THESIS

’ﬁ,
a?

m

L.»

 

 

Iiiiii'riijlﬂijiiﬂiﬂiiiiiiiitir‘niiiil

3 2058 9978

This is to certify that the

dissertation entitled

R-Phycoerythrin as a Fluorescent Time-Temperature
Integrator for Monitoring Thermal Processing of
Beef Products

presented by

Alicia Orta-Ramirez

has been accepted towards fulﬁllment
of the requirements for

 

PhD degree in msmme

Mm m MM

Major professor

Date [2”q‘qﬁ

MSU i: an Afﬁrmative Action/Equal Opportunity Institution 0-12771

r‘:\\
MBRARY
Michigan State

University

  
  

 

 

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

 

DATE DUE I DATE DUE DATE DUE

 

rim/toiling

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 c/CIRCJDOaDuast-p.“

R-PHYCOERYTHRIN AS A FLUORESCENT TIME-TEMPERATURE INTEGRATOR
FOR MONITORING THERMAL PROCESSING OF BEEF PRODUCTS

By

Alicia Orta-Ramirez

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1999

ABSTRACT
R-PHYCOERYTHRIN AS A FLUORESCENT TIME-TEMPERATURE INTEGRATOR
FOR MONITORING THERMAL PROCESSING OF BEEF PRODUCTS
By

Alicia Orta-Ramirez

Time—temperature integrators (TTIS) are a rapid and accurate method to verify thermal
processing in foods because they can predict the lethality of a target attribute when the two
systems are subjected to the same thermal process. The objective of this study was to
characterize and verify the applicability of R-phycoerythrin (R-PE) to be used as the detector
component of a TTI for monitoring of thermal processing in beef products.

R-PE, a protein which ﬂuoresces in the visible range, was puriﬁed from algal tissues
of Porphyra yezoensis by precipitation with ammonium sulfate and gel ﬁltration
chromatography. The isolated R-PE had a purity greater than 5 as determined by Asos/Azso.
Isothermal experiments were conducted to determine the thermal inactivation parameters of
R-PE calculated on the basis of ﬂuorescence loss under different conditions of pH and
concentrations of sucrose, NaCl, sodium-dodecyl sulfate (SDS), urea and B-mercaptoethanol
(ME). R-PE was most thermostable at pH between 5.0 and 9.0, but became more heat
sensitive at pH 4.0 and 10.0. Sucrose and ME had a thermostabilizing effect, while SDS,
NaCl and urea decreased thermal stability of R-PE. The variations obtained in the thermal
inactivation parameters showed that the kinetics of R-PE can be modiﬁed by altering the
solution composition.

When R-PE was placed in borate buffer, pH 9.0, the resulting 2 value (599°C) fell

within the range of 2 values reported for Salmonella in ground beef (5.6-6.2°C). Salmonella
is the target microorganism in thermal processing of beef products.

In non-isothermal experiments, R-PE ﬂuorescence could distinguish between
adequate and inadequate heat processes according to USDA requirements for hamburger and
roast beef. The results suggested that thermal inactivation kinetics of R-PE do not follow a
ﬁrst-order reaction. A non-linear mathematical model was developed to ﬁt thermal
inactivation kinetics of R-PE and S. senftenberg. The reaction orders (n) calculated for R-PE
and S. senftenberg were 0.55 and 1.2, respectively. The activation energies (E A) were
estimated using the Arrhenius model. The calculated E A for R-PE and S. senftenberg were
347.4 and 220.84 1d mole“, respectively. The lethalities calculated for R-PE and S.
senftenberg assuming non-linear kinetics were highly predictable (R2 of 0.98 for R—PE and
0.93 for S. senftenberg). Results strongly suggest that, despite the differences in thermal

inactivation parameters, the lethality of Salmonella can be predicted from R-PE ﬂuorescence.

T o tears and laughter,
T o sunshine and rain.
T 0 family and friends,
T o Lluna, Pruna and Xaloc,
But most of all...

