most other vegetative pathogenic microorganisms, and although Listeria monocytogenes

82

is typically more heat resistant than Salmonella, Listeria-related recalls of RTE meat and
poultry products have almost invariably been traced to post-processing contamination
rather than survival during processing.

To verify compliance with performance standards, FSIS supports the use of
Salmonella cocktails containing relatively heat resistant strains and particularly those
implicated in outbreaks (FSIS 2001). While these cultures are normally prepared under
optimal laboratory conditions, a cocktail containing stressed cells more truly represents
the physiological state of organisms that may enter products from the food processing
environment, with the latter being a far better choice for thermal inactivation and other
challenge studies (Johnson 2003, Juneja and Novak 2003).

Raw meat and poultry is generally perceived as an ideal growth medium for
bacteria; however, ahnost all such contaminating microorganisms have been subjected to
environmental and processing stresses. Physical, chemical and nutritional stresses can
occur at various stages within the farm-to-fork continuum (Yousef and Courtney 2003).
Fermentation, the addition of preservatives and the use of acid washes to decontaminate
carcasses can lead to acid stress (Abee and Wouters 1999, Dickson and Siragusa 1994).
Starvation stress can occur in low nutrient or nutrient competitive environments, which
can include animal carcasses, food, equipment surfaces, walls, floors and water (Lou and
Yousef 1996, Dickson and Frank 1993). Rapid rehydration of foods and exposure to high
salt concentrations can cause osmotic injury (Van Schothorst and Maggie Duke 1984,
Bremer and Kramer 2000), with exposure to certain sanitizers leading to oxidative stress
(Yousef and Courtney 2003). Microorganisms inhabiting refrigerated foods are subject

to cold stress. The cold shock phenomenon occurs when growing bacteria are exposed to

83

a sudden temperature decrease of approximately 15°C or greater (Mackey 1984, Jones et
al. 1996).

Briefly shifting an organism from lower to higher temperatures within or above
their normal growth range (Bunning et al. 1990, Mackey and Derrick 1986, Pagan et al.
1997, Farber and Brown 1990) can lead to heat shock. Conditions favorable for the
development of heat-stressed cells include slow heating (e.g., pasteurization of egg
products), slow cooking (e. g., meats and roasts), holding products on warming trays or
under heat lamps, interrupted cooking cycles due to equipment failure and exposure to
hot acid sprays used on carcasses (Mackey and Derrick 1987b, Murano and Pierson 1993,
Bunning et al. 1990, Juneja and Novak 2003, Yousef and Courtney 2003).

Heat shock (Bunning et al. 1990, Juneja et al. 1998, Linton et al. 1992, Murano
and Pierson 1993, Pagan et al. 1999), as well as starvation (Leenanon and Drake 2001,
Lou and Yousef 1996, Rowe and Kirk 2000, Tolker-Nielsen and Molin 1996) and acid
stress (Farber and Pagotto 1992, Leenanon and Drake 2001, Leyer and Johnson 1993,
Lou and Yousef 1996), reportedly enhance the thermotolerance of S. Typhimurium,
Escherichia coli 0157:H7 and L. monocytogenes. In addition to this enhanced or
acquired thermotolerance, exposure to sublethal stresses may potentiate the virulence of
foodborne pathogens, as the expression of many virulence factors is dependent upon
environmental cues such as low pH and elevated temperatures (Mekalanos 1992, Knachel
and Gould 1995).

Potentially enhanced survival of stressed pathogens in foods greatly impacts the
reliability of current lethality standards, risk assessment, predictive models and HAACP

programs in meeting food safety goals (Johnson 2003). Studies addressing the thermal

84

inactivation of stressed pathogens in meat products rather than laboratory broth systems
are needed to better assess the adequacy of the current USDA-F SIS lethality standards.
Hence, the aim of this study was to investigate the effects of heat-, cold- and starvation-

stress on the thermal inactivation kinetics of Salmonella in comminuted turkey.

4.3 Materials and Methods
4.3.] Bacterial Strains

The following eight strains of Salmonella were used: S. Thompson FSIS 120
(chicken isolate), S. Enteritidis H3527 (clinical isolate phage type 13A), S. Enteritidis
H3502 (clinical isolate phage type 4), S. Typhimurium H33 80 (DT104 clinical isolate), S.
Hadar MF60404 (turkey isolate), S. Copenhagen 8457 (pork isolate) and S. Heidelberg
F5038BG1 (clinical isolate), all obtained from V.K. Juneja (Agricultural Research
Service, Eastern Regional Research Center, USDA-ARS, Philadelphia, PA), and S.
Typhimurium 420, a DT104 strain isolated item a pork processing facility. All strains
were maintained at -—70°C in Trypticase soy broth (T SB) (Difco Laboratories, Detroit,
MI) containing 10% (v/v) glycerol (J.T. Baker, Mallinckbrodt Baker, Inc., Phillipsburg,
NJ).
4.3.2 Culture Preparation

Each culture was propagated by transferring one loopful of fiozen cell suspension
to 9 mL of TSB containing 0.6% yeast extract (Difco) (TSB-YE). Cultures were
subjected to a minimum of two consecutive overnight transfers (37°C, 18-24 h) in TSB-

YE before use.

85

4.3.3 Preparation of the Unshocked Salmonella Cocktail

An 8-strain Salmonella cocktail was prepared by combining 2 rnL of each
overnight (stationary phase) culture in a sterile 50 mL centrifuge tube (Clear Propylene,
Plug Seal Cap, Corning Inc., Corning, NY), centrifuging at 10,000 rpm for 15 min at 4°C
(Super T21, Sorvall® Products, L.P., Newtown, CT) and re-suspending the pellet in 1 mL
0.1% peptone (Difco) so as to contain approximately 1010 CFU/mL. The unshocked
Salmonella culture, designed to serve as the control, was placed on ice a maximum of 2 h
until it was inoculated into comminuted turkey.

4.3.4 Heat Shock Treatment

An 8-strain Salmonella cocktail was prepared by combining 3 mL of the
individual overnight (stationary phase) cultures, centrifuging at 10,000 rpm for 15 min at
4°C and re-suspending the pellet in 3 rnL 0.1% peptone to a concentration of
approximately 10lo CFU/mL. Heat-injured cells were obtained using the method of
Mathew and Ryser (2002) and Busch and Donnelly (1992), with minor modifications. In
this procedure, 200 mL of TSB-YE (in a 2800 mL wide-mouth Fembach flask) was
tempered to 54°C in a shaking (30 rpm) water bath (Reciprocal Shaking Bath, Precision
Scientific, Winchester, VA), inoculated with 2 mL (~1010 CFU/mL) ofthe Salmonella
cocktail and heated at 54°C for 30 min.

The initial temperature for heat shock was 48°C, which was also used by Mackey
and Derrick (1986, 1987a, 1987b) and Bunning et al. (1990) for Salmonella. However,
for the individual Salmonella strains in the cocktail, the extent of injury observed at 48°C
(6-54%), as well as at 52°C (23-68%), was decided to be insufficient for the current study

(see Appendix B, Table B.1). At 56°C, a 30 min heat shock was lethal for 4 of the 8

86

strains. As 54°C resulted in approximately 90% injury in preliminary studies, it was this
temperature that was incorporated into the heat shock procedure.

After 30 min, heat-shocked cells were immediately cooled in an ice-water bath for
3 to 5 min. Thereafter, the suspension was transferred from the Fembach flask to a sterile
250 mL polypropylene centrifuge bottle (Sorvall® Products, L.P., Newtown, CT),
pelleted by centrifugation at 10,000 rpm for 15 min at 4°C and re-suspended in 1 mL
0.1% peptone to obtain a heat-Shocked cocktail containing approximately 109 CFU/mL.
The cocktail was then placed on ice a maximum of 2 h until inoculation into comminuted
turkey.
4.3.5 Cold Shock Treatment

The 8-strain Salmonella cocktail was prepared as previously described for heat
shock, except the pellet was re-suspended in 4 mL 0.1% peptone to obtain a culture
containing approximately 1010 CFU/mL. Cold-stressed cells were obtained by
inoculating 200 mL TSB-YE (in a 500 mL Erlenmeyer flask) previously tempered to 4°C
with 3 mL (~10lo CFU/mL) of the Salmonella cocktail. This cell suspension was then
gently agitated for 1 min and held quiescently at 4°C for 2 h. The 2 h time period for the
cold shock treatment was chosen because preliminary work showed, that by 4 h, injury
had decreased for the individual Salmonella strains in the cocktail (see Appendix B,
Table B2).

The cold-shocked suspension was prepared for inoculation into comminuted
turkey using the same method as for heat shock, producing a culture containing

approximately 1010 CFU/mL.

87

4.3.6 Starvation Treatment

The method for starvation was similar to that of Dickson and Frank (1993). The
8-strain Salmonella cocktail was prepared as previously described for heat shock, except
the pellet was re-suspended in 24 mL Butterfield’s phosphate buffer (BPB), pH 7.18
(prepared according to the Bacteriological Analytical Manual (FDA 1998)). To prevent
carry—over of nutrients, the culture was centrifuged again at 10,000 rpm for 15 min at 4°C
and the pellet re-suspended in 3 mL BPB to obtain a cocktail containing approximately
1010 CFU/mL. Cells were starved by transferring 1 mL of the Salmonella cocktail into 99
rnL BPB in a 160-mL screw-capped glass bottle, gently agitating the cell suspension for 1
min and holding it quiescently at 4°C for 10 d. The starved suspension was prepared for
inoculation into comminuted turkey using the same method as previously described for
heat shock, with the resulting culture containing approximately 109 CFU/mL.

Preliminary work showed that 5 d of starvation resulted in insufficient injury for
the individual Salmonella strains in the cocktail and by 14 d, injury had begun to decrease
from those levels observed at 10 d (see Appendix B, Table B.3). As a result, the 10 d
period was incorporated into the starvation procedure. To verify that the injury observed
at 10 d was due to starvation in BPB and not to the effects of low temperature, injury was
also determined after 10 d at 4°C for cultures of the Salmonella strains grown and held in
TSB-YE (see Appendix B, Table 3.4).

4.3.7 Comminuted Turkey

Skinless turkey breast meat was acquired fiom Michigan Turkey Producers, Inc.

(Wyoming, NH) and transported at 0°C to the Michigan State University Meat

Laboratory. The breast meat was chopped in a bowl chopper (Hobart Manufacturing Co.,

88

Model 841810, Troy, OH) until the temperature increased from 0°C to 13°C. The
comminuted turkey was double-bagged in po lyethylene- laminated nylon pouches,
vacuum packaged in approximately 30 g portions, frozen at -12°C and irradiated (12
kGy) at Iowa State University to eliminate indigenous microflora.

Triplicate samples of irradiated comminuted turkey were tested for sterility by
enriching 1 g of turkey in 9 mL TSB-YE at 37°C for 18-24 h. Following incubation,
three plates of Trypticase soy agar (Difco) containing 0.6 % yeast extract (TSA-YE) were
inoculated by spread plating with 0.33 rnL and incubated at 37°C for 18-24 h.

Fat, moisture and protein contents were determined by AOAC (1996) methods
991.36, 950.46B, and 981.1, respectively. For determination of pH, 10 g of comminuted
turkey were added to 90 g of distilled water and homogenized using a Polytron
homogenizer (Model PT 10/35, Brinkman Instruments, Westbury, NJ) at a speed setting
of 3 for 305. The pH of the homogenized comminuted turkey was measured with a
combination electrode (Model 145, Coming, Medfield, MA). Analyses were carried out
in triplicate.

4.3.8 Inoculation of Comminuted Turkey

Pouches of comminuted turkey were thawed under cold running water
approximately 1 h prior to performing thermal inactivation studies. Using aseptic
procedures, 0.3 mL of the unshocked or stressed Salmonella cocktail were added
dropwise to 10 g of comminuted turkey to achieve an approximate concentration for
unshocked, cold-shocked and starved cells of 108 CFU/g and 107 CFU/g for heat-shocked

cells. The meat was manually mixed in a sterile beaker for 3 to 5 min using sterile

89

gloves. Even distribution was verified by replacing the inoculum with green food dye
(McCormick and Company, Inc., Hunt Valley, MD) and plating sub-samples of the meat.

One g of inoculated meat was transferred aseptically to 5 x 25.5 cm polyethylene
laminated nylon bags (Butcher and Packer Supply Co., Detroit, MI). The bags were
tested for sterility by depositing 9mL TSB-YE in 10 random bags that were then
incubated at 37°C for 18-24 h. Following incubation, 0.33 rnL of this enrichment was
spread-plated on each of three plates of TSA-YE. Background microflora were not
detected.

Bags containing the meat were rolled, with a large glass test tube, between two
pre-measured templates to a thickness of 1mm and then heat sealed using a soldering iron
or the heat selector switch of a Freshlock Turbo IITM Premium Home-use Vacuum Sealer
(Model 1830, Keystone Manufacturing Company, Inc., Buffalo, NY).

4.3.9 Thermal Inactivation

The heat-sealed bags were fit into a rack and submerged in a temperature-
controlled waterbath (NESLAB Instruments, Inc., Newington, NH) set at 605°C to
obtain an actual water temperature of 60°C (DuaLogRTM thermocouple thermometer,
Cole Parmer Instrument Co., Model No. 01100-50, Vernon Hills, IL).

Thermal lag time (the time for the meat to reach within 0.5°C of the waterbath
temperature) was determined by inserting a T-type thermocouple into the geometric
center of a l-g comminuted turkey sample, submerging the sample in the waterbath and
logging the temperature (DuaLong thermocouple thermometer, Model No. 01100-50).

Following 7 trials, the lag time was determined to be 14.57 i: 1.27 s, which was adjusted

90

to 15 s for the thermal inactivation studies. The end of the lag time was identified as
“time zero”, the beginning of the inactivation time period.
4.3.10 Enumeration

Bags of comminuted turkey were removed fiom the 60°C waterbath at 13 or 25 s
intervals and submerged in an ice-water bath. Thereafter, the cooked comminuted turkey
was aseptically transferred to sterile WhirlpakTM bags (Nasco, Ft. Atkinson, WI) and
manually homogenized for 1 min in 9 mL 0.1% peptone. Samples were placed in an ice-
water bath and plated within 6 h. Samples removed at 25 and 13 3 intervals were serially
diluted in 0.1% peptone and spread plated in duplicate on TSA-YE (a medium that will
recover both healthy and injured cells) and xylose lysine deoxycholate (XLD) agar
(Difco) (a medium that recovers uninjured cells), respectively, with the plates counted
after 18-24 h of incubation at 37°C.

Counts also were determined for each cocktail before and after stressing and after
inoculation into comminuted turkey before cooking. Cells were enumerated by serial
dilution in 0.1% peptone followed by spiral plating (Autoplate® 4000, Spiral Biotech,
Inc., Norwood, MA) in duplicate on TSA-YE and XLD. All plates were counted after
18-24 h of incubation at 37°C.

Percent injury after the individual stress treatments and after inoculation into
comminuted turkey was determined for each replication according to the formula: %
injury = [(count on non-selective agar — count on selective agar)/count on non-selective

agar] x 100, with the results reported as averages.

91

4.3.1] Data Analysis

Thermal inactivation of the unshocked Salmonella cocktail was tested six times,
while the heat-, cold- and starve-stressed Salmonella cocktails were tested four times.
The raw data were analyzed with an analysis of covariance model (Proc GLM, SAS®
Version 8.02, SAS Institute, Inc., Cary, NC), with independent variables being treatment
(heat, cold, starvation, unshocked), medium (TSA-YE, XLD), time (thermal inactivation
treatment) and. their interactions, where time is considered to be a continuous predictor
(or covariate). For each trial, D-values (the time required to reduce the microbial
population by 90%) were calculated for each treatment, using at least seven data points
on TSA-YE and four on XLD, by linear regression of surviving bacteria vs. time using
Microsoft Excel 2000 (Microsoft Corp, Redmond, VA). Assuming linear inactivation
kinetics, the D-value is the negative reciprocal of the slope of the regression line. Results

are reported as averages.

4.4 Results
4.4.1 Composition of Comminuted Turkey

The fat, moisture and protein content of comminuted turkey was 1.50 :l: 0.27%,
70.46 at 0.17% and 23.47 :h 0.61%, respectively. The pH was 5.79 i 0.04. The
background microflora in irradiated comminuted turkey was determined to be present at
<1 CFU/g, with none of the enrichments showing growth.
4.4.2 Sublethal Injury

The three stress treatments produced Salmonella cocktails containing 89.02 :I:

13.18% heat-injured, 44.71 :t 20.07% cold-injured and 67.72 d: 19.54% starvation-injured

92

cells, respectively. After inoculating the comminuted turkey approximately 1 h later,
88.79 :1: 12.39%, 72.19 3: 7.14% and 72.35 :1: 8.14% of these previously stressed cells
were heat-, cold- and starve-injured, respectively. Except for cold shock, the initial
injuries and the injuries after inoculation into comminuted turkey were not significantly
different (P>0.05). All comminuted turkey samples were plated within 6 h, with injury

remaining the same during this time, indicating absence of repair.

4.4.3 Thermal Resistance of Salmonella Cocktails

Analysis of Variance (ANOVA) results are shown in Appendix B (Table B.7).
Analysis of the raw data showed a significant effect (P<0.05) of treatment, medium, time
and the interactive effects of these factors on thermal inactivation. Thermal inactivation
curves for each treatment (unshocked, heat-injured, cold-injured, starvation-injured) are
presented in Figures 4.1, 4.2, 4.3 and 4.4. Thermal inactivation was carried out over a
period of 175 and 91 s for samples plated on TSA-YE and XLD, respectively. TSA-YE
thermal inactivation curves are plotted with all 8 data points sampled (up to 175 3), while
fewer data points were available for XLD curves. By 91 5, there were no detectable (<10
CFU/g) unshocked and heat-shocked survivors on XLD, and no cold-Shocked and starved
survivors by 78 and 52 s, respectively. Overall, for all treatments, XLD yielded
significantly lower recovery (P<0.05) compared to TSA-YE, indicating that the
population of healthy cells alone appears less thermotolerant compared to the total
population of healthy and injured cells.

For all treatments and media, the data, as evidenced by the thermal inactivation

curves, are clearly non-linear. The deviations fiom linearity are most likely because

93

 

 

 

 

 

10
9 ..
8
a 7 ‘ o TSA-YE (healthy +
E 6 . injured)
9' 5 - l XLD (healthy)
3‘ 4 ~
.4
3 a
2 ‘ f
1 a
0 r t r
0 50 100 150 200
Time (seconds)

 

 

 

Fig. 4.1 Thermal inactivation at 60°C of the unshocked Salmonella cocktail. Values
plotted are the arithmetic means of six trials, with standard deviations shown as error
bars.

 

 

  
 

 

 

 

10 ,
9
8
o TSA-YE (healthy +

A 7 injured)
:2; 6 - XLD (healthy)
u.
g 5
3 4
.l

3

2

1

o 1 F T

0 50 100 1 50 200
Time (seconds)

 

 

 

Fig. 4.2 Thermal inactivation at 60°C of the heat-shocked Salmonella cocktail. Values
plotted are the arithmetic means of four trials, with standard deviations shown as error
bars.

94

 

_L
o

 

 
 
  

o TSA-YE (healthy +
injured)

. XLD (healthy)

Log(CFUIg)
o - N or a or or <1 on to

 

 

 

O 50 100 150 200
Time (seconds)

 

 

 

Fig. 4.3 Thermal inactivation at 60°C of the cold-shocked Salmonella cocktail. Values
plotted are the arithmetic means of four trials, with standard deviations shown as error
bars.

 

 

   

 

 

 

10 -
9
8
7 o TSA-YE (healthy +
.3 injured)
3 ° ‘ - XLD (healthy)
IL.
2 5~
8' 4 ~
.1
3 i
2 _
1 - I
0 t t .
0 50 100 150 200
Time (seconds)

 

 

 

Fig. 4.4 Thermal inactivation at 60°C of the starved Salmonella cocktail. Values plotted
are the arithmetic means of four trials, with standard deviations shown as error bars.

95

thermal inactivation curves represent a population of bacterial cells (including different
strains), which have varying degrees of thermal resistance (Whiting 1995, Peleg and Cole
1998). However, as the emphasis was on testing and comparing treatments and there was
a relatively good fit (R2 values >0.95), the log-linear model was used for all analyses.
Dam-values and regression parameters are presented in Table 4.1 for TSA-YE
and Table 4.2 for XLD. The differences, for each treatment, between Dam-values on
non-selective and selective media (injury index) are shown in Table 4.3. Slopes and
Dw-c-values on XLD for the heat-shocked and starved Salmonella cocktails were
significantly different (P<0.05) than for the unshocked control, but not for the cold-
shocked cocktail (P>0.05). 0n TSA-YE, only the heat-shocked cocktail had a
significantly different slope and Dare-value than the unshocked cocktail (see Appendix

B, Table B8).

96

Table 4.1 Dam-values and regression parameters on TSA-YE for unshocked and heat-,
cold- and starvation-stressed Salmonella.

 

 

Treatment y-intercept Slope R2 D-value (min)
Unshocked (n=6) 8.37 i 0.33 -0.04 i 0.002 0.97 0.41 i 0.02
Heat (n=4) 6.67 i 0.26 -0.03 :t 0.003 0.99 0.64 :t 0.08a
Cold (n=4) 8.32 i 0.51 -0.04 :t 0.004 0.96 0.38 :I: 0.03
Starvation (n=4) 7.74 :t 0.17 -0.04 :1: 0.002 0.97 0.41 i 0.02

 

Arithmetic Means :1: Standard Deviations
alSignificantly different (P<0.05) from the unshocked control

Table 4.2 Dwoc-values and regression parameters on XLD for unshocked and heat-, cold-

and starvation-stressed Salmonella.

 

 

Treatment y- intercept Slope RI D-value (min)
Unshocked (n=6) 8.25 :t 0.10 -0.10 :1: 0.01 0.98 0.17 i 0.02
Heat (n=4) 5.25 i 0.20 -0.05 :I: 0.003 0.97 0.35 i 0.02ll
Cold (n=4) 8.06 d: 0.52 -0.10 i 0.01 0.98 0.17 :1: 0.02
Starvation (n=4) 7.04 :1: 0.14 -0.13 i 0.02 0.99 0.14 d: 0.02II

 

Arithmetic Means a: Standard Deviations
aSignificantly different (P<0.05) from the unshocked control

Table 4.3 Injury index‘I following thermal treatment (60°C) of Salmonella.

 

 

Treatment Injury Index
Unshocked 0.24
Heat 0.29
Cold 0.21
Starvation 0.27

 

aInjury index = D-value (non-selective) — D-value (selective)

97

4.5 Discussion

While comparisons in comminuted turkey are limited, the Dam-value (on TSA-
YE) reported here for the unshocked Salmonella cocktail (0.41 d: 0.02 min) is in close
agreement with the Dam-value of 0.46 min for S. Thompson in ground beef reported by
Mackey and Derrick (1987b). Similarly, Maurer (2001) reported a D-Value at 61°C of
0.44 min for an 8-strain Salmonella cocktail containing 7 of our 8 strains (S. Montevideo
FSIS 051 used in place of S. Typhimurium 420) in ground beef, and Smith (2000)
reported Dace-values of 0.43 and 0.41 min for two strains of S. Typhimurium DT104 in
ground beef.

Looking at Dam-values for all treatments, the population of healthy cells alone on
XLD was less thermotolerant compared to the entire population of healthy and injured
cells as enumerated on TSA-YE. 0n XLD, Salmonella was non-detectable (<10 CFU/g)
by 91 s of cooking, whereas 175 s of cooking still resulted in detectable levels on TSA-
YE. As this is true for unstressed as well as stressed Salmonella, cooking appears to
injure a portion of the population, altering its resistance profile. McClements et al.
(2001) reported the same phenomenon in L. monocytogenes using high hydrostatic
pressure, a treatment known to cause sublethal injury (Wuytack et al. 2003), with an 8-
log reduction on Listeria selective agar (Oxford formulation) after 24 min of exposure to
400 MPa compared to a 4-log reduction on TSA-YE. The shifting of cells fi'om the
healthy to injured state is not reflected when using a selective medium or a non-selective
medium alone, as non-detectable cells (on selective media) are often incorrectly assumed

to be dead.

98

Differences between D-values on non-selective and selective media have been
used to assess bacterial injury following thermal stress, with similar D-Values indicating
no sublethal injury. Murano and Pierson (1993), who refer to this relationship as an
Injury Index, found that the injury index for heat-shocked E. coli 0157:H7 was greater
than for non-heat-shocked controls, suggesting that exposure to a heat shock did not
protect the cells against further thermal injury and that pre-heat shock enhanced sublethal
injury. The same could be considered true for heat Shock and starvation in the current
work, as both have a higher injury index than the unshocked control.

The Dam-values for control and stressed salmonellae, on both selective and non-
selective media, should also be examined within the context of the FSIS lethality
performance standards. A 7-log reduction applies to poultry products, with processors
either validating the lethality of their treatment or following pre-determined
time/temperature requirements (safe harbor guidelines). At 60°C, the minimum
processing time required to achieve a 7-log reduction is 12 min (F SIS 1999), which
corresponds to a l-log reduction in 1.71 min. For all treatments tested here, this lethality
is exceeded in comminuted turkey (Dam-values less than 1.71 min).

Although thermal inactivation of Salmonella in comminuted turkey exceeded the
lethality compliance guidelines, safe harbor requirements are based on the processing
time after the minimum temperature is reached. They do not account for long come-up
times or slow-cooking of meat products (e.g. large roasts) that could heat shock
pathogens. Results presented in this study indicate that heat shock enhances the thermal

resistance of Salmonella.

99

Our heat-shocked Salmonella had a higher Dam-value on XLD than did the
unshocked Salmonella, indicating that more time is needed for the healthy fraction of the
heat-shocked population to lose tolerance to selective agents (due to injury by the
cooking process) compared to the primarily healthy unshocked population. As bacterial
cells are assumed to go from the healthy to injured state before dying, this effect of heat
shock may be the first stage in acquired thermotolerance.

Heat shock produced a significantly higher Dance-value on TSA-YE, indicating
that a certain degree of therml resistance was conferred on the Salmonella population.
In experiments with S. Typhimurium, Bunning et al. (1990) characterized the degree of
thermal resistance by the ratio of the D57_3oc-value after heat shock to the Dsmoc-value of
the control. They found that this ratio was 1.1, 1.3 and 3.0 for S. Typhimurium heat-
shocked in TSB-YE for 30 min at 52°C, 42°C and 48°C, respectively, before thermal
inactivation. While investigating the thermotolerance of heat-shocked (48°C for 30 min)
and non-heat-shocked S. Thompson in minced beef; Mackey and Derrick (1987b)
calculated D-Value ratios of 2.38 and 2.74 at 54°C and 60°C, respectively. The Dm-
value ratio in our study was 1.56, which is in the range reported by Banning et al. (1990)
for S. Typhimurium, but somewhat lower than that observed for S. Thompson.
Differences in the degree of protection afforded by heat shock may be related to the
inoculum level and serotype of Salmonella, as well as the heat shock temperature,
medium and heating menstra. Given that Mackey and Derrick (1987b) also heat-shocked
S. Thompson directly in minced beef, the results and implications of our current work
could be expanded if the Salmonella cocktail were heat-shocked in comminuted turkey

rather than in a broth prior to inoculation.

100

Acquired thermotolerance following heat shock has been reported in other
species. Juneja et al. (1998) sublethally heated samples of beef gravy and ground beef
previously inoculated with a four-strain cocktail of E. coli 0157:H7 at 46°C for 15 to 30
min. Heat-shocked cells were more thermotolerant than unshocked cells, with a 1.56-
and 1.50-fold increase in the time to achieve a 4-log reduction at 60°C in beef gravy and
ground beef, respectively. When treated at 62°C, cells of L. monocytogenes heat-shocked
in TSB-YE at 45°C for 180 min were six-fold more heat resistant than unshocked cells
(Pagan et al. 1999), with heat shocking in TSB-YE at 48°C for 10 min increasing the
DssoC-value of L. monocytogenes more than two-fold (Linton et al. 1992).

Stress responses, including general and heat shock specific responses, may
prepare cells for additional thermal stress with the synthesis of heat shock proteins
(HSPS). Many HSPS are chaperones that serve primarily to bind to and stabilize proteins,
monitor protein folding to prevent misfolding or aggregation of proteins and re-fold
improperly folded or denatured proteins (Juneja and Novak 2003). However, the
presence of HSPS does not always correlate with acquired thermotolerance (Watson
1990). Heat shock may also act by enhancing repair (including DNA repair) mechanisms
of cells (Murano and Pierson 1993) and promoting the accumulation of osrnolytes such as
trehalose, which may stabilize proteins, enzymes and membranes (Abee and Wouters
1999, Rowbury 2003).

Starvation-stressed cells also had a higher injury index than did unstressed cells.
However, in contrast to heat shock, starvation did not increase the Dmoc-values on TSA-
YE or XLD. In fact, the Dare-value on XLD was significantly lower for starved

Salmonella than for the control, suggesting that the healthy fraction of the starved

101

population is more readily injured during cooking compared to the primarily healthy
unshocked population. The significance of this observation is not yet clear, because the
Dam-value of the entire population on TSA-YE was unaffected by starvation, indicating
acquired thermotolerance was not a consequence of the given starvation treatment.

L. monocytogenes starved in phosphate buffer for 156 h at 30°C exhibited a 13-
fold increase in the D56oc-value on non-selective media, with starvation apparently
protecting injured cells from the lethal effect of extended heating (Lou and Yousef 1996).
Leenanon and Drake (2001) reported D56oc-values for starved (0.85% NaCl, 48 h, 37°C)
and non-starved E. coli 0157:H7 of 9.7 and 7.1 min, respectively. Differences in the
protective effect of starvation may be due to the length of starvation. Maximum thermal
resistance for L. monocytogenes was achieved between 138 and 156 h of starvation (Lou
and Yousef 1996), while starvation for 4 h provided maximum thermal resistance for E.
coli (Jenkins et al. 1988). In our work, 10 d may have exceeded the starvation period for
maximum thermal resistance in Salmonella, with Smith (2000) reporting that starvation
in peptone water for 14 d (4°C) actually reduced thermal resistance of Salmonella at
55°C, 58°C, 61°C and 63°C in ground beef.

Additionally, our overnight cultures were in stationary phase, the phase that
predominates in natural settings (Rowe and Kirk 2000). The stationary phase is
accompanied by a general state of stress (including heat) resistance (Matin 1991, Givskov
et al. 1994, Hengge-Aronis 1993, Lazazzera 2000), which may be due to the synthesis of
protective proteins that facilitate resistance to chemical and physical challenges (Rowe
and Kirk 2000). These proteins include enzymes for trehalose synthesis and uptake,

H202 degradation, DNA repair and glycogen synthesis (Rowbury 2003).

102

Inability to detect an increase in thermal resistance in our starved cocktail may be
due to a thermal protective effect from the stationary phase control culture. The
significantly increased Dssoc-value reported by Lou and Yousef (1996) for starved L.
monocytogenes was based on a comparison to control cells in the exponential phase.
While Tolker-Nielsen and Molin (1996) observed a significant increase in heat resistance
with starved S. Typhimurium, they also were using exponential phase control cells.
While, some authors have used stationary phase control cells, the results have been
variable. Rowe and Kirk (2000) reported enhanced thermotolerance in two of three
starved strains of E. coli 0157:H7, and Bang and Drake (2002) reported a significant
increase in the Dace-value of only one of three strains of Vibrio vulnificus.