...To James

...F or believing in me.

iv

"Stop this day and night with me
and you shall possess the origin of all poems

You shall possess the good of the earth and sun,
(there are millions of suns left)

You shall no longer take things at second or third hand,
nor look through the eyes of the dead,
nor feed on the spectres in books
You shall not look through my eyes either, nor take things from me,

You shall listen to all sides and ﬁlter them from your self".

(Walt Whitman, Song of Myself)

ACKNOWLEDGMENTS

First and foremost, I would like to thank my advisor, Dr. Denise M. Smith for her
constant guidance and support throughout the program. I am most grateful for her patience
and understanding and her continued encouragement in my work and professional activities.
Words cannot express how much I have learned with and from her and this project would not
have been possible at all without her.

I wish to thank the members of my committee, Drs. John E. Merrill, Robert Y. Ofoli,
Gale M. Strasburg and Kris Berglund, for agreeing to participate in this project and for their
guidance and counseling during this work.

I would like to express my appreciation to Dr. James Steffe, Dr. Virginia
Vega-Warner and MS. Sarah Smith for their help, suggestions and opinions. I needed all the
help I could get and they were there when I needed them. Thanks also to Dr. Donald G.
Uzarski and Dr. Robert Templeman for the statistical advice.

I would like to thank two generations of "labmates" that made my work in the lab so
much more bearable, especially those long days and sleepless nights trying to make progress
in my project. Thank you to Dr. Stephanie Smith, Dr. Anne Smyth, Mr. Arnie Sair and Dr.
Ivy Hsu (lst generation), and to Ms. Carolyn Ross, Ms. Betsy Booren, Ms. Jennifer Maurer,
Mr. Jin-Shan Shie and Mr. Rob Slonaker (2nd generation). A special mention goes to Dr.
Manee Vittayanont who "survived" with me through both generations.

I feel most in debt to Mr. Matt Rarick for his invaluable time and effort put in this

vi

dissertation, as well as for his friendship and encouragement. I would like to acknowledge
Drs. Jose Candace Jackson, Jamie Sue Pratter, Arti Arora, Marivi Tejada, Dilum Dunuwila
and family; and Mrs. Becky Uzarski, Mr. Jay Chick, Mr. Eric Cole, Dr. Mike Miller, Mr.
Chris Sypien, Mr. Craig Banotai and many others because their friendship made my life at
MSU an unforgettable experience. I would also like to thank my ﬁ'iends at home, Drs. Alicia
Calvet, Jordi Torren, Quim Porcar, Rosa Ma. Annengol and Anna Ortuno; and Ms. Olga
Comas, Ms. Lali Ferret, Ms. Bea Baldo, MS. Montse Coma and Ms. Zoren Basualdo, as well
as the Orta-Calvet and Gines-Orta families, because life without them would not have been
the same. Thanks are also extended to Mrs. Shelli Pfeifer and Ms. Debi Bewsaert, and the
Department of Food Science and Human Nutrition at MSU.

Finally, this would have not been possible without the love and moral support of the
Orta-Ramirez family and, most of all, Dr. James Roger Clarke. It is because of them that I am
who I am and I got to this point, and I will never be able to repay them for that.

This work was supported by Michigan State University Crop and Food Bioprocessing
Center and the Cooperative State Research, Education and Extension Service, United States
Department of Agriculture under agreement number 96-35500-3551. Any opinions, ﬁndings,
conclusions or recommendations expressed in this document are those of the authors and do

not necessarily reﬂect the view of the US Department of Agriculture.