In our work, similar Dam-values were obtained on TSA-YE and XLD for both
the cold-shocked and control cocktails. However, reports of decreased heat resistance
with cold stress can be found in the literature. Leenanon and Drake (2001) cold-stressed
(TSB, 1 week, 5°C) E. coli 0157:H7 and non-pathogenic E. coli and reported D561;-
values of 6.4 and 5.1 min, respectively, compared to 6.8 and 5.5 min for the non-cold-
stressed controls. Additionally, Miller et al. (2000) observed a 25% reduction in the
Dwoc-value for L monocytogenes inoculated onto fi'ankfurter skins following 3 h of cold
shocking at 0°C. A 15% reduction was observed when L. monocytogenes was inoculated
into UHT milk and cold shocked. Storing homogenized whole egg at 4°C for as little as
4 h significantly increased the thermal sensitivity of S. Enteritidis at 60°C (Humphrey
1990), with maximum thermal sensitivity achieved within 24 h. In the case of the current
work, extending the cold shock period beyond 2 h might also result in enhanced thermal

sensitivity.

103

While the mechanism of thermal sensitivity is unknown, alterations in membrane
permeability, reductions in ATP levels and RNA damage have been proposed (Humphrey
1990). Given that no thermal protection was afforded by cold shock in our work and the
apparent thermal sensitivity reported by other authors, cold shock (e.g., refrigeration)
prior to heating may be an effective pathogen intervention strategy (Miller et al. 2000).
Heat sensitive cells of S. Enteritidis rapidly regain heat resistance when incubated at
37°C, with their recovery far less rapid at 20°C (Humphrey 1990). Thermal sensitivity in
L. monocytogenes is sustained even if cells are shifted to 28°C for 3 h before heating
(Miller et al. 2000). Therefore, it seem unlikely that brief exposure to ambient
temperatures between refrigeration and thermal challenge would affect the heat resistance
of sensitized cells.

The cold shock response and its relationship to other stresses are complex. Cold
stress can reportedly enhance freeze and freeze/thaw resistance (Jeffreys et al. 1998,
Bollman et al. 2001, Bang and Drake 2002, Leenanon and Drake 2001), and exposure to
refiigeration temperatures may also maintain the heat shock response and its associated
thermotolerance. Heat-shocked (48°C, 30 min) cells of L. monocytogenes shifted to 4°C
retained their thermotolerance in a sausage mix at least 24 h after initial heat shock
(Farber and Brown 1990). While thermotolerance was lost for heat-shocked (46°C, 15-
30 min) E. coli 0157:H7 after storage at 4°C for 14 h, it was sustained after at least 24 h
of storage at 15°C or 28°C (Juneja et al. 1998). Bunning et al. (1990) reported that the
D57_goC-value for L monocytogenes heat-shocked at 48°C returned to control levels within
1 h when the cells were shifted to 35°C, while the elevated heat resistance reported by

Mackey and Derrick (1986) for heat-shocked (42°C, 45°C or 48°C) S. Typhimurium

104

persisted for at least 10 h of incubation at the heat shock temperatures. Duration of
thermotolerance is likely impacted by the temperatures for growth, heat shock and post-
heat shock storage. The acquisition of thermotolerance associated with heat shock as
observed in our study should be further investigated to identify conditions that allow it to
persist and those that do not, with the latter used to develop pathogen intervention

strategies applicable in thermlly processed foods.

4.6 Conclusion

Only heat shock resulted in elevated thermotolerance of the Salmonella cocktail.
Starvation did not protect against heat, although the physiological state of the control
culture may have affected the comparisons. Because starvation results in cells entering
the stationary phase, thermal resistance of the starved culture did not differ significantly
from the stationary phase control cells. Cold shock also did not impact thermal
resistance, and filrther studies will indicate whether different cold shock treatments can
actually sensitize Salmonella to thermal challenges. Given the outcome with heat shock,
great care should be taken to include the product history and physiological state of
Salmonella when conducting thermal inactivation and challenge studies and when
developing guidelines for thermal processing. Also, non-selective culture media should
be used to assess the response of microorganisms to physical and chemical challenges, as
selective media can underestimate the number of cells remaining in a population. This is
especially important, as the sublethally injured organisms that do not proliferate on

selective media still retain their pathogenicity.

105

CHAPTER 5

CONCLUSIONS AND FUTURE RECOMMENDATIONS

The safety of thermally and otherwise processed foods is based on the detection
of all pathogenic organisms, including those sublethally injured cells tlmt retain their
pathogenicity. Recovery of injured organisms is often highly sensitive to factors that do
not affect, to the same degree, the recovery of uninjured organisms, especially oxidative
stress. Several studies have successfully enhanced recovery of sublethally injured
microorganisms by supplementing various media with reducing agents, 02 scavengers
and/or H202-degrading agents. However, our work showed that TSA-YE with 0.1% or
0.6% sodium pyruvate or Oxyrase® was no more effective than unsupplemented TSA-YE
at recovering heat-, cold- or starvation-stressed Salmonella.

The effects of toxic oxygen species and neutralizing supplements on the recovery
of injured microorganisms are complex. A number of factors can affect the outcome,
including the type of supplement and the concentration at which it is used. As such, the
effectiveness of other concentrations or other reducing agents or oxygen-scavengers
warrants further investigation.

Physical, chemical and nutritional stresses can occur at various stages within the
farm-to-fork continuum, and exposing Salmonella to various stresses in the meat
processing environment can induce sublethal injury and stress response mechanisms that
may alter the microorganism’s thermotolerance during cooking. Although starvation and
cold shock did not protect against heat, thermotolerance of Salmonella was elevated by

heat shock, making it important for the product history and physiological state of

106

Salmonella to be considered when developing and validating thermal processes. In
addition, different inactivation kinetics were observed on non-selective and selective
media. As a result, non-selective culture media should be used to assess the response of
microorganisms to physical and chemical challenges, as selective media can
underestimate the number of cells remaining in a population. This is especially
important, as the sublethally injured organisms that do not proliferate on selective media
still retain their pathogenicity.

The duration of thermotolerance may depend on a number of factors, including
the growth temperature, the heat shock temperature and the post-heat shock storage
temperature. The acquired thermotolerance associated with heat shock in this study
should be investigated in more detail to identify conditions that allow it to persist and
those that do not. The latter could be used to develop pathogen control strategies
applicable to a food processing environment. One of these control strategies may be cold
storage. While cold shock did not affect the thermal resistance of Salmonella in this
study, other authors have indicated that it may increase the thermal sensitivity of bacterial
cells. Therefore, further studies should be undertaken to identify different cold shock
treatments that may sensitize the Salmonella cocktail to a thermal challenge.

The differential plating results indicate that repair of sublethal injury did not occur
during the time between the stress treatments and inoculation into ground turkey or
plating of the uncooked control samples (~ 6 h in ice-water). Additional research on the
tirneframe (and conditions) required for optimal repair of sublethal injury should be

conducted, as repair of injury will correlate with proliferation of bacteria within a food

107

product. The relationship between repair time and the duration of enhanced
thermotolerance would also be of interest.

Finally, as there were significant differences in thermal inactivation observed on
non-selective and selective media, models and D-values developed with a selective
medium should be identified and compared to results obtained using a non-selective

medium.

108

APPENDIX A
DEVELOPMENT OF A NOVEL GRADIENT PLATE METHOD

FOR ASSESSING SUBLETHAL INJURY

A.1 Abstract

Currently, assessment of sublethal injury in a bacterial population following
preservation treatments relies on standard differential plating with non-selective and
selective media, a method that is both time and labor-intensive. This study determined
the effectiveness of two novel gradient plates, poured with either selective BSA or XLD
as the bottom layer and non-selective TSA-YE as the top layer, for determining sublethal
injury. The diffusion of selective components into the non-selective medium affected the
ability of the gradient plates to recover heat-injured Salmonella. Because of this, it was
determined that the gradient plates were inappropriate in the quantitative assessment of
injury. However, the gradient plates did recover a range of cells across the plate and the

technique is a viable one-plate screening method for qualitatively identifying injury.

A.2 Introduction

Heat treatment is the most widely applied method of food preservation (Wuytack
et al. 2003), with many foods dependent upon thermal processing for their stability and
safety. Temperatures and treatments meant to be lethal will not necessarily destroy all
microorganisms, as a potentially significant number of cells actually survive a given heat
process and are merely injured (Murano and Pierson 1993). Injured cells are considered

capable of causing disease if ingested by susceptible individuals (Reissbrodt et a1. 2002).

109

Heat-injured bacteria exhibit increased sensitivity to selective agents such as bile
salts, tellurite, sodium desoxycholate, bismuth sulfite, sodium chloride, crystal violet and
brilliant green (Ray 2001). Because of this characteristic, growth on selective media is
the primary method of assessing sublethal injury (Hurst 1984, Gilbert 1984, Czechowicz
et al. 1996, Semanchek and Golden 1998, Brashears et a1. 2001). The difference in plate
counts between selective and non-selective media is used to quantify sublethal injury as a
proportion or percentage of the entire population (Kang and Siragusa 1999, Semanchek
and Golden 1998, Baylis et al. 2000a, Dickson and Frank 1993, Restaino et al. 1980,
Brashears et al. 2001).

Gradient plates have been used to identify bacterial mutagens (Rexroat et a1.
1995) and antibiotic resistant mutants, with mutants appearing in areas of higher
antibiotic concentration. Two-dimensional gradient plates have been used to characterize
the simultaneous bacterial growth response to two different environmental parameters,
such as pH and NaCl (Caldwell and Hirsch 1973, Wirnpenny and Waters 1984, Thomas
et al. 1992). Use of the gradient plate technique to characterize injured bacterial
populations is limited to one report in which the growth limits for heat-injured
Brochothrix thermosphatica were assessed on two-dimensional pH/NaCl gradient plates
using image analysis (Rattanasomboon et al. 2001).

The present study tested the ability of gradient plates, prepared with non-selective
and selective media, to quantify thermal injury in Salmonella. Since the bacterial
population may be best represented as a continuum of cells ranging ficm mildly to

severely injured, along with potentially unaffected and dead cells following heat

110

treatment (Hurst 1984, Mackey 2000, Stephens et al. 1997), the gradient plates were also

examined for their ability to characterize this range and distribution of injury.

A.3 Materials and Methods
A.3.l Bacterial Strains

The following eight strains of Salmonella were used: S. Thompson FSIS 120
(chicken isolate), S. Enteritidis H3527 (clinical isolate plmge type 13A), S. Enteritidis
H3502 (clinical isolate phage type 4), S. Typhimurium H3380 (DT104 clinical isolate), S.
Hadar MF60404 (turkey isolate), S. Copenhagen 8457 (pork isolate) and S. Heidelberg
F 5038BG1 (clinical isolate), all obtained fi'om V.K Juneja (Agricultural Research
Service, Eastern Regional Research Center, USDA-ARS, Philadelphia, PA) and S.
Typhimurium 420, a DT104 strain isolated from a pork processing facility. All strains
were maintained at —70°C in Trypticase soy broth (TSB) (Difco Laboratories, Detroit,
MI) containing 10% (V/v) glycerol (J .T. Baker, Mallinckbrodt Baker, Inc., Phillipsburg,
NJ).
A.3.2 Culture Preparation

Each culture was propagated by transferring one loopful of the fi‘ozen cell
suspension into 9 mL of TSB containing 0.6% yeast extract (TSB-YE) (Difco). Cultures
were subjected to a minimum of two consecutive overnight transfers (37°C, 18-24 h) in
TSB-YE before use.
A.3.3 Heat Shock Treatment

An 8-strain Salmonella cocktail was prepared by combining 3 mL of the

individual overnight cultures in a sterile 50 mL centrifuge tube (Clear Propylene, Plug

111

Seal Cap, Corning Inc., Corning, NY), centrifuging at 10,000 rpm for 15 min at 4°C
(Super T21, Sorvall® Products, L.P., Newtown, CT) and re-suspending the pellet in 3 rnL
of 0.1% peptone (Difco) to obtain a culture containing approximately 10lo CFU/mL.
Heat-injured cells were obtained using the method of Mathew and Ryser (2002) and
Busch and Donnelly (1992), with minor modifications. In this procedure, 200 mL of
TSB-YE (in a 2800 mL wide-mouth Fembach flask) was tempered to 54°C in a shaking
(30 rpm) water bath (Reciprocal Shaking Bath, Precision Scientific, Winchester, VA),
inoculated with 2 mL (~10lo CFU/mL) of the Salmonella cocktail and heated at 54°C for
30 min. After 30 min, heat-shocked cells were immediately cooled in an ice-water bath
for 3 to 5 min.

The initial temperature for heat Shock was 48°C, which was also used by Mackey
and Derrick (1986, 1987a, 1987b) and Bunning et al. (1990) for Salmonella. However,
for the individual Salmonella strains in the cocktail, the extent of injury observed at 48°C
(6-54%), as well as at 52°C (23-68%), was decided to be insufficient for the current study
(see Appendix B, Table 8.1). At 56°C, a 30 min heat shock was lethal for 4 of the 8
strains. As 54°C resulted in approximately 90% injury in preliminary studies, it was this
temperature that was incorporated into the heat shock procedure.

A.3.4 Gradient Plate Production

Gradient plates were prepared in 100 x 100 x 15-mm square petri dishes (Integrid,
BDFalcon, Franklin Lakes, NJ) divided into an equal 6 x 6 grid (Figure AD, with
numerical columns and alphabetical rows. The following media were used: Trypticase
soy agar (Difco) containing 0.6% yeast extract (TSA-YE), bismuth sulfite agar (Difco)

(BSA) and xylose lysine deoxycholate (XLD) agar (Difco). For the bottom layer, 25 mL

112

of either XLD or BSA were pipetted into the petri dishes. The plates were elevated to a
height of 0.6 cm (creating an angle with the bench of ~ 4°) (Figure A.2) and the agar
allowed to harden (~15 min). After solidification, the plates were placed flat and 25 rnL
of non-selective TSA-YE were pipetted over the selective medium (Figure A.3). The

plates were allowed to harden, stored overnight and then dried for 30 min before use.

113

 

Fig. A.1 Integrid 100 x 100 x 15 mm square plates with 6 x 6 grid.

 

Fig. A.2 Preparation of the bottom (selective) layer of gradient plates.
* Angle measurement: ~ 4°

 

Fig. A.3 Completed gradient plate, with a selective (bottom) and non-selective (top)
layer.

114

A.3.5 Recovery of Cells

Before and after heat stressing, the cells were enumerated by serial dilution in
0.1% peptone followed by spiral plating (Autoplate® 4000, Spiral Biotech, Inc.,
Norwood, MA) in duplicate on TSA-YE, XLD and BSA in standard 100 x 15-mm round
petri dishes. Plates were counted after 18 to 24 h of incubation at 37°C. Counts on the
non-selective medium (TSA-YE) represent the total viable population, while those on the
selective media represent the uninjured fi'action. Percent injury following heat treatment
was determined according to the formula: % injury = [(count on non-selective agar —
count on selective agar)/count on non-selective agar] x 100. Percent injury was
calculated for each replication, with the results reported as averages.

Cells were also recovered before and after heat stressing by spread plating in
duplicate on XLD/ISA-YE and BSA/TSA-YE gradient plates. Plates were counted after
18 to 24 h of incubation at 37°C. Populations recovered on the entire plate and in the six
vertical columns representing the six different gradations of selectivity were calculated,
the latter according to the formula: CFU/mL = [(# of colonies in column)(dilution
factor)] + 0.022. The denominator is the assumed volume deposited within each column,
or 1/6 of the entire volume (0.13 mL) spread on the gradient plate. With selectivity
decreasing across the plate, column 1 is assumed to be equal in selectivity to the selective
medium and column 6 to the non-selective medium. Percent injury on the gradient plates
was calculated as follows: % injury = [(count in column 6 — count in column 1)/count in
column 6] x 100. Percent injury was calculated for each replication, with the results

reported as averages.

115

A.3.6 Data Analysis

All experiments were repeated six times, with the results reported as averages.
An unpaired Student’s t-test (Proc TIEST, SAS® Version 8.02, SAS Institute, Inc., Cary,
NC) was used to compare column 1 and column 6 data to the selective and non-selective
media, respectively. A one-way Analysis of Variance (AN OVA) (Proc GLM, SAS®

Version 8.02, SAS Institute, Inc., Cary, NC) was used for all other analyses.

A.4 Results
A.4.1 Sublethal Injury

Based on traditional differential plating on TSA-YE and XLD or BSA, 93.48 i
6.61% and 76.72 at: 27.47% of the original Salmonella cocktail was sublethally injured,
respectively. When determined by gradient plating, injury on XLD/TSA-YE and
BSA/TSA-YE gradient plates was 65.62 i 15.29% and 46.53 :1: 16.95%, respectively.
The two XLD injuries are significantly (P<0.05) different, while the two BSA injuries are
not (P=0.045). The BSA injury results were affected by a potential outlier from standard
differential plating (only 24.70% injury as compared to 94.50%, 98.10%, 93.20%,
77.80% and 72% for the other trials), which is likely the cause of the large standard
deviation (27.47%). When this outlier is excluded, the BSA injuries also become
significantly different.
A.4.2 XLD/TSA-YE Gradient Plates

Before heat shock, no significant (P>0.05) differences were seen between the
numbers of Salmonella recovered on TSA-YE and column 6 (non-selective) of the

gradient plates or between XLD and column 1 (selective) of the gradient plates (Table

116

A.1). After heat shock, standard XLD plates and column 1 did not differ, while standard
TSA-YE plates and column 6 were significantly different (P<0.05). The averages for
each individual column of the gradient plates are presented in Table A.2. While initial
counts did not differ fiom column 1 to column 6, post-heat shock counts gradually
increased across the plate (Figure A.4).
A.4.3 BSA/TSA-YE Gradient Plates

Both before and after heat shock, the numbers of Salmonella recovered on TSA-
YE and column 6 were not significantly different (P>0.05), as was also true of BSA and
column 1 (Table A.3). As with the analysis of BSA injury, if the potential outlier is
removed, TSA-YE and column 6 counts after heat shock are significantly different
(P<0.05) (Table A.4). The data for each individual column of the gradient plates are
found in Table A5. Initial Salmonella levels in each column were the same, while those

following heat shock were not (Figure A.5).

Table A.l Growth of Salmonella on standard TSA-YE and XLD plates and in selected
columns of XLD/TSA-YE gradient plates.

 

 

Initial Post-Heat Shock
TSA-YE (NS) 10.74 a 0.41 9.51 r 0.221‘
XLD (S) 10.71 a 0.40 8.14 s: 0.35
Column 1 (S) 10.25 a 0.40 8.23 a 0.48
Column 6 (NS) 10.38 a 0.36 8.74 a 0.31b

 

NS-non-selective, S-selective
Arithmetic Means [Log (CFU/mL)] 1: Standard Deviations, n=6
Means for the same treatment and medium selectivity with different superscripts are

Significantly (P<0.05) different.

117

Table A.2 Recovery of Salmonella on XLD/TSA-YE gradient plates

 

 

Initial Post-Heat Shock
Column 1 (S) 10.25 i 0.40 8.23 i 0.48
Column 2 10.09 A: 0.39 8.22 :h 0.42
Column 3 10.13 :t 0.37 8.23 :t 0.48
Column 4 10.08 :1: 0.41 8.40 :r 0.46
Column 5 10.22 i: 0.40 8.53 i 0.34
Column 6 (NS) 10.38 i: 0.36 8.74 :1: 0.31

 

NS-non-selective, S-selective

Arithmetic Means [Log (CFU/mL)] i Standard Deviations, n=6

Table A.3 Growth of Salmonella on standard TSA-YE and BSA plates and in selected

columns of BSA/TSA-YE gradient plates.

 

 

Initial Post-Heat Shock
TSA-YE (NS) 10.74 :1: 0.41 9.51 i 0.22
BSA (S) 10.74 :1: 0.41 8.60 i 0.44
Column 1 (S) 10.53 :t 0.26 9.02 i 0.17
Column 6 (NS) 10.50 i 0.35 9.29 d: 0.07

 

NS-non-selective, S-selective

Arithmetic Means [Log (CFU/mL)] :t Standard Deviations, n=6
Means for the same treatment and medium selectivity with different superscripts are

significantly (P<0.05) different.

118

Table A.4 Grth of Salmonella on standard TSA-YE and BSA plates and in selected
columns of BSA/TSA-YE gradient plates.

 

 

Initial Post-Heat Shock
TSA-YE (NS) 10.78 :1: 0.44 9.60 i: 0.06“
BSA (S) 10.78 :1: 0.44 8.53 a 0.46
Column 1 (S) 10.55 a: 0.29 9.02 a 0.19
Column 6 (NS) 10.53 a 0.38 9.31 a: 0.06”

 

NS-non-selective, S-selective

Arithmetic Means [Log (CFU/mL)] 3: Standard Deviations, n=5

Means for the same treatment and medium selectivity with different superscripts are
significantly (P<0.05) different.

Table A.5 Recovery of Salmonella on BSA/TSA-YE gradient plates

 

 

Initial Post-Heat Shock
Column 1 (S) 10.53 i 0.26 9.02 i 0.17
Column 2 10.45 :t 0.33 8.89 i 0.22
Column 3 10.17 :1: 0.48 8.88 a: 0.21
Column 4 10.18 i 0.38 8.93 :1: 0.22
Column 5 10.42 :t: 0.37 9.10 d: 0.20
Column 6 (NS) 10.50 i: 0.35 9.29 i 0.07

 

NS-non-selective, S-selective
Arithmetic Means [Log (CFU/mL)] i Standard Deviations, n=6

119

 

 

 

 

 

 

3 8.7 1
E
D
u.
9, 8.2 ~
O)
O
_|
7.7 -
. . ’ y= 8.03+0.10x
‘ R2: 0.17
7.2 . . , . .
0 1 2 3 4 5 6
Column

 

Fig. A.4 Recovery on XLD/TSA-YE gradient plates following heat shock.

 

5°
or

 

5°
.5

5"
to

Log (CFUImL)
‘f’
00 09 O
O
0 0
one
0

 

 

 

8.8 - °
0 O .

8.6 -
O O

84, 0 y=8.81+0.06x

R2 = 0.20
8.2 i . . - .
0 1 2 3 4 5 6

Column

 

Fig. A.5 Recovery on BSA/TSA-YE gradient plates following heat shock.

120

 

 

 

 

A.5 Discussion

Analysis of the XLD/TSA-YE gradient plates was straightforward, while analysis
of the BSA/TSA-YE gradient plates considers the exclusion of the potential outlier
associated with standard BSA plates. The presence of this outlier, and the overall
variability associated with BSA injury, is likely due to the medium itself. Additional
measurements of injury on BSA and XLD following heat shock were compared (see
Appendix B, Table B.9). From this analysis, it is apparent that the range of injuries (and
standard deviation) associated with BSA is much larger than for XLD and the average
injury is lower.

Moriffigo et al. (1989) also observed greater injury with XLD following seawater
stress, reporting a 50-60% reduction in Salmonella levels on XLD and only a 25%
reduction on BSA. Of the different media (BSA, XLD, brilliant green agar, brilliant
green-phenol red-lactose sucrose-agar, eosin-methylene blue agar, Hektoen enteric agar,
Salmonella-Shigella agar, tryptic soy-brilliant green agar and tryptic soy-xylose-lysine
agar) tested, Morifiigo et al. (1989) found that XLD and brilliant green-phenol red-
lactose-sucrose agars were the most selective for Salmonella, with the selective agents in
XLD perhaps being more inhibitory to heat-injured Salmonella than those in BSA. In
addition, problems with preparation, storage and aging of BSA before use can affect
selectivity and overall performance (Warburton et al. 1994).

Recovery of fewer salmonellae after heat treatment in column 6 (non-selective) of
the XLD/1‘ SA-YE and BSA/TSA-YE gradient plates (n=5) than on standard TSA-YE
plates suggests a problem with difiusion of the selective media. Diffilsion of the

selective agents in XLD and BSA apparently inhibited heat-injured Salmonella that were

121

able to recover on TSA-YE plates. The selective components would not be expected to
affect healthy cells, which explains why diffusion was not a problem for initial plate
counts.

The gradient plates could be improved by replacing XLD and BSA with a
selective medium that would allow measurement of the gradient, with the relationship of
concentration as a function of distance along the plate. However, even plates with a pre-
determined gradient are affected by diffusion and must be used in a timely manner. A
major limitation of the gradient plate technique is that experiments must be performed
over a relatively short period of time due to the diffusive breakdown of the gradients
(Rattanasomboon et al. 2001, McClure et at 1989).

Because of problems with diffusion and the inability to measure a true selective
gradient, the gradient plates developed here cannot replace standard differential plating
for the quantitative assessment of bacterial injury. However, these plates Show potential
for qualitative assessment of sublethal injury. According to the statistical analysis,
initially all six columns across the BSA/TSA-YE and XLD/'1‘ SA-YE gradient plates did
not differ significantly. However, following the heat treatment, the six columns on each
plate were not the same. The slope of the regression line for both types of gradient plates
was >1, indicating a general trend of increasing salmonellae numbers across the plates
(Figures A.4 and A5).

The numerical value of the slope may prove useful as an estimation of the degree
of injury sustained by a population. However, the graphical representation, with a slope
>1, of bacteria across the gradient plate can be used to Visualize the distribution of cells,

with a larger number of injured cells able to grow as one proceeds from column 1 to

122

column 6, with the additional bacteria recovered in each column hypothetically more
injured than those in the previous columns. If the gradient plates were prepared with a
measurable gradient (e.g., NaCl) and used before the diffusive breakdown of that
gradient, the ability to proliferate in the presence of a certain concentration (or column)
could be used to classify and/or quantify injury. Subtracting the total number of colonies
in previous columns would give the number of injured cells in a given column.

Perhaps the most promising aspect of the gradient plates is the possibility for the
development of a screening assay for sublethal injury. The six columns yielded similar
results before heat shock but not after, generally exhibiting a gradual increase instead.
Although this method requires some refinement, the observation of a slope >1 can be
used as a simple indicator of injury. If injury is identified, further analyses (e.g., standard
differential plating, protein analysis, thermal inactivation studies to assess cross-
protection) can be conducted to quantify and characterize this injury. Such gradient
plates may be especially useful for determining the potential for non-thermal and non-
traditional (e.g., pulsed electric field) processing methods to cause sublethal injury. For
example, using differential plating with TSA and three selective media (TSA with 3%
NaCl, TSA adjusted to pH 5.5 and violet red bile glucose agar), Wuytack et al. (2003)
found that high-pressure homogenization and pulsed electric field treatments caused little
or no injury. Confirmation of such results using the gradient plate technique would

conserve both time and resources.

123

A.6 Conclusion

While the current XLD/TSA-YE and BSA/TSA-YE gradient plates cannot yet
replace standard differential plating for quantitative assessment of bacterial injury, these
plates do show potential as a screening method to identify sublethal injury in a bacterial
population. In addition, if advancements are made to the technique, including the
preparation of plates with a measurable gradient, the method could be used to quantify or
classify injury. Sublethal injury has many consequences, such as cross-protection against
other, more severe stresses and the potential for increased virulence of injured cells. The
ability of a single-plate method to identify injured populations could become a crucial
first step in characterizing the bacterial response to microbial reduction strategies and
preservation treatments. Based on the qualitative identification of injury, future efforts
should address environmental conditions that protect injured cells and processing

parameters that control injury and lethality.

124

APPENDIX B

SUPPLEMENTARY DATA AND STATISTICS

Table B.l Preliminary heat shock data comparing percent injury (on BSA) at different

 

 

temperatures.

Salmonella Strain 48°C (n=2)a 52°C (n=2)“ 54°C (n=3)ab 56°C (n=1)
S. Typhimurium 420 61.20 91.93 NDc
S. Thompson FSIS 120 23.20 67.60 91.13 59.60
S. Enteritidis H3527 6.13 66.45 90.80 98.40
S. Enteritidis H3502 42.05 58.25 93.53 88.40
S. Typhimurium H3380 54.05 56.22 91.43 100
S. Hadar MF60404 35.82 35.90 88.67 90.90
S. Copenhagen 8457 42.10 49.30 92.47 100
S. Montevideo F SIS 051 100 98.70 NDc 100
S. Heidelberg F5038BG1 16.05 22.60 91.40 100

 

aArithmetic mean

bFor at least three trials, S. Montevideo F SIS 051 was unable to grow on BSA, even if
not heat-treated (i.e., plated after being grown overnight at 37°C). As a result, this strain
was replaced with S. Typhimurium 420, a DT104 environmental isolate.

ND-not determined

Table B.2 Preliminary cold Shock (4°C) data comparing percent injury (on BSA) at
different time intervals.

 

 

Salmonella Strain 30 minutes 2 hours 4 hours

S. Typhimurium 420 29.23 i 12.81 75.90 a: 2.72 30.97 :E 18.00
S. Thompson FSIS 120 61.73 i 19.90 85.10 d: 12.18 43.53 1 22.70
S. Enteritidis H3527 45.10 :t 32.57 76.90 3: 4.17 38.90 :E 20.14
S. Enteritidis H3502 55.90 i 19.32 74.87 i 8.28 38.23 :1: 16.77
S. Typhimurium H3380 54.63 :t 16.72 81.83 i 3.43 18.93 :1: 19.36
S. Hadar MF60404 52.40 i 20.65 73.20 i 6.51 19.23 i 13.40
S. Copenhagen 8457 53.00 i 18.76 79.33 i 1.16 16.23 i 4.09

S. Heidelberg F503SBG1 44.60 3: 11.13 79.47 i 1.86 58.17 i 2.60

Percentages are arithmetic mean i standard deviation for n=3.

125

Table B.3 Preliminary starvation data comparing percent injury (on BSA) at different

time intervals.

 

 

Salmonella Strain 5 days (n=1) 10 days (n=6)ab 14 days (n=1)
S. Typhimurium 420 NDc 84.15 NDc

S. Thompson FSIS 120 0 84.85 74.90

S. Enteritidis H3527 0 83.77 55.50

S. Enteritidis H3502 18.06 79.58 74.70

S. Typhimurium H3380 0 79.31 27.10

S. Hadar MF60404 0 79.92 74.40

S. Copenhagen 8457 0 84.98 40.20

S. Montevideo FSIS 051 0 NDc 39.30

S. Heidelberg F5038BG1 0 86.23 40.10

 

aArithmetic mean

l’For at least three trials, S. Montevideo FSIS 051 was unable to grow on BSA, even if
not heat-treated (i.e., plated after being grown overnight at 37°C). As a result, this strain

was replaced with S. Typhimurium 420, a DT104 environmental isolate.

ND-not determined

Table BA Injury (on BSA) for 8 individual Salmonella strains following cold storage in

TSB-YE at 4°C for 10 d.

 

 

Salmonella Strain % Injury

S. Typhimurium 420 0 i 0

S. Thompson FSIS 120 0 :l: 0

S. Enteritidis H3527 0 i 0

S. Enteritidis H3502 3.23 i 3.30
S. Typhimurium H3380 2.40 :t 4.16
S. Hadar MF60404 0.33 d: 0.58
S. Copenhagen 8457 2.10 :1: 2.05
S. Heidelberg F5038BG1 3.00 i 5.20

 

Percentages are arithmetic mean :1: standard deviation for n=3.

126

Table B.5 Analysis of variance (AN OVA) table for injury of 8 individual Salmonella
strains following heat shock (54°C/30 min), cold shock (4°C/2 h) or starvation (4°C/10 d)
(Proc Mixed, SAS® Version 8.02, SAS Institute, Inc., Cary, NC).

 

 

Effect Num DF Den DF F value Pr>F‘l
Heat Shock (n=3)

Strain 44 1.18 0.3348
Treatment (Heat) 1 44 8108.69 <.0001
Strain*Treatment 8 44 0.63 0.7456
Cold Shock (n=3)

Strain 8 42 0.34 0.9439
Treatment (Cold) 1 42 1938.59 <.0001
Strain*Treatment 8 42 0.63 0.7452
Starvation (n=3)

Strain 8 42 1.64 0.1438
Treatment (Starvation) 1 42 5093.37 <.0001
Strain*Treatment 8 42 0.79 0.6171

 

aPr>F values less than 0.05 indicates significance of the given effect.

Table B.6 ANOVA table for recovery of Salmonella following heat Shock, cold shock or
starvation (Proc Mixed, SAS® Version 8.02, SAS Institute, Inc., Cary, NC).