vii

TABLE OF CONTENTS

LIST OF TABLES .......................................................... x

LIST OF FIGURES ....................................................... xiii

CHAPTER 1: Introduction ................................................... 1

CHAPTER 2: Objectives .................................................... 3

CHAPTER 3: Signiﬁcance and justiﬁcation ..................................... 4

CHAPTER 4: Literature review ............................................... 6

4.1. THERMAL PROCESSING REGULATIONS ........................... 6

4.2. Salmonella ...................................................... 10

4.2.1. Characteristics of the microorganism. ......................... 10

4.2.2. Clinical presentations. ..................................... 12

4.2.3. Pathogenesis. ............................................ 13

4.2.4. Epidemiology. ........................................... 13

4.2.5. Prevention and control. .................................... 14

4.3. TIME-TEMPERATURE INTEGRATORS ............................ 16

4.3.1. In situ methods. .......................................... 17

4.3.2. Physical-mathematical methods. ............................. 17

4.3.3. Time-temperature integrators (TTIS). ......................... 18

4.4. PHYCOBILIPROTEIN S .......................................... 28

4.5. PRINCIPLES OF FLUORESCENCE SPECTROSCOPY ................ 31
CHAPTER 5: Effect of pH on the thermal inactivation of R-phycoerythrin from Porphyra

yezoensis .......................................................... 33

5.1. ABSTRACT .................................................... 33

5.2. INTRODUCTION ............................................... 33

5.3. MATERIALS & METHODS ....................................... 36

5.3.1. Source of R-phycoerythrin. ................................. 36

5.3.2. Puriﬁcation of R-phycoerythrin .............................. 36

5.3.2.1. Procedure for fresh algal tissue ....................... 36

5.3.2.2. Procedure for dry algal tissue ........................ 38

5.3.2.3. Purity and yield of extracted R-phycoerythrin ........... 39

5.3.3. Determination of molecular weight and isoelectric point. .......... 39

5.3.4. Absorbance and ﬂuorescence measurements. ................... 40

5.3.5. Effects Of pH on thermal inactivation of R-PE. .................. 40

5.3.6. Calculation of D and 2 values. ............................... 41

5.4. RESULTS & DISCUSSION ....................................... 42

5.4.1. Determination of molecular mass and p1. ...................... 42

5.4.2. Effect of pH on R-PE spectral properties. ...................... 43

5.4.3. Effect of pH on thermal inactivation of R-PE .................... 46

viii

5.5. CONCLUSIONS ................................................. 49
CHAPTER 6: Effects of sucrose, SDS, NaCl, urea and B-mercaptoethanol on the thermal

inactivation of R-phycoerythrin ......................................... 51

6.1. ABSTRACT .................................................... 51

6.2. INTRODUCTION ............................................... 51

6.3. MATERIALS AND METHODS .................................... 54

6.3.1. Materials. ............................................... 54

6.3.2. Solution conditions. ....................................... 55

6.3.3. Thermal inactivation. ...................................... 55

6.3.4. Fluorescence measurement. ................................. 56

6.3.5. Data analysis. ............................................ 56

6.4. RESULTS AND DISCUSSION ..................................... 57

6.4.1. Effects of sucrose. ........................................ 57

6.4.2. Effects of SDS ............................................ 64

6.4.3. Effects ofNaCl. .......................................... 68

6.4.4. Effects of urea. ........................................... 72

6.4.5. Effects of B-mercaptoethanol (ME). .......................... 79

6.4.6. Implications for food processing. ............................ 82

6.5. CONCLUSIONS ................................................. 84
CHAPTER 7: Validation of R-phycoerythrin as a time-temperature integrator to monitor

thermal processes in beef products ...................................... 85

7.1. ABSTRACT .................................................... 85

7.2. INTRODUCTION ............................................... 85

7.2.1. Theoretical considerations. ................................. 87

7.3. MATERIALS AND METHODS .................................... 89

7.3.1. Source and preparation of R-phycoerythrin. .................... 89

7.3.2. Isothermal experiments. .................................... 89

7.3.3. Non-isothermal experiments. ................................ 90

7.3.4. Fluorescence measurement. ................................. 91

7.3.5. Experimental heat inactivation data of S. senftenberg. ............ 91

7.3.6. Data analysis. ............................................ 93

7.4. RESULTS AND DISCUSSION ..................................... 94

7.4.1. Isothermal treatment. ...................................... 94

7.4.2. Non-isothermal treatment. .................................. 95

7.4.3. The nth-order model. ..................................... 100
7.4.4. Validation of parameter estimates for the nth-order model to predict