 

 

Effect Num DF Den DF F value Pr>F°
Heat Shock

Medium 3 18 0.65 0.5945
Treatment 1 8 134.20 <.0001
Medium" Treatment 3 1 8 2.47 0.0947
Cold Shock

Medium 3 12 0.82 0.5099
Treatment 1 5 33.02 0.0022
Medium*Treatment 3 12 0.52 0.6781
Starvation

Medium 3 12 0.99 0.4319
Treatment 1 5 0.01 0.9094
Medium*Treatment 3 12 0.74 0.5458

 

aPr>F values less than 0.05 indicates significance of the given effect.

127

Table B.7 AN OVA table for the effects of treatment, medium, time and their interactions

on the number of survivors following thermal (60°C) treatment of Salmonella (Proc

GLM, SAS® Version 8.02, SAS Institute, Inc., Cary, NC).

 

 

Source DF Sum of Mean F value Pr>Fa
Squares Square

Treatment 3 68.01 22.67 72.06 <0.0001
Medium 1 8.28 8.28 26.31 <0.0001
Treatment*Medium 3 4.92 1.64 5.22 0.0017
Time 1 682.39 682.39 2169.07 <0.0001
Time*Medium 1 1 12.53 112.53 357.69 <0.0001
Time*Treatment 3 60.24 20.08 63.83 <0.0001
Time*Trt*Medium 3 20.68 6.89 21.91 <0.0001

 

aPr>F values less than 0.05 indicates significance of the given effect.

Table B.8 Statistical analysis (Proc GLM, SAS® Version 8.02, SAS Institute, Inc., Cary,
NC) of the thermal inactivation of Salmonella, sorted by medium (Proc Sort, SAS®

Version 8.02), with unshocked as the control on each medium.

 

 

 

Parameter Estimate Standard t value Pr>|t|a
Error

TSA-YE

Time -0.04083 0.00132 -30.98 <0.0001

Time*Treatment Unshocked 0.00000

Tirne*Treatment Heat 0.01406 0.00214 6.57 <0.0001

Time*Treatment Cold -0.003 12 0.00214 -1 .46 0. 1473

Time*Treatment Starvation 0.0001 7 0.00208 0.08 0.9362

XLD

Time -0.09873 0.00353 -27.94 <0.0001

Tilne*Treatment Unshocked 0.00000

Time*Treatment Heat 0.051 19 0.00509 10.06 <0.0001

Time*Treatment Cold 0.00227 0.00606 0.38 0.7083

Tirne*Treatment Starvation -0.02294 0.00870 -2.64 0.0099

 

aPr>|t| values less than 0.05 indicates significance of the given effect.

128

Table B.9 Percent injury on three different selective media following heat shock.

 

 

XLDa BSAa BGa
Observationsb 19 35 9
Average 92.12 79.20 70.57
Minimum 70.80 24.70 7.10
Maximum 98.90 98.10 81.30
Standard Deviation 6.97 18.08 23.90

 

XLD- xylose lysine deoxycholate, BSA-bismuth sulfite agar, BG-brilliant green
aMedia ingredients. Per liter of water: XLD (3 g yeast extract, 3.75 g xylose, 7.5 g
lactose, 7.5 g sucrose, 5 g L-Lysine, 5 g NaCl, 6.8 g sodium thiosulfate, 0.8 g ferric
ammonium citrate, 15 g agar, 0.08 g phenol red), BSA (5 g beef extract, 10 g peptone, 5 g
dextrose, 4 g disodium phosphate, 0.3 g ferrous sulfate, 2 g bismuth ammonium citrate, 6
g sodium sulfite, 0.025 g brilliant green, 20 g agar), BG (10 g proteose peptone, 3 g yeast
extract, 10 g lactose, 10 g saccharose, 5 g NaCl, 20 g agar, 0.0125 g brilliant green, 0.08
g phenol red).

Observations were collected from several different projects, including work involving
gradient plates, media supplementation and thermal inactivation in ground turkey.

Table B.l0 Percent injury on three different selective media following cold shock.

 

 

XLD BSA BG
Observationsa 14 26 9
Average 59.67 60.07 67.04
Minimum 20.10 0 56.20
Maximum 90.70 92.50 89.30
Standard Deviation 20.49 22.59 10.83

 

XLD- xylose lysine deoxycholate, BSA-bismuth sulfite agar, BG-brilliant green
aObservations were collected fi'om several different projects, including work involving
gradient plates, media supplementation and thermal inactivation in ground turkey.

129

Table B.ll Percent injury on three different selective media following starvation

 

 

XLD BSA BG
Observationsa 8 20 6
Average 65.76 61.10 78.12
Minimum 3 1.60 0 64.80
Maximum 80.2 94.30 91.60
Standard Deviation 19.25 33.14 1 1.25

 

 

XLD- xylose lysine deoxycholate, BSA-bismuth sulfite agar, BG-brilliant green
aObservations were collected fi'om several different projects, including work involving
media supplementation and thernml inactivation in ground turkey.

130

REFERENCES

Abee, T. and Wouters, J.A. 1999. Microbial stress response in minimal processing. Int. J.
Food Microbiol. 50: 65-91

Abshire, KZ. and Neidhardt, RC. 1993. Analysis of proteins synthesized by Salmonella
typhimurium during growth within a host macrophage. J. Bacteriol. 175: 3734-3743

Adler, H.I. 1990. The use of microbial membranes to achieve anaerobiosis. Crit. Rev.
Biotechnol. 10: 119-127

Adler, H.I., Carrasco, A., Crow, W. and Gill, J .S. 1981. Cytoplasmic membrane-fraction
that promotes septation in an Escherichia coli lon mutant. J. Bacteriol. 147: 326-332

Ahmed, M.N., Conner, DE. and Huffman, D.L. 1995. Heat-resistance of Escherichia coli
0157:H7 in meat and poultry as affected by product composition. J. Food Sci. 60: 606-
610

Allen, LA, Pasley, SM. and Pierce, M.S.F. 1952. Conditions affecting the growth of
Bacterium coli on bile salts media. Enumeration of this organism in polluted waters. J.
Gen. Microbiol. 7: 257-267

AMI. 2000. The role of meat and poultry industry in the US. economy. American Meat
Institute. http://www.meatami.org/indupg02.htm

Amin, V. and Olson, N. 1968. Influence of catalase activity on resistance of a coagulase-
positive staphylococci to hydrogen peroxide. Appl. Microbiol. 16: 267-270

Andrews, GP. and Martin, SE. 1979. Catalase activity during recovery of heat-stressed
Staphylococcus aureus MF-3 1. Appl. Environ. Microbiol. 38: 390-394

Andrews, W.H. 1986. Resuscitation of injured Salmonella spp. and coliforms from foods.
J. Food Prot. 49: 62-75

AOAC. 1996. oaicial Methods of Analysis, 16th ed. Association of Official Analytical
Chemists, Gaithersburg, MD

Archer, D.L. 1996. Preservation microbiology and safety: evidence that stress enhances
virulence and triggers adaptive mutations. Trends Food Sci. Technol. 7: 91-95

Archer, J ., Jervis, E.T., Bird, J. and Gaze, J.E. 1998. Heat resistance of Salmonella
weltevreden in low-moisture enviromnents. J. Food Prot. 61: 969-973

131

Audia, J.P., Webb, CC. and Foster, J.W. 2001. Breaking through the acid barrier: an
orchestrated response to proton stress by enteric bacteria. Int. J. Med. Microbiol 291: 97-
106

Baird-Parker, AC. and Davenport, E. 1965. The effect of recovery medium on isolation
of Staphylococcus aureus after heat treatment and after the storage of frozen or dried
cells. J. Appl. Bacteriol. 28: 390-402

Bang, W. and Drake, MA. 2002. Resistnace of cold- and starvation-stressed Vibrio
vulnificus to heat and fieeze-thaw exposure. J. Food Prot. 65: 975-980

Bang, I.S., Audia, J .P., Park, Y.K. and Foster, J.W. 2002. Autoinduction of the ompR
response regulator by acid shock and control of the Salmonella enterica acid tolerance
response. Mol. Microbiol. 44: 1235-1250

Bayles, D.O., Annous, BA. and Wilkinson, B.J. 1996. Cold stress proteins induced in
Listeria monocytogenes in response to temperature downshock and growth at low
temperatures. Appl. Environ. Microbiol. 62: 1116-1119

Baylis, C.L., MacPhee, S. and Betts, RP. 2000a. Comparison of two commercial
preparations of buffered pepton water for the recovery and growth of Salmonella bacteria
from foods. J. Appl. Microbiol. 89: 501-510

Baylis, C.L., MacPhee, S. and Betts, RP. 2000b. Comparison of methods for the
recovery and detection of low levels of injured Salmonella in ice cream and milk powder.
Lett. Appl. Microbiol. 30: 320-324

Bayne, H.G., Garibaldi, J .A. and Lineweaver, H. 1965. Heat resistance of Salmonella
Typhimurium and Salmonella Seftenberg 775W in chicken meat. Poultry Sci. 44: 1281-
1284

Bearson, S.M., Benjamin, W.H., Jr., Swords, W.E. and Foster, J.W. 1996. Acid shock
induction of RpoS is mediated by the mouse virulence gene mviA of Salmonella
typhimurium. J. Bacteriol. 178: 2572-2579

Bell, C. and Kyriakides, A. 2002. Salmonella. A Practical Approach to the Organism and
its Control in Food. Blackwell Science, Oxford

Berlutti, F., Casalino, M., Zagaglia, C., Fradiani, P.A., Visca, P. and Nicoletti, M. 1998.
Expression of the virulence plasmid-carried apyrase gene (apy) of enteroinvasive
Escherichia coli and Shigellaflexneri is under the control of H-NS and the VirF and VirB
regulatory cascade. Infect. Immun. 66: 4957-4964

Beuchat, LR. 1978. Injury and repair of Gram-negative bacteria, with special
consideration of the involvement of the cytoplasmic membrane. Adv. Appl. Microbiol.
23: 219-243

132

Blackburn, CW. and McCarthy, J .D. 2000. Modifications to methods for enumeration
and detection of injured Escherichia coli 0157:H7 in foods. Int. J. Food Microbiol. 55:
285-290

Bollman, J., Ismond, A. and Blank, G. 2001. Survival of Escherichia coli 0157 :H7 in
frozen foods: impact of the cold shock response. Int. J. Food Microbiol. 64: 127-138

Boziaris, LS. and Adams, MR. 2001. Temperature shock, injury and transient sensitivity
to nisin in Gram negatives. J. Appl. Microbiol. 91: 715-724

Brackett, RE, Schuman, J.D., Ball, HR. and Scouten, A.J. 2001. Thermal inactivation
kinetics of Salmonella spp. within intact eggs heated using humidity-controlled air. J.
Food Prot. 64: 934-938

Brashears, M.M., Amezquita, A. and Stratton, J. 2001. Validation of methods used to
recover Escherichia coli 0157 :H7 and Salmonella spp. subjected to stress conditions. J.
Food Prot. 64: 1466-1471

Bremer, E. and Kramer, R. 2000. Coping with osmotic challenges: osmoregulation
through accumulation and release of compatible solutes. In Storz and Hengge-Aronis
(ed.), Bacterial Stress Responses. ASM Press, Washington D. C., p. 79-97.

Brenner, F .W., Villar, R.G., Angulo, F .J., Tauxe, R., and Swaminathan, B. 2000.
Salmonella Nomenclature. J. Clin. Microbiol. 38: 2465-2467

Bromberg, R., George, SM. and Peck, M.W. 1998. Oxygen sensitivity of heated cells of
Escherichia coli 0157:H7. J. Appl. Microbiol. 85: 23 1-237

Buchmeier, NA. and Hefli'on, F. 1990. Induction of Salmonella stress proteins upon
infection of macrophages. Science. 248: 730-732

Bunning, V.K., Crawford, R.G., Tierney, J.T. and Peeler, J .T. 1990. Thermotolerance of
Listeria monocytogenes and Salmonella typhimurium after sublethal heat shock. Appl.
Environ. Microbiol. 56: 3216-3219

Busch, S.V. and Donnelly, CW. 1992. Development of a repair-enrichment broth for

resuscitation of heat-injured Listeria monocytogenes and Listeria innocua. Appl. Environ.
Microbiol. 58: 14-20

Busta, F .F. 1976. Practical implications of injured microorganisms in food. J. Milk Food
Technol. 39: 138-145

Busta, RF. 1994. Importance of injured food-bome microorganisms in minimal
processing. In Singh and Oliveira (ed.), Minimal Processing of Foods and Process
Optimization: An Interface. CRC Press, Boca Raton, p. 227-231

133

Buzby, J.C., Frenzen, P.D., Rasco, B. 2001. Product Liability and Microbial Foodbome
Illness. ERS Agricultural Economic Report No. 799, April 2001

Calabrese, JP. and Bissonnette, GK. 1990. Improved membrane filtration method
incorporating catalase and sodium pyruvate for detection of chlorine-stressed coliform
bacteria. Appl. Environ. Microbiol. 56: 3558-3564

Caldwell, DE. and Hirsch, P. 1973. Growth of microorganisms in two-dimensional
steady-state diffusion gradients. Can. J. Microbiol. 19: 53-58

Carlson, TR. 2002. Effect of water activity and humidity on the thermal inactivation of
Salmonella during heating of meat. MS. Thesis. Michigan State University, East
Lansing, Nfl

Carmel, P. 2000. The Meat and Poultry Industry. October 2000 report for B28 Electronic
Commerce lecture. Management of Global Information Technology, Kogod School of
Business, American University, Washington DC.
http://www.American.edu/carmel/ma590la/industry.htm

Casadei, I.R., Hitchings, E., Archer, J. and Gaze, J.E. 2001. Heat resistance of Bacillus
cereus, Salmonella typhimurium and Lactobacillus delbrueckii in relation to pH and
ethanol. Int. J. Food Microbiol. 63: 125-134

Castillo, A., Lucia, L.M., Mercado, I. and Acuff, GR. 2001. In-plant evaluation of a
lactic acid treatment for reduction of bacteria on chilled beef carcasses. J. Food Prot. 64:
738-740

CDC. 1997a Salmonella Surveillance: Annual Summary, 1996. Atlanta, Georgia: US.
Department of Health and Human Services.

CDC 1997b. Foodbome Outbreak Response and Surveillance Unit, Foodbome Disease
Outbreak Line Listing, 1997. Atlanta, Georgia: US. Department of Health and Human
Services. http://www.cdc.gov/foodborneoutbreaks/us_outbpd17fbofinal1997.pdf

CDC. 1998. Salmonella Surveillance: Annual Summary, 1997. Atlanta, Georgia: US.
Department of Health and Human Services.

CDC. 1999. Salmonella Surveillance: Annual Summary, 1998. Atlanta, Georgia: US.
Department of Health and Human Services.

CDC. 2000. Salmonella Surveillance: Annual Summary, 1999. Atlanta, Georgia: US.
Department of Health and Hunum Services.

CDC. 2001a. FoodNet Surveillance Report for 2000: Final Report. Atlanta: Centers for
Disease Control and Prevention

134

CDC. 2001b. Salmonella Surveillance: Annual Summary, 2000. Atlanta, Georgia: US.
Department of Health and Human Services.