thermal inactivation of S. senftenberg. ...................... 104

7.5. CONCLUSIONS ................................................ 106

7.6. NOMENCLATURE AND UNITS .................................. 106

CHAPTER 8: Conclusions ................................................. 108

CHAPTER 9: Future research .............................................. 110

BIBLIOGRAPHY ........................................................ 112

ix

LIST OF TABLES

Table 4.1. Absorption and ﬂuorescence characteristics of phycobiliproteins. ........... 29
Table 5.1. D values for R-phycoerythrin at different pHs. .......................... 47

Table 5.2. Summary of 2 values and regression analysis for R-phycoerythn'n at different pHs.
.................................................................. 48

Table 6.1. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in sucrose solutions in 50 mM NaCl, 20 mM phosphate buffer, pH 7.0
.................................................................. 61

Table 6.2. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in sucrose solutions in 50 mM NaCl, 20 mM glycine buffer, pH 10.0
.................................................................. 61

Table 6.3. The 2 values and regression parameters for thermal inactivation curves of R-
phycoerythrin in sucrose solutions in 50 mM NaCl, 20 mM phosphate buffer, pH 7.0
and and 50 mM NaCl, 20 mM glycine buffer, 10.0 .......................... 63

Table 6.4. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in sodium-dodecyl sulfate (SDS) solutions in 50 mM NaCl, 20 mM
phosphate buffer, pH 7.0 .............................................. 63

Table 6.5. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in sodium-dodecyl sulfate (SDS) solutions in 50 mM NaCl, 20 mM
glycine buffer, pH 10.0 ............................................... 66

Table 6.6. The 2 values and regression parameters for thermal inactivation curves of R-
phycoerythrin in sodium-dodecyl sulfate (SDS) solutions in 50 mM NaCl, 20 mM
phosphate buffer, pH 7.0 and 50 mM NaCl, 20 mM glycine buffer, 10.0 ......... 66

Table 6.7. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in NaCl solutions in 20 mM phosphate buffer, pH 7.0 ........... 71

Table 6.8. D values and regression parameters for thermal inactivation curves of R-
phycoerythn'n in NaCl solutions in 20 mM glycine buffer, pH 10.0 ............. 71

Table 6.9. The 2 values and regression parameters for thermal inactivation curves of R-
phycoerythrin in NaCl solutions in 20 mM phosphate buffer, pH 7.0 and 50 mM
NaCl, 20 mM glycine buffer, pH 10.0 .................................... 73

Table 6.10. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in urea solutions in 50 mM NaCl, 20 mM citrate buffer, pH 4.0 . . . . 73

Table 6.11. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in urea solutions in 50 mM NaCl, 20 mM phosphate buffer, pH 7.0
.................................................................. 76

Table 6.12. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in urea solutions in 50 mM NaCl, 20 mM glycine buffer, pH 10.0 . . 76

Table 6.13. The 2 values and regression parameters for thermal inactivation curves of R-
phycoerythiin in urea solutions in 50 mM NaCl, 20 mM citrate buffer, 4.0, 50 mM
NaCl, 20 mM phosphate buffer, pH 7.0 and 50 mM NaCl, 20 mM glycine buffer, 10.0

.................................................................. 78
Table 6.14. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in B-mercaptoethanol (ME) solutions in 50 mM NaCl, 20 mM
phosphate buffer, pH 7.0 .............................................. 78

Table 6.15. D values and regression parameters for thermal inactivation curves of R-
phycoerythrin in B-mercaptoethanol (ME) solutions in 50 mM NaCl, 20 mM glycine
buffer, pH 10.0

 

.................................................................. 81
Table 6.16. The z values and regression parameters for thermal inactivation curves of R-
phycoerythrin in B-mercaptoethanol (ME) solutions in 50 mM NaCl, 20 mM
phosphate buffer, pH 7.0 and 50 mM NaCl, 20 mM glycine buffer, 10.0 ......... 81
Table 6.17. Effects of the holding temperature and 2 value (E A) of a time-temperature
integrator on the actual processing value reported (Source: Van Loey et al., 1995b)
.................................................................. 83
Table 7.1. Time-temperature schedules for non-isothermal heating of R-phycoerythrin in
0.012 M borate buffer, pH 9.0. ......................................... 92

Table 7.2. Summary of D and 2 values and regression analysis for R-phycoerythrin in 0.012
M borate buffer, pH 9.0. .............................................. 94

Table 7.3. Residual ﬂuorescence Of R-phycoerythrin after thermal processing under USDA
cooking regulations. .................................................. 97

Table 7.4. Estimation of reaction order (n) for R-phycoerythrin in 0.012 M borate buffer, pH

xi

9.0, using an nth—order model. ......................................... 101

Table 7.5. Estimation of reaction order (n) for Salmonella senftenberg in ground beef using
an nth-order model. ................................................. 101

Table 7.6. Estimation of reaction constant (k) for R-phycoerythrin in 0.012 M borate buffer,
pH 9.0, using an nth-order model. ...................................... 101

Table 7.7. Estimation of reaction constant (k) for Salmonella senftenberg using an nth-order
model ............................................................. 101

Table 7.8. Kinetic parameters (k0, EA and n) of R-phycoerythrin and Salmonella senftenberg.
................................................................. l 04

xii

LIST OF FIGURES

Figure 5.1. Flow diagram for the puriﬁcation Of R-phycoerythrin from P. yezoensis. All steps
were performed at 4°C. ............................................... 37

Figure 5.2. Isoelectric focusing of R-phycoerythrin from P. yezoensis in a 7.5 % acrylamide
gel covering the pH range 4.0 to 7.0 1) Isoelectric focusing standard, 2) R-
phycoerythrin ....................................................... 43

Figure 5.3. Absorption spectrum of R-phycoerythrin from P. yezoensis in SOmM NaCl, 20
mM sodium phosphate buffer, pH 7.0. ................................... 44

Figure 6.1. Thermal inactivation curves of phycoerythrin in sucrose solutions in 50 mM
NaCl, 20 mM phosphate or glycine buffer, pH 7.0 and 10.0, respectively. ....... 58

Figure 6.2. Thermal inactivation curves of phycoerythrin in 50 mM, 20 mM citrate buffer,
pH 4.0, 50 mM, 20 mM phosphate buffer, pH 7.0 and 50 mM, 20 mM glycine buffer,
pH 10.0. ........................................................... 59

Figure 6.3. Thermal inactivation curves of phycoerythrin in sodium dodecyl sulfate solutions
in 50 mM NaCl, 20 mM phosphate or glycine buffer, pH 7.0 and 10.0, respectively.
.................................................................. 65

Figure 6.4. Thermal inactivation curves of phycoerythrin in NaCl solutions in 50 mM NaCl,
20 mM phosphate or glycine buffer, pH 7.0 and 10.0, respectively. ............. 69

Figure 6.5. Thermal inactivation curves of phycoerythrin in urea solutions in 50 mM NaCl,
20 mM citrate, phosphate or glycine buffer, pH 4.0, 7.0 and 10.0, respectively. . . . 74

Figure 6.6. Thermal inactivation curves of phycoerythrin in B-mercaptoethanol solutions in
50 mM NaCl, 20 mM phosphate and glycine buffer, pH 7.0 and 10.0, respectively.
.................................................................. 80

Figure 7.1. Residual ﬂuorescence in arbitrary units (an) of R-phycoerythrin in 0.012 M
borate buffer, pH 9.0, after non-isothermal heating .......................... 96

Figure 7.2. Comparison of Observed residual ﬂuorescence and predicted ﬂuorescence as

calculated by the General Method of R-phycoerythrin in 0.012 M borate buffer, pH
9.0 aﬁer non-isothermal experiments. .................................... 99

xiii

Figure 7.3. Comparison of ﬂuorescence of R-phycoerythrin in 0.012 M borate buffer, pH
9.0, after non-isothermal heating, and experimental lethality of the thermal process.
.................................................................. 99