CDC. 20010. Preliminary FoodNet Data on the Incidence of Foodbome Illnesses-
Selected Sites, United States, 2000. MMWR. 50: 241-246

CDC. 2002a. Preliminary FoodNet Data on the Incidence of Foodbome Illnesses-
Selected Sites, United States, 2001. MMWR. 51: 325-329

CDC 2002b. Multistate outbreaks of Salmonella serotype Poona infections associated
with eating cataloupe fi'om Mexico-United States and Canada, 2000-2002. MMWR. 51:
1044-1047

CDC 2003. Outbreaks of Salmonella serotype Enteritidis infection associated with eating
shell eggs-United States, 1999-2001. MMWR. 51: 1149-1152

Chantarapanont, W., Slutsker, L., Tauxe, R.V. and Beuchat, LR. 2000. Factors
influencing inactivation of Salmonella enteritidis in hard-cooked egg. J. Food Prot. 63:
36-43

Cheng, H.-Y., Yang, H.-Y. and Chou, C.-C. 2002. Influence of acid adaptation on the
tolerance of Escherichia coli 0157:H7 to some subsequent stresses. J. Food Prot. 65:
260-265

Costerton, J.W., Lewandowski, Z., Caldwell, D.E., Korber, DR. and Lappin-Scott, HM.
1995. Microbial biofilms. Annu. Rev. Microbiol. 49: 711-745

Craig, J.E., Boyle, D., Francis, K.P. and Gallagher, MP. 1998. Expression of the cold-
shock gene cspB in Salmonella lyphimurium occurs below a threshold temperature.
Microbiol. 144: 697-704

Crutchfield, S. 2001. ERS Briefing Room-Economics of foodborne disease: overview.
http://www.ers.usda.gov/briefing/FoodbomeDisease/overview.htm

Crutchfield, S.R., Buzby, J.C., Roberts, T., Ollinger, M., and Jordan Lin, C.-T. 1997. An
economic assessment of food safety regulations: the new approach to meat and poultry
inspection. ERS Agricultural Economic Report No. 755, July 1997

Czechowicz, S.M., Santos, O. and Zottola, EA. 1996. Recovery of therrnally-stressed
Escherichia coli 0157:H7 by media supplemented with pyruvate. Int. J. Food Microbiol.
33: 275-284

Dallmier, AW. and Martin, SE. 1988. Catalase and superoxide dismutase activities after
heat injury of Listeria monocytogenes. Appl. Environ. Microbiol. 54: 581-582

135

D’Aoust, J.-Y. 1981. Update on preenrichment and selective enrichment conditions for
detection of Salmonella in foods. J. Food Prot. 44: 369-374

D’Aoust, J.-Y. 1997. Salmonella Species. In Doyle, Beuchat and Montville (ed.), Food
Microbiology. Fundamentals and Frontiers. ASM Press, Washington DC, p. 129-158

D’Aoust, J.-Y., Warburton, D.W. and Sewell, A.M. 1985. Salmonella and food safety. J.
Food Prot. 48: 1062-1066

Datta, AR. and Kothary, M.H. 1993. Effects of glucose, growth temperature, and pH on

listeriolysin 0 production in Listeria monocytogenes. Appl. Environ. Microbiol. 59:
3495-3497

Dickson, J .S. and Frank, J .F . 1993. Bacterial starvation stress and contamination of beef.
Food Microbiol. 10: 215-222

Dickson, J .S. and Siragusa, GR. 1994. Survival of Salmonella typhimurium, Escherichia
coli 0157:H7 and Listeria monocytogenes Scott A during storage on beef sanitized with
organic acids. J. Food Safety.14: 313-327

Dillon, J .R. and Bezanson, GS. 1984. Microbial response to injuries from host defense
mechanisms and antibiotics. In Hurst and Nasirn (ed.), Repairable Lesions in
Microorganisms. Academic Press, London, p. 187-215

Dodd, CE. and Aldsworth, T.G. 2002. The importance of RpoS in the survival of
bacteria through food processing. Int. J. Food Microbiol. 74: 189-194

Dorman, C.J., Bhriain, N.N. and Higgins, CF. 1990. DNA supercoiling and
environmental regulation of virulence gene expression in Shigellaflexneri. Nature. 344:

789-792

Doyle, ME. and Mazzotta, AS. 2000. Review of studies on the thermal resistance of
salmonellae. J. Food Prot. 63: 779-795

ERS. 2001. Statistical Indicators, Summary Data. Economic Research Service,
Agricultural Outlook Bulletin, April 2001, AGO-280, p. 28-56

ERS. 2002. Statistical Indicators, Summary Data. Economic Research Service,
Agricultural Outlook, August 2002, AGO-293, p. 35-64

Farber, J .M. and Brown, BE. 1990. Effect of prior heat shock on heat resistance of
Listeria monocytogenes in meat. Appl. Environ. Microbiol. 56: 1584-1587

Farber, J.M. and Pagotto, F. 1992. The effect of acid shock on the heat resistance of
Listeria monocytogenes. Lett. Appl. Microbiol. 15: 197-201

136

Farr, SB. and Kogoma, T. 1991. Oxidative stress responses in Escherichia coli and
Salmonella typhimuium. Microbiol. Rev. 55: 561-585

FDA 1998. Bacteriological Analytical Manual, 8th Edition, Revision A, 1998
http://www.cfsan. fda.gov/~ebam/rl 1.html

Flowers, R.S., D’Aoust, J.-Y., Andrews, W.H. and Bailey, J.S. 1992. Salmonella. In
VanderZant and Splittstoesser (ed.), Compendium of Methods for the Microbiological
Examination of Foods. 3rd ed. APHA, Washington, D.C., p. 371-422

Flowers, R.S., Martin, S.E., Brewer, D.G. and Ordal, Z]. 1977. Catalase and
enumeration of stressed Staphylococcus aureus cells. Appl. Environ, Microbiol. 33:
1 1 12-1 1 17

Foster, J .W. 1991. Salmonella acid shock proteins are required for the adaptive tolerance
response. J. Bacteriol. 173: 6896-6902

Foster, J .W. 1995. Low pH adaptation and the acid tolerance response of Salmonella
typhimurium. Crit. Rev. Microbiol. 2 1 : 2 1 5-23 7

Foster, J .W. 2000. Microbial responses to acid stress. In Storz and Hengge-Aronis (ed.),
Bacterial Stress Responses. ASM Press, Washington D.C., p. 99-115

Foster, J .W. and Moreno, M. 1999. Inducible acid tolerance mechanisms in enteric
bacteria. Novartis Found. Symp. 221: 55-69

Foster, J .W. and Spector, MP. 1995. How Salmonella survive against the odds. Annu.
Rev. Microbiol. 49: 145-174

Frenzen, P.D., Riggs, T.L., Buzby, J.C., Breuer, T., Roberts, T., Voetsch, D. and Reddy,
S. 1999. Salmonella cost estimate updated using FoodNet data Food Safety. 22: 10-15

Frenzen, RD. 2002. ERS Briefing Room-Economics of foodborne disease: Salmonella.
http://www.ers.usdagov/briefing/FoodborneDisease/Salmonella/indexhtm

FSIS. 1999. Performance standards for the production of certain meat and poultry
products. Federal Register. Vol. 64, No. 3, p. 732-749

FSIS. 2001. Performance standards for the production of processed meat and poultry
products. Federal Register. Vol. 66, No. 39, p. 12590-12636

F SIS. 2002a. United States Department of Agriculture Food Safety and Inspection
Service, Recall Information Center. http://www.fsis.usda.gov/OA/recalls/rec_summ.htm

137

FSIS. 2002b. Progress report on Salmonella testing of raw meat and poultry products,
1998-2001. United States Department of Agriculture, Office of Public Health and

Science, Biosciences Division

Gahan, C.G.M. and Hill, C. 1999. The relationship between acid stress responses and
virulence in Salmonella typhimurium and Listeria monocytogenes. Int. J. Food Microbiol.
50: 93-100

Genther, F.J. and Martin, SE. 1990. Ribosome assembly in Salmonella typhimurium
7136 during recovery from heat shock. Food Microbiol. 7: 53-67

Gerris, CV. 2000. Use of lactic acid for carcass decontamination. Meat & Poultry. 1
ApriL 2000

Gilbert, P. 1984. The revival of micro-organisms sublethally injured by chemical
inhibitors. In Andrew and Russell (ed.), The Revival of Injured Microbes. Academic
Press, London, p. 175-197

Gilbert, RJ. and Humphrey, T.J. 1998. F ood-Bome Bacterial Gastroenteritis. In Hausler
Jr. and Sussman (ed.), Topley & Wilson’s Microbiology and Microbial Infections. Vol. 3
Bacterial Infections. 9th ed. Arnold, London, p. 539-565

Girgis, H.S., Smith, J ., Luchansky, J .B. and Klaenhammer, TR. 2003. Stress adaptations
of lactic acid bacteria. In Yousef and Juneja (ed.), Microbial Stress Adaptation and Food
Safety. CRC Press, Boca Raton, p. 159-211

Givskov, M., Eberl, L., Meller, S., Poulsen, L.K and Molin, S. 1994. Responses to
nutrient starvation in Pseudomonas putida KT2442: analysis of general cross-protection,

cell shape, and macromolecular content. J. Bacteriol. 176: 7-14

Goldstein, J., Pollitt, NS. and Inouye, M. 1990. Major cold-shock protein of Escherichia
coli. Proc. Natl. Acad. Sci. USA. 87: 283-287

Gomez, RF. and Sinskey, A.J. 1975. Effect of aeration on minimal medium recovery of
heated Salmonella (whimurium. J. Bacteriol. 122: 106-109

Gort, AS. and lmlay, J .A. 1998. Balance between endogenous superoxide stress and
antioxidant defenses. J. Bacteriol. 180: 1402-1410

Gould, G.W. 1984. Injury and repair mechanisms in bacterial spores. In Andrew and
Russell (ed.), The Revival of Injured Microbes. Academic Press, London, p. 199-220

Graumann, P. and Marahiel, MA. 1996. Some like it cold: response of microorganisms
to cold shock. Arch Microbiol. 166: 293-300

138

Graumann, P., Schroder, K, Schmid, R. and Marahiel, MA. 1996. Cold shock stress-
induced proteins in Bacillus subtilis. J. Bacteriol. 178: 4611-4619

Guidot, P.M., McCord, J.M., Wright, R.M. and Repine, J .E. 1993. Absence of electron
transport restores growth of a manganese-superoxide dismutase deficient Saccharomyces
cerevisiae. J. Biol. Chem. 68: 26699-26703

Hansen, N.-H. and Riemann, H. 1963. Factors affecting the heat resistance of nonsporing
organisms. J. Appl. Bacteriol. 26: 314-333

Harris, ND. 1963. The influence of the recovery medium and the incubation temperature
on the survival of damaged bacteria. J. Appl. Bacteriol. 26: 387-397

Hecker, M., Schumann, W. and Volker, U. 1996. Heat-shock and general stress response
in Bacillus subtilis. Mol. Microbiol. 19: 417-428

Hengge-Aronis, R. 1993. Survival of hunger and stress: the role of lpoS in early
stationary phase gene regulation in E. coli. Cell. 72: 165-168

Hengge-Aronis, R. 2000. The general stress response in Escherichia coli. In Storz and
Hengge-Aronis (ed.), Bacterial Stress Responses. ASM Press, Washington D.C., p. 161-
178

Humphrey, T.J. 1988. Peroxide sensitivity and catalase activity in Campylobacterjejuni
after injury and during recovery. J. Appl. Bacteriol. 64: 337-343

Humphrey, T.J. 1990. Heat resistance in Salmonella enteritidis phage type 4: the
influence of storage temperatures before heating. J. Appl. Bacteriol. 69: 493-497

Humphrey, T.J., Wilde, SJ. and Rowbury, R.J. 1997. Heat tolerance of Salmonella
typhimurium DT104 isolates attached to muscle tissue. Lett. Appl. Microbiol. 25: 265-
268

Humphrey, T.J., Williams, A., McAlpine, K., Lever, M.S., Guard-Fetter, J. and Cox, J.M.
1996. Isolates of Salmonella enterica Enteritidis PT4 with enhanced heat and acid
tolerance are more virulent in mice and more invasive in chickens. Epidemiol. Infect.
177: 79-88

Hurst, A. 1977. Bacterial injury: a review. Can. J. Microbiol 23: 935-944

Hurst, A. 1984. Revival of vegetative bacteria after sublethal heating. In Andrew and
Russell (ed.), The Revival of Injured Microbes. Academic Press, London, p. 77-103

Hurst, A., Hendry, G.S., Hughes, A. and Paley, B. 1976. Enumeration of sublethally
heated Staphylococci in some dried foods. Can. J. Microbiol. 22: 677-683

139

Hurst, A. and Hughes, A. 1981. Repair of salt tolerance and repair of lost D-alanine and
magnesium following sublethal heating of Staphylococcus aureus are independent events.
Can. J. Microbiol. 27: 627-632

IFT. 2002. Expert Report on Emerging Microbiological Food Safety Issues. Implications
for Control in the 21“ Century. Institute of Food Technologists, International Food Safety
and Quality Conference and Exposition, Atlanta, GA. 20 February 2002

Jackson, H. and Woodbine, M. 1963. The effect of sublethal heat treatment on the growth
of Staphylococcus aureus. J. Appl. Bacteriol. 26: 152-158

Jay, J.M. 1998. Modern Food Microbiology. 5th ed. Aspen Publishers, Inc., Gaithersburg

Jeffreys, A.G., Hak, KM., Steffan, R.J., Foster, J.W. and Bej, AK. 1998. Growth,
survival and characterization of cspA in Salmonella enteritidis following cold shock.
Curr. Microbiol. 36: 29-35

Jenkins, D.E., Schultz, J.E. and Matin, A. 1988. Starvation-induced cross protection
against heat or H202 challenge in Escherichia coli. J. Bacteriol. 170: 3910-3914

Johnson, EA. 2003. Microbial adaptation and survival in foods. In Yousef and Juneja
(ed.), Microbial Stress Adaptation and Food Safety. CRC Press, Boca Raton, p. 75-103

Johnson, KM. and Busta, RF. 1984. Detection and enumeration of injured bacterial

Spores in processed foods. In Andrew and Russell (ed.), The Revival of Injured Microbes.
Academic Press, London, p. 241-256

Jones, K. 2001. ERS Agricultural Economist, Personal Communication

Jones, P.G., Mitta, M., Kim, Y., Jiang, W. and Inouyi, M. 1996. Cold shock induces a
major ribosomal-associated protein that unwinds double-stranded RNA in Escherichia
coli. Proc. Natl. Acad. Sci. USA. 93: 76-80

Jones, P.G., VanBogelen, R. and Neidhardt, RC. 1987. Induction of proteins in response
to low temperature in Escherichia coli. J. Bacteriol. 169: 2092-2095

Juneja, V.K. and Eblen, BS. 1999. Predictive thermal inactivation model for Listeria
monocytogenes with temperature, pH, NaCl, and sodium pyrophosphate as controlling
factors. J. Food Prot. 62: 986-993

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

Juneja, V.K., Klein, P.G. and Marmer, BS. 1998. Heat shock and thermotolerance of

Escherichia coli 0157:H7 in a model beef gravy system and ground beef. J. Appl.
Microbiol. 84: 677-684

140

Juneja, V.K. and Novak, J .S. 2003. Adaptation of foodborne pathogens to stress from
exposure to physical intervention strategies. In Yousef and Juneja (ed.), Microbial Stress
Adaptation and Food Safety. CRC Press, Boca Raton, p. 31-53

Kang, DH. and Siragusa, GR. 1999. Agar underlay method for recovery of sublethally
heat-injured bacteria. Appl. Environ. Microbiol. 65: 5334-5337

Kapoor, S., Singh, R.D., Sharma, RC. and Khullar, M. 2002. Anaerobiosis induced
virulence of Salmonella typhi. Indian J. Med. Res. 115: 184-188

Kaufinann, SH. 1989. Stress proteins: virulence factors of intracellular disease agents?
Immun. Infekt. 17: 124-128

Kennedy, J .E., Oblinger, J.L. and West, KL. 1980. Fate of Salmonella infantis,
Staphylococcus aureus, and Hafnia alvei in vacuum packaged beef plate pieces during
refi'igerated storage. J. Food Sci. 45: 1273-1277, 1300

Knabel, SJ. and Thielen, SA. 1995. Enhanced recovery of severely heat-injured,
thermotolerant Listeria monocytogenes from FDA primary enrichment media using a
novel, simple, strictly anaerobic method. J. Food Prot. 58: 29-34

Knabel, S.J., Walker, H.W., Hartman, RA. and Mendonca, AF. 1990. Effects of growth
temperature and strictly anaerobic recovery on the survival of Listeria monocytogenes
during pasteurization. Appl. Environ. Microbiol. 56: 370-376

Knechel, S. and Gould, G. 1995. Preservation microbiology and safety: Quo vadis?
Trends Food Sci. Technol. 6: 127-131

Kushal, R and Anand, SK. 1999. Repair and recovery of thermally injured cells of
Yersinia enterocolitica in milk. J. Food Prot. 62: 1203-1205

Lazazzera, BA. 2000. Quortun sensing and starvation: signals for entry into stationary
phase. Curr. Opin. Microbiol. 3: 177-182