Figure 7.4. Thermal resistance curves of R-phycoerythrin in 0.012 M borate buffer, pH 9.0,
calculated using linear (n=1) and non-linear kinetics (n=0.55). ............... 103

Figure 7.5. Thermal resistance curves Of Salmonella senftenberg in ground beef calculated

using linear (n=1) and non-linear kinetics (n=1.2). ......................... 103
Figure 7.6. Arrhenius plot of R-phycoerythrin in 0.012 M borate buffer, pH 9.0. ....... 105
Figure 7.7. Arrhenius plot of Salmonella senftenberg in ground beef. ................ 105

xiv

CHAPTER 1: Introduction

Thermal processing is the most common method used for food preparation and
preservation. The thermal processing kinetics can be deﬁned in terms of D, z and F values
(Singh and Heldman, 1993; Hendrickx et al., 1995). D value is the time in minutes required
to decrease a quality attribute by 90% at a constant temperature. The 2 value is the
temperature increase necessary to reduce the D value by 90%. The F value is the time
required to achieve a stated reduction in a population of microorganisms or spores. This time
is usually expressed as a multiple of the D value. Thermal processes describing changes in a
quality attribute such as a microorganism or a protein can also be expressed in terms of
chemical kinetics using the Arrhenius equation

k = k0 exp(—E A/R T)
where k0 is the pre-exponential factor, E A is the activation energy of the process, R is the
universal gas constant, and T is the temperature. In this context, the D value or thermal
reduction time is equivalent to the reaction rate constant, k, and 2 value or thermal resistance
constant can be compared to the activation energy, E A (Singh and Heldman, 1993). Over a
certain range of temperature, both models (D-z and k- E A) can be applied with the same
accuracy (Hendrickx et al., 1995).

Even with reliable thermal processing schedules and equipment, though, there is a
need to determine process compliance to ensure food safety. The evaluation of thermal

processes in foods can be done using in situ methods, physical-mathematical models or

time-temperature integrators. Time-temperature integrators (TTIS) are devices capable of
predicting the time-temperature response of a quality or safety index in a food product. This
occurs when the thermal destruction of both TTI and target index are identical (zTTI = 2mg“)
(Hendrickx et al., 1995; Van Loey et al., 1996).

The United States Department of Agriculture Food Safety Inspection Service
(U SDA-F SIS) has recently changed the thermal processing regulations for meat products to
include a lethality standard based on destruction of the pathogenic microorganism Salmonella
expressed as log reductions. The USDA now requires that manufacturers use a combination
of thermal and non-thermal processes to achieve a 6.5-log Salmonella reduction in
ready-to-eat, cooked beef, roasted beef and cooked corned beef; a 7-log reduction in
ready-to-eat poultry products (Federal Register, 1999a). The USDA has proposed a 5-log
reduction in cooked uncured meat patties, although this may be increased (Federal Register,
1999a). The development of a TTI with the same 2 value as Salmonella in beef and poultry
products could be used to determine the adequacy of any log reduction process.

The main objective of this project was to study the thermal inactivation kinetics of the
ﬂuorescent protein R-phycoerythrin and assess its usefulness as a TTI for ensuring the proper
cooking of beef products. By manipulation of the thermal inactivation kinetics of
R-phycoerythrin, to adjust the 2 value of the protein to that of Salmonella in beef,
R-phycoerythrin could be used as the direct sensing component of a new type of TTI to

model Salmonella destruction in cooked beef products.

CHAPTER 2: Objectives

The overall goal of this research is to develop a time-temperature integrator (TTI) to
monitor thermal processing of foods with special emphasis in meat products. The main
component of this TTI is a naturally-ﬂuorescent pigment protein called R-phycoerythrin
(R-PE).

The speciﬁc objectives are:

1. To determine the thermal inactivation kinetics of R-PE as affected by pH and addition
of other chemicals (sucrose, sodium-dodecyl sulfate, NaCl, urea and
(ﬂ-mercaptoethanol);

2. To compare thermal inactivation kinetics of R-PE with experimental heat inactivation
data of Salmonella senftenberg in beef products;

3. TO verify the ability of R-PE to quantify the integrated time-temperature impact
during isothermal and non-isothermal heating proﬁles; and

4. To develop and verify a mathematical expression that ﬁts the thermal inactivation
kinetics of S. senftenberg in cooked beef products and of R-PE as measured by

ﬂuorescence loss.

CHAPTER 3: Significance and justification

Adequate thermal processing is the most important method to eliminate pathogenic
bacteria and viruses from meat and poultry products to prevent foodborne diseases.
According to Todd (1996), the estimated number of cases resulting from foodborne
infections and intoxications were 55-62 million costing $5.8-8.6 billion. Salmonella, E. coli
0157:H7, Campylobacter and Staphylococcus are among the most common causative agents
of foodborne illnesses in the US. Rapid and accurate methods to verify compliance of
thermal treatments are needed to ensure safety of cooked meat products. Current USDA
methodology for monitoring adequacy of heat treatments is based on determination of the
endpoint temperature to which the product has been cooked. Neither of the ofﬁcial tests use
the time-temperature integrator concept nor are based on thermal destruction of bacterial
pathogens (Doyle and Schoeni, 1984; Goodfellow and Brown, 1978; Line et al., 1991). In
addition, these methods are inaccurate, subjective and do not provide enough margin of
safety (Smith and Orta-Ramirez, 1995).

This project prOposes the development of a TTI for monitoring of thermally processed
beef products (for example, roast beef and hamburger patties). Manufacturers and consumers
will beneﬁt from improved food quality and safety through the increase in the frequency,
accuracy and convenience of thermal process testing. It is envisioned that these TTIs will be
devices with a very low cost per unit, ready to be incorporated into the thermal processing

procedure, easy to recover, and with a fast and user-friendly readout. Because of the high

convenience, they could be applied to routine monitoring at the retail level (i.e. fast food
restaurants, processing plants). More frequent routine testing would yield safer cooked beef
products while avoiding overcooking, which in turn results in lower quality foods and higher
energy consumption.

We will evaluate the applicability of a protein, R-phycoerythrin (R-PE), to be used as
the detecting component of the TTI. By determining the kinetics Of thermal inactivation of
R-PE and characterizing the factors or conditions that alter such parameters, we can evaluate
the modiﬁed inactivation kinetics to deﬁne the usefulness of the protein to predict destruction
of Salmonella in beef products.

Although the use of colorimetric-based or protein-based thermal monitoring tests is
not totally new, this TTI offers advantages over other alternatives that justify its research and
development:

0 R-phycoerythrin is a natural pigment-protein derived from an edible algae.

This is of interest because of its inherent safety in food applications.

0 The algae has abundant pigment (up to 20% of dry weight) which is easily
extracted and puriﬁed.

- Encapsulation in a deﬁned chemical environment allows known kinetics even
in foods of different composition.

0 When compared with enzymes, R-phycoerythrin shows a wider extent of
applications because of its tolerance over a broader range of chemical
conditions.

° The ﬂuorescence loss of R-phycoerythrin is directly and immediately

detectable, without the need for additional steps prior to analysis.

CHAPTER 4: Literature review

4.1. THERMAL PROCESSING REGULATIONS

Microbial contamination is considered the most important hazard associated with
foodborne illnesses (Todd, 1996; Centers for Disease Control and Prevention, 1998a). From
1988 to 1992, a total of 2,423 foodborne outbreaks were recorded, of which, 79% were
caused by bacterial pathogens (Centers for Disease Control and Prevention, 1998a).
Inadequate cooking and improper storage and holding temperatures are the most common
errors that lead to foodborne outbreaks (Todd, 1996; Centers for Disease Control and
Prevention, 1998a). Recent outbreaks of food poisoning due to Escherichia coli 0157:H7
associated with consumption of underprocessed beef products, have led to a severe scrutiny
of the methodology used to ensure safety Of cooked meat products. Traditionally, Title 9 of
the Code of Federal regulations outlines the requirements for the processing of meat products
needed for destruction of pathogenic microorganisms that can cause foodborne disease
(U SDA-FSIS, 1990). The government regulatory agencies reviewed and updated the current
thermal processing regulations after the 1993 outbreak of E. coli 01572H7. This outbreak
was associated with consumption Of undercooked ground beef patties and involved more than
500 people, including four deaths.