Lee, L.S., Lin, J., Hall, H.K., Bearson, B. and Foster, J.W. 1995. The stationary-phase
sigma factor as (RpoS) is required for a sustained acid tolerance response in virulent
Salmonella typhimurium. Mol. Microbiol. 17: 155-167

Leenanon, B. and Drake, MA. 2001. Acid stress, starvation, and cold stress affect
poststress behavior of Escherichia coli 0157:H7 and nonpathogenic Escherichia coli. J.
Food Prot. 64: 970-974

Leimeister-Wachter, M., Domann, E. and Chakraborty, T. 1992. The expression of
virulence genes in Listeria monocytogenes is therrnoregulated. J. Bacteriol. 174: 947-952

141

Le Minor, L. 1992. The Genus Salmonella. In Balows, Triiper, Dworkin, Harder and
Schleifer (ed.), The Prokaryotes. A Handbook on the Biology of Bacteria:
Ec0physiology, Isolation, Identification, Applications. Vol. 3, 2“d ed. Springer-Verlag,
New York. p. 2760-2774

Levine, P., Rose, B., Green, S., Ransom, G. and Hill, W. 2001. Pathogen testing of
Ready-to-Eat meat and poultry products collected at Federally inspected establishments
in the United States, 1990 to 1999. J. Food Prot. 64: 1188-1193

Levinson, R. 2000. Testimony on Food Safety and Microbial Contamination presented by
Richard Levinson, Associate Executive Director of the American Public Health
Association to the Senate Committee on Agriculture, Nutrition, and Forestry. September
20, 2000

Leyer, OJ. and Johnson, EA. 1993. Acid adaptation induces cross-protection against
environmental stresses in Salmonella typhimurium. Appl. Environ. Microbiol. 59: 1842-
1847

Lindquist, S. 1992. Heat-shock proteins and stress tolerance in microorganisms. Curr.
Opin. Gen. Develop. 2: 748-755

Line, J.E., Fain, A.R., Moran, A.B., Martin, L.M., Lechowich, R.V., Carosella, J.M. and
Brown, W.L. 1991. Lethality of heat to Escherichia coli 0157:H7: D-value and z-value
determinations in ground beef. J. Food Prot. 54: 762-766

Linton, R.H., Webster, J.B., Pierson, M.D., Bishop, JR. and Hackney, CR. 1992. The
effect of sublethal heat shock and grth atmosphere on the heat resistance of Listeria
monocytogenes ScottA. J. Food Prot. 55: 84-87

Loewen, P.C., Hu, 3., Strutinsky, J. and Sparling, R. 1998. Regulation in the rpoS
regulon of Escherichia coli. Can. J. Microbiol. 44: 707-717

Lottering, EA. and Streips, UN. 1995. Induction of cold shock proteins in Bacillus
subtilis. Curr. Microbiol. 30: 193-199

Lou, Y. and Yousef, AB. 1996. Resistance of Listeria monocytogenes to heat after
adaptation to environmental stresses. J. Food Prot. 59: 465-471

Lou, Y. and Yousef, AB. 1997. Adaptation to sublethal environmental stresses protects
Listeria monocytogenes against lethal preservation factors. Appl. Environ Microbiol. 63:
1252-1255

Mackey, BM. 1984. Lethal and sublethal effects of refrigeration, fi'eezing and freeze-

drying on micro-organsisms. In Andrew and Russell (ed.), The Revival of Injured
Microbes. Academic Press, London, p. 45-75

142

Mackey, BM. 2000. Injured bacteria. In Land, Baird-Parker and Gould (ed.), The
Microbiological Safety and Quality of Food. Aspen Publishers, Inc., Gaithersburg, p.
315-341

Mackey, B.M., Boogard, E., Hayes, CM. and Baranyi, J. 1994. Recovery of heat-injured
Listeria monocytogenes. Int. J. Food Microbiol. 22: 227-237

Mackey, HM. and Derrick, CM. 1982. The effect of sublethal injury by heating,
fieezing, drying and gamma-radiation on the duration of the lag phase of Salmonella
typhimurium. J. Appl. Bacteriol. 53: 243-251

Mackey, B.M. and Derrick, CM. 1986. Elevation of the heat resistance of Salmonella
typhimurium by sublethal heat shock. J. Appl. Bacteriol. 61: 389-393

Mackey, B.M. and Derrick, C.M. 1987a. Changes in the heat resistance of Salmonella
typhimurium during heating at rising temperatures. Lett. Appl. Microbiol. 4: 13-16

Mackey, B.M. and Derrick, C.M. 1987b. The effect of prior heat shock on the
thermoresistance of Salmonella thompson in foods. Lett. Appl. Microbiol. 5: 115-118

Mackey, B.M. and Seymour, DA. 1987. The effect of catalase on recovery of heat-
injured DNA-repair mutants of Escherichia coli. J. Gen. Microbiol. 133: 1601-1610

Madigan, M.T., Martinko, J.M., and Parker, J. 1997. Brock Biology of Microorganisms.
8tln ed. Prentice Hall, Upper Saddle River

Matias, P., Pagan, R., Leguérinel, 1., Condon, S., Mafart, P. and Sala, F. 2001. Effect of
sodium chloride concentration on the heat resistance and recovery of Salmonella
typhimurium. Int. J. Food Microbiol. 63: 209-216

Martin, S.E., Flowers, RS and Ordal, Z.J. 1976. Catalase: its efiect on microbial
enumeration Appl. Environ Microbiol. 32: 731-734

Mathew, F .P. and Ryser, ET. 2002. Competition of thermally injured Listeria
monocytogenes with a mesophilic lactic acid starter culture in milk for various heat
treatments. J. Food Prot. 65: 643-650

Matin, A. 1991. The molecular basis of carbon-starvation-induced general resistance in
Escherichia coli. Mol. Microbiol. 5: 3-10

Mattick, K.L., Jergensen, F ., Legan, J .D., Lappin-Scott, HM. and Humphrey, T.J. 2000.

Habituation of Salmonella spp. at reduced water activity and its effect on heat tolerance.
Appl. Envrion. Microbiol. 66: 4921-4925

143

Mattick, K.L., Jargensen, R, Wang, P., Pound, J., Vandeven, M.H., Ward, L.R., Legan,
J.D., Lappin-Scott, HM. and Humphrey, T.J. 2001. Effect of challenge temperature and
solute type on heat tolerance of Salmonella serovars at low water activity. Appl. Environ
Microbiol. 67: 4128-4136

Maurelli, A.T. 1989. Regulation of virulence genes in Shigella. Mol. Biol. Med. 6: 425-
432

Maurer, J .L. 2001. Environmental effects on the thermal resistance of Salmonella,
Escherichia coli 0157:H7, and triose phosphate isomerase in ground turkey and beef.
MS. Thesis. Michigan State University, East Lansing, MI

Mayr, B., Kaplan, T., Lechner, S. and Scherer, S. 1996. Identification and purification of
a family of dirneric major cold shock protein homologs fiom the psychrotrophic Bacillus
cereus WSBC 10201. J. Bacteriol. 178: 2916-2925

Mazzotta, AS. 2001. Thermal inactivation of stationary-phase and acid-adapted
Escherichia coli 0157:H7, Salmonella, and Listeria monocytogenes in fruit juices. J.
Food Prot. 64: 315-320

McClements, J.M.J, Patterson, M.F. and Linton, M. 2001. The effect of growth stage and
growth temperature on high hydrostatic pressure inactivation of some psychrotrophic
bacteria in milk. J. Food Prot. 64: 514-522

McClure, P.J., Roberts, TA. and Oguru, PD. 1989. Comparison of the effects of sodium
chloride, pH and temperature on the growth of Listeria monocytogenes on gradient plates
and in a liquid medium. Lett. Appl. Microbiol. 9: 95-99

McDonald, L.C., Hackney, CR. and Ray, B. 1983. Enhanced recovery of injured
Escherichia coli by compounds that degrade hydrogen peroxide or block its formation.
Appl. Environ. Microbiol. 45: 360-365

McGovern, VP. and Oliver, J .D. 1995. Induction of cold responsive proteins in Vibrio
vulnificus. J. Bacteriol. 177: 4131-4133

McMahon, C.M.M., Byrne, C.M., Sheridan, J.J., McDowell, D.A., Blair, LS. and
Hegarty, T. 2000. The effect of culture growth phase on induction of the heat shock
response in Yersinia enterocolitica and Listeria monocytogenes. J. Appl. Microbiol. 89:

198-206

Mead, P.S., Laurence, S., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., Griffin,
P.M., and Tauxe, R.V. 1999. Food-related illness and death in the United States. Emerg.
Infect. Dis. 5: 607-625

Mekalanos, J .J . 1992. Environmental signals controlling expression of virulence
determinants in bacteria. J. Bacteriol. 174: 1-7

144

 

Meyer, DH. and Donnelly, CW. 1992. Effect of incubation temperature on repair of
heat-injured Listeria in milk. J. Food Prot. 55: 579-582

Miller, A.J., Bayles, DO, and Eblen, BS. 2000. Cold shock induction of thermal
sensitivity in Listeria monocytogenes. Appl. Environ. Microbiol. 66: 4345-4350

Mitchell, DO. and Starzyk, M.J. 1975. Survival of Salmonella and other indicator
microorganisms. Can. J. Microbiol. 21: 1420-1421

Mizunoe, Y., Wai, S.N., Takade, A. and Yoshida, S. 1999. Restoration of culturability of
starvation-stressed and low-temperature-stressed Escherichia coli 0157 cells by using
H202-degrading compounds. Arch. Microbiol. 172: 63-67

Moats, W.A., Dabbah, R. and Edwards, V.M. 1971. Survival of Salmonella anatum
heated in various media. Appl. Microbiol 21: 476-481

Montville, T.J. 1997. Principles which influence microbial growth, survival, and death in
foods. In Doyle, Beuchat and Montville (ed.), Food Microbiology, Fundamentals and
Frontiers. ASM Press, Washington D.C., p. 13-29

Morifiigo, M.A., Martinez-Manzanares, E., Mufioz, A., Cornax, R., Romero, P. and
Borrego, J.J. 1989. Evaluation of different plating media used in the isolation of
salmonella from environmental samples. J. Appl. Bacteriol. 66: 353-360

Mossel], D.A.A., and van Netten, P. 1984. Harmful effects of selective media on stressed
micro-organisms: nature and remedies. In Andrew and Russell (ed.), The Revival of
Injured Microbes. Academic Press, London, p. 329-369

Musarrat, J. and Ahmad, M. 1988. pH induced damage and repair in E. coli. Mutat. Res.
193: 219-227

Murano, EA. and Pierson, MD. 1992. Effect of heat shock and growth atmosphere on
the heat resistance of Escherichia coli 0157:H7. J. Food Prot. 55: 171-175

Murano, EA. and Pierson, MD. 1993. Effect of heat shock and incubation atmosphere
on injury and recovery of Escherichia coli 0157:H7. J. Food Prot. 56: 568-572

Murphy, R.Y., Marks, B.P., Johnson, ER, and Johnson, M.G. 1999. Inactivation of
Salmonella and Listeria in ground chicken breast meat during thermal processing. J. Food
Prot. 62: 980-985

Murphy, RY, Duncan, L.K., Johnson, E.R., Davis, MD. and Smith, J.N. 2002. Thermal
inactivation D- and z-values of Salmonella serotypes and Listeria innocua in chicken
patties, chicken tenders, flanks, beef patties, and blended beef and turkey patties. J. Food
Prot. 65: 53-60

145

Nebra, Y., Jofre, J. and Blanch, AR. 2002. The effect of reducing agents on the recovery
of injured Bifidobacterium cells. J. Microbiol. Methods. 49: 247-254

Neidhardt, F .C. and VanBogelen, RA. 1987. Heat Shock Response. In Neidhardt,
Ingraham, Low, Magasanik, Schaechter and Umbarger (ed.), Escherichia coli and
Salmonella Typhimurium: Cellular and Molecular Biology. ASM Press, Washington
D.C., p. 1334-1345

Novak, J .S., Tunick, M.H. and Juneja, V.K. 2001. Heat treatment adaptations in
Clostridium perfringens vegetative cells. J. Food Prot. 64: 1527-1534

O’Driscoll, B., Gahan, C.G.M. and Hill, C. 1996. Adaptive acid tolerance response in
Listeria monocytogenes: isolation of an acid-tolerant mutant which demonstrates
increased virulence. Appl. Environ. Microbiol. 62: 1693-1698

Oerskov, J .L. 1925. Les bacilles appartenant aux groupes typhique, paratyphique,
paradysentérie, et coli, deviennent, aprés chauffagé hypersensibles a l’action de
l’oxygene atmosphérique. CR Soc. Biol, Paris. 92: 402

Oliver, J .D. 1981. Lethal cold stress of Vibrio vulnificus in oysters. Appl. Environ.
Microbiol. 41: 710-717

Pagan, R., Condon, S., and Sala, F.J. 1997. Effects of several factors on the heat-shock-
induced thermotolerance of Listeria monocytogenes. Appl. Environ Microbiol. 63: 3225-
3232

Pagan, R., Matias, P., Palop, A. and Sala, F .J. 1999. Resistance of heat-shocked cells of
Listeria monocytogenes to mano-sonication and mano-therrno-sonication. Lett. Appl.
Microbiol. 28: 71-75

Palumbo, SA. 1984. Heat injury and repair in Campylobacterjejuni. Appl. Environ
Microbiol. 48: 477-480

Pascual, C., Robinson, T.P., Ocio, M.J., Aboaba, 0.0. and Mackey, BM. 2001. The
effect of inoculum size and sublethal injury on the ability of Listeria monocytogenes to
initiate growth under suboptimal conditions. Lett. Appl. Microbiol. 33: 357-361

Patel, J .R and Beuchat, LR. 1995. Enrichment in Fraser broth supplemented with
catalase of Oxyrase®, combined with the microcolony immunoblot technique, for
detecting heat-injured Listeria monocytogenes in foods. Int. J. Food Microbiol. 26: 165-
176

Patel, J.R., Hwang, C.-A., Beuchat, L.R., Doyle, MP. and Brackett, RE. 1995.

Comparison of oxygen scavengers for their ability to enhance resuscitation of heat-
injured Listeria monocytogenes. J. Food Prot. 58: 244-250

146

Pearson, A.M. and Gillett, TA. 1996. Processed Meats. 3rd ed. Chapman and Hall, New
York

Peleg, M. and Cole, MB. 1998. Reinterpretation of microbial survival curves. Crit. Rev.
Food Sci. 38: 353-380

Pellon, IR and Sinskey, A.J. 1984. Heat-induced damage to the bacterial chromosome
and its repair. In Andrew and Russell (ed.), The Revival of Injured Microbes. Academic
Press, London, p. 106-125

Poppe, C. 1999. Epidemiology of Salmonella enterica Serovar Enteritidis. In Saeed,
Potter and Wall (ed.), Salmonella enterica Serovar Enteritidis in Humans and Animals.
Epidemiology, Pathogenesis and Control. Iowa State University Press, Ames, p. 3-18

Postgate, JR. and Hunter, JR. 1963. The survival of starved bacteria. J. Appl. Bacteriol.
26: 295-306

Przybylski, KS. and Witter, L.D. 1979. Injury and recovery of Escherichia coli after
sublethal acidification Appl. Environ. Microbiol. 37: 261-265

Quesnel, L.B. 1984. Biological indicators and sterilization processes. In Andrew and
Russell (ed.), The Revival of Injured Microbes. Academic Press, London, p. 257-291

Quintavalla, S. and Campanini, M. 1991. Effect of rising temperature on the heat
resistance of Listeria monocytogenes in meat emulsion Lett. Appl. Microbiol. 12: 184-
187

Quintavalla, S., Larini, S., Mutti, P. and Barbuti, S. 2001. Evaluation of the thermal
resistance of different Salmonella serotypes in pork meat containing curing additives. Int.
J. Food Microbiol. 67: 107-114

Rattanasomboon, N., Bellara, S.R., Fryer, P.J., Thomas, CR. and McFarlane, CM. 2001.
The gradient plate technique as a means of studying the recovery of heat-injured
Brochothrix thermosphacta. Int. J. Food Sci. Technol. 36: 369-3 76

Ray, B. 1984. Reversible freeze-injury. In Hurst and Nasim (ed.), Repairable Lesions in
Microorganisms. Academic Press, London, p. 237-271

Ray, B. 1986. Impact of bacterial injury and repair in food microbiology: its past, present
and filture. J. Food Prot. 49: 651-655

Ray, B. 1989. Injured Index and Pathogenic Bacteria: Occurrence and Detection in
Foods, Water and Feeds. CRC Press, Inc. Boca Raton

Ray, B. 2001. Fundamental Food Microbiology. 2"‘1 ed. CRC Press, Boca Raton

147

Rayman, M.K., Aris, B. and El Derea, HE. 1978. The effect of compounds which
degrade hydrogen peroxide on the enumeration of heat-stressed cells of Salmonella
seftenberg. Can J. Microbiol. 24: 883-885

Rees, C.E.D., Dodd, C.E.R., Gibson, P.T., Booth, LR. and Stewart, G.S.A.B. 1995. The
significance of bacteria in stationary phase to food microbiology. Int. J. Food Microbiol.
28: 263-275

Reichart, 0. 1994. Modelling the destruction of Escherichia coli on the base of reaction
kinetics. Int. J. Food Microbiol. 23: 449-465

Reissbrodt, R, Rienaecker, 1., Romanova, J.M., Freestone, P.P.E., Haigh, R.D., Lyte, M.,
Tschape, H. and Williams, RH. 2002. Resuscitation of Salmonella enterica serovar
Typhimurium and enterohemorrhagic Escherichia coli from the viable but nonculturable
state by heat-stable enterobacterial autoinducer. Appl. Environ. Microbiol. 68: 4788-4794

Restaino, L., Jeter, W.S. and Hill, W.M. 1980. Thermal injury of Yersinia enterocolitica.
Appl. Environ Microbiol. 40: 939-949

Rexroat, M.A., Oberly, T.J., Bewsey, B.J. and Garriott, ML. 1995. The gradient plate
assay: a modified Ames assay used as a prescreen for the identification of bacterial
mutagens. Mutat. Res. 341: 185-192

Ribot, E., Wierzba, R., Angulo, F. and Barrett, T. 2002. Salmonella enterica serotype
Typhimurium DT104 isolated from humans, United States, 1985, 1990, and 1995.
Emerg. Infect. Dis. 8: 387-391

Romans, J.R., Costello, W.J., Wendell Carlson, C., Greaser, M.L., and Jones, K.W. 2001.
The Meat We Eat. l4til ed. Interstate Publishers, Inc., Danville

Rowbury, R.J. 2003. Physiology and molecular basis of stress adaptation, with particular
reference to the subversion of stress adaptation, and to the involvement of extracellular
components in adaptation In Yousef and Juneja (ed.), Microbial Stress Adaptation and
Food Safety. CRC Press, Boca Raton, p. 247-302

Rowe, MT and Kirk, RB. 2000. Effect of nutrient starvation on the resistance of
Escherichia coli 0157:H7 to subsequent heat stress. J. Food Prot. 63: 1745-1748

Russell, AD. 1984. Potential sites of damage in micro-organisms exposed to chemical or
physical agents. In Andrew and Russell (ed.), The Revival of Injured Microbes.
Academic Press, London, p. 1-18

Sato, M. and Takahashi, H. 1970. Cold-shock of bacteria. IV. Involvement of DNA-

ligase reaction in recovery of Escherichia coli fi'om cold shock. J. Gen. Appl. Microbiol.
16: 279-290

148

Schellhorn, HE. and Hassan, HM. 1988. Response of hydroperoxidase and superoxide
dismutase deficient mutants of Escherichia coli K-12 to oxidative stress. Can. J.
Microbiol. 34: 1171-1176

Semanchek, J .J . and Golden, DA. 1998. Influence of growth temperature on inactivation
and injury of Escherichia coli 0157:H7 by heat, acid, and freezing. J. Food Prot. 61: 395-
401

Shallow, S., Vugia, D., Fiorentino, T., Mshar, P., Farley, M., Lance-Parker, 8., Pass, M.,
Karchmer, T., Carter, M., Smith, K, Zansky, S., and Cassidy, M. 2000. Preliminary
FoodNet data on the incidence of foodborne illnesses-selected sites, United States, 1999.
MMWR Weekly. 49: 201-205

Simpson, J.A., Narita, S., Gieseg, S., Gebicki, S., Gebicki, J.M. and Dean, RT. 1992.
Long-lived reactive species on free-radical-damaged proteins. Biochem. J. 282: 621-624

Singh, A. and McFeters, GA. 1987. Survival and virulence of copper- and chlorine-
stressed Yersinia enterocolitica in experimentally infected mice. Appl. Environ.
Microbiol. 53: 1768-1774

Skurnik, M. and Toivanen, P. 1992. Lch is the temperature-regulated activator of the
yadA gene of Yersinia enterocolitica and Yersinia pseudotuberculosis. J. Bacteriol. 174:
2047-2051

Sleator, RD. and Hill, C. 2002. Bacterial osmoadaptation: the role of osmolytes in
bacterial stress and virulence. FEMS Microbiol. Rev. 26: 49-71

Smirnova, G.V., Zakirova, 0N. and Oktiabr’skii, 0.N. 2001a. The role of antioxidant
systems in response of Escherichia coli to heat shock. Mikrobiol. 70: 595-601

Smirnova, G.V., Zakirova, 0N. and Oktiabr’skii, O.N. 2001b. Role of the antioxidant
system in response of Escherichia coli to cold stress. Mikrobiol. 70: 55-60

Smith, J.L. and Dell, B.J. 1990. Capability of selective media to detect heat-injured
Shigellaflexneri. J. Food Prot. 45: 360-365

Smith, J.L. and Palumbo, SA. 1978. Injury to Staphylococcus aureus during sausage
fermentation. Appl. Environ Microbiol. 36: 857-860

Smith, SE. 2000. R-phycoerythrin as a time-temperature integrator, and triose phosphate
isomerase as a temperature indicator of thermal processing in beef patties. MS. Thesis.
Michigan State University, East Lansing, MI

Snyder, L. and Champness, W. 1997. Molecular Genetics of Bacteria. ASM Press,
Washington DC.

149

Sokolovic, Z., Fuchs, A. and Goebel, W. 1990. Synthesis of species-specific stress
proteins by virulent strains of Listeria monocytogenes. Infect. Immun 58: 3582-3587

Sokolovic, Z., Riedel, J., Wuenscher, M. and Goebel, W. 1993. Surface-associated, PrfA-
regulated proteins of Listeria monocytogenes synthesized under stress conditions. Mol.
Microbiol. 8: 219-227

Southard, L., Gustafson, R., Miller, J., and Harvey, D. 2000. Meat and poultry production
to continue record-setting pace. ERS Agricultural Outlook Bulletin, June-July 2000, p. 5-
6

Stead, D. and Park, SF. 2000. Roles of superoxide dismutase and catalase in resistance of
Campylobacter coli to fi'eeze-thaw stress. Appl. Environ. Microbiol. 66: 3110-3112

Stephens, P.J., Cole, MB. and Jones, M.V. 1994. Effect of heating rate on the thermal
inactivation of Listeria monocytogenes. J. Appl. Bacteriol. 77: 702-708

Stephens, P.J., Druggan, P. and Nebe-von Caron, G. 2000. Stressed Salmonella are

exposed to reactive oxygen species from two independent sources during recovery in
conventional culture media. Int. J. Food Microbiol. 60: 269-285

Stephens, P.J., Joynson, J.A., Davies, K. W., Holbrook, R., Lappin-Scott, HM. and
Humphrey, T.J. 1997. The use of an automated growth analyzer to measure recovery
times of single heat-injured Salmonella cells. J. Appl. Microbiol. 83: 445-455

Storz, G. and Hengge-Aronis, R. 2000. Preface. In Storz and Hengge-Aronis (ed.),
Bacterial Stress Responses. ASM Press, Washington D.C., p. xii-xiv

Storz, G. and Zheng, M. 2000. Oxidative stress. In Storz and Hengge-Aronis (ed.),
Bacterial Stress Responses. ASM Press, Washington D.C., p. 47-59

Suh, J.-H. and Knabel, SJ. 2000. Comparison of different reducing agents for enhanced
detection of heat-injured Listeria monocytogenes. J. Food Prot. 63: 1058-1063

Sunil, B. 2000. The true impact of food borne infections. J. Indian Vet. Assoc., Kerala.
vol. 5. http://www.jivaonline.com

Sykes, G. 1963. The phenomenon of bacterial survival. J. Appl. Bacteriol. 26: 287-294

Takeda, Y. and Avila, H. 1986. Structure and gene expression of the E. coli Mn-
superoxide dismutase gene. Nucleic Acids Res. 14: 4577-4589

Taormina, P.J. and Beuchat, LR. 2001. Survival and heat resistance of Listeria

monocytogenes after exposure to alkali and chlorine. Appl. Environ Microbiol. 67: 2555-
2563

150

Teo, A.Y. and Knabel, SJ. 2000. Development of a simple recovery-enrichment system
for enhanced detection of heat-injured Listeria monocytogenes in pasteurized milk. J.
Food Prot. 63: 462-472

Thanomsub, B., Anupunpisit, V., Chanphetch, S., Watcharachaipong, T., Poonkhum, R.
and Srisukonth, C. 2002. Effects of ozone treatment on cell growth and ultrastructural
changes in bacteria. J. Gen. Appl. Microbiol. 48: 193-199

Thippareddi, H., Phebus, R.K., Fung, D.Y.C. and Kastner, CL. 1995. Use of universal
preenrichment medium supplemented with OxyraseTM for the simultaneous recovery of
Escherichia coli 0157:H7 and Yersinia enterocolitica. J. Rap. Methods Auto. Microbiol.
4: 37-50

Thomas, L.V., Wimpenny, J .W.T. and Peters, AC. 1992. Testing multiple variables on

the growth of a mixed inoculum of Salmonella strains using gradient plates. Int.J. Food
Microbiol. 15: 165-175

Talker-Nielsen, T. and Molin, S. 1996. Role of ribosome degradation in the death of
heat-stressed Salmonella typhimurium. FEMS Microbiol. Lett. 142: 155-160

Tompkin, RB. 1986. Microbiology of Ready-to-Eat Meat and Poultry Products. Adv.
Meat Research. 2: 89-121

Toomey, K.E., Blake, PA. and Ralston, M. 2001. Foodbome outbreaks associated with
eating pork barbecue-Georgia, 1996-2000. Georgia Epidemiology Report. 17: 1-3

Tran, T.T. 1995. Evaluation of Oxyrase® enrichment method for isolation of
Campylobacterjejuni from inoculated foods. Lett. Appl. Microbiol. 21: 345-347

Tsuchido, T., VanBogelen, RA. and Neidhardt, F .C. 1986. Heat shock response in
Escherichia coli influences cell division. Proc. Natl. Acad. Sci. USA. 83: 6959-6963

Valentine, J .S., Foote, C.S., Greenberg, A. and Liebman, J.F. 1995. Active Oxygen in
Biochemistry. Blackie Academic & Professional, London

Valone, S.E., Chikami, GK. and Miller, V.L. 1993. Stress induction of the virulence
proteins (Sva, -B, and -C) from native plasmid pSDL2 of Salmonella dublin. Infect.
Imrnun 61: 705-713

VanBogelen, RA. and Neidhardt, RC. 1990. Ribosomes as sensors of heat and cold
shock in Escherichia coli. Proc. Natl. Acad. Sci. USA. 87: 5589-5593

Van Schothorst, M. and Maggie Duke, A. 1984. Effect of sample handling on microbial

limits laid down in standards. In Andrew and Russell (ed.), The Revival of Injured
Microbes. Academic Press, London, p. 309-328

151

Vasconcelos, J .A. and Deneer, H.G. 1994. Expression of superoxide dismutase in Listeria
monocytogenes. 60: 2360-2366

Vtilker, U., Mach, H., Schmid, R. and Hecker, M. 1992. Stress proteins and cross-
protection by heat shock and salt stress in Bacillus subtilis. J. Gen. Microbiol. 138: 2125-
2135

Warburton, D.W., Bowen, B., Konkle, A., Crawford, C., Durzi, 8., Foster, R., Fox, C.,
Gour, L., Krohn, G., LaCasse, P., Larnontagne, G., McDonagh, S., Arling, V.,
Mackenzie, J., Todd, E.C.D., Oggel, J., Plante, R., Shaw, 8., Tiwari, N.P., Trottier, Y.
and Wheeler, B.D. 1994. A comparison of six different plating media used in isolation of
Salmonella. Int. J. Food Microbiol. 22: 277-289

Watson, K. 1990. Microbial Stress Proteins. Adv. Microb. Pysiol. 31: 183-223

Whiting, RC. 1995. Microbial modeling in foods. Crit. Rev. Food Sci. Nutrit. 35: 467-
494

Wimpenny, J.W.T. and Waters, P. 1984. Growth of microorganisms in gel-stabilised two-
dirnensional diffusion J. Gen. Microbiol. 130: 2921-2926

Wong, H.-C., Peng, P.-Y., Lan, S.-L., Chen, Y.-C., Lu, K.-H., Shen, C.-T. and Lan, S.-F.
2002. Effects of heat shock on the thermotolerance, protein composition, and toxin
production of Vibrio parahaemolyticus. J. Food Prot. 65: 499-507

Wouters, J.A., Rombouts, F.M., Kuipers, 0.P., de Vos, W.M. and Abee, T. 2000. The
role of cold-shock proteins in low-temperature adaptation of food-related bacteria. Syst.
Appl. Microbiol. 23: 165-173

Wouters, J.A., Sanders, J.-W., Kok, J., De Vos, W.M., Kuipers, OR and Abee, T. 1998.
Clustered organization and transcriptional analysis of a family of five csp genes of
Lactococcus lactis MG1363. Microbiol. 144: 2885-2893

Wright, QT. and Klaenhammer, T.R 1981. Calcium-induced alterations in cellular
morphology affecting the resistance of Lactobacillus acidophilus to fi'eezing. Appl.
Environ Microbiol. 41 : 807-815

Wuytack, E.Y., Phuong, L.D.T., Aertsen, A., Reyns, K.M.F., Marquenie, D., De
Ketelaere, B., Masschalck, B., Van Opstal, I., Diels, A.M.J. and Michiels, CW. 2003.
Comparison of sublethal injury induced in Salmonella enterica serovar Typhimurium by
heat and by different nontherrnal treatments. J. Food Protection. 66: 31-37

Xavier, U. and Ingham, S. 1993. Increased D-Values for Salmonella enteritidis resulting
from the use of anaerobic enumeration methods. Food Microbiol. 10: 223-228

152

 

 

Yamamoto, T., Hanawa, T. and Ogata, S. 1994. Induction of Yersinia enterocolitica
stress proteins by phagocytosis with macrophage. Microbiol. Immunol. 38: 295-300

Yousef, AB. and Courtney, PD. 2003. Basics of stress adaptation and implications in
new-generation foods. In Yousef and Juneja (ed.), Microbial Stress Adaptation and Food
Safety. CRC Press, Boca Raton, p. 1-30

Yu, L.S.L. and Fung, D.Y.C. 1991. Effect of OxyraseTM enzyme on Listeria
monocytogenes and other facultative anaerobes. J. Food Safety. 11: 163-175

Zemser, RB. and Martin, SE. 1998. Heat stability of virulence-associated enzymes fiom
Listeria monocytogenes SLCC 5764. J. Food Prot. 61: 899-902

Zhao, T. and Doyle, MP. 2001. Evaluation of universal preenrichment broth for growth
of heat-injured pathogens. J. Food Prot. 64: 1751-1755

Zook, C.D., Busta, F .F . and Brady, L.J. 2001. Sublethal sanitizer stress and adaptive
response of E. coli 0157:H7. J. Food Prot. 64: 767-769

Zugel, U. and Kaufmann, SH. 1999. Role of heat shock proteins in protection from
pathogenesis of infectious diseases. Clin. Microbiol. Rev. 12: 19-39

153