In January 1994, the Food and Drug Administration (FDA) published a new Food
Code (FDA, 1993) establishing new recommendations for cooking meat products. It was

recommended that pork products and comminuted meats be cooked to internal temperatures

of either 63°C (145°F) for 3 min, 66°C (150°F) for 1 min or 68°C (155°F) for 15 sec. Roast
beef and corned beef were recommended to be cooked to any of the following
time/temperature protocols: 54°C (130°F )/ 121 min, 56°C (132°F)/77 min, 57°C (134°F)/47
min, 58°C (136°F)/32 min, 59°C (138°F)/19 min, 60°C (140°F)/12 min, 61°C (142°F)/8 min,
62°C (144°F)/5 min or 63°C (145°F)/3 min. Poultry products were recommended to be
cooked to 74°C (165°F) or above for 15 sec. In addition, according to USDA-FSIS
regulations, ofﬁcial establishments that manufacture fully-cooked patties were required to
use one of the following time/temperature protocols: 66.1°C (151°F)/41 sec, 66.7°C
(152°F)/32 sec, 67.2°C (153°F)/26 sec, 67.8°C (154°F)/20 sec, 683°C (155°F)/ 16 sec,
689°C (156°F)/ 13 sec, and 694°C (157°F) (and up)/ 10 sec. Also, ofﬁcial establishments that
manufacture partially cooked patties were required to heat patties to a minimum internal
temperature 60°C (140°F) followed by cooling to a maximum internal temperature of 4°C
(40°F) within 2 h (USDA-F SIS, 1993). These schedules were based on destruction Of
pathogenic microorganisms (FDA, 1993; Goodfellow and Brown, 1978).

In 1996, the FSIS proposed to amend the Federal meat and poultry inspection
regulations (Federal Register, 1996). The proposed modiﬁcations deﬁned the performance
standards that establishments should meet to produce safe products while allowing each
processor to use their own speciﬁc plant-processing procedures, even if these differed from
the ones speciﬁed in the current regulations. Three performance standards were proposed by
F SIS: lethality, stabilization and handling. The purpose of the lethality standard was to
effectively eliminate pathogenic microorganisms that could be present in the product. The
reduction of a microbial population can be expressed in terms of decimal reduction time or D

value. The D value is deﬁned as the time necessary to achieve a 90% reduction in the

microbial population at a given temperature (Singh and Heldman, 1993).

Traditionally, Salmonella was the microorganism of concern in cooked beef products
(Goodfellow and Brown, 197 8). Although E. coli 01572H7 has become of greater concern,
especially in meat patties, it is less heat resistant than Salmonella (Doyle and Schoeni, 1984;
Line et al., 1991; Orta—Ramirez, 1994). In addition, although Listeria monocytogenes is more
heat resistant than Salmonella, its incidence in meat products is usually a result of
recontamination. Thus, expected levels of L. manocytogenes are much lower than those Of
Salmonella. Therefore, the thermal inactivation of Salmonella in meat products is indicative
of destruction of other pathogenic microorganisms. For cooked beef, roast beef, cooked
corned beef and cooked poultry products, FSIS recommended a lethality resulting in a 7-D
reduction in Salmonella; for fully cooked uncured meat patties the lethality standard should
be a process resulting in a 5-D Salmonella reduction. It is important to notice that all
time/temperature combinations previously required by USDA for cooked/roast beef, cooked
corned beef, poultry products and fully cooked uncured patties, would result in a 7-D or 5-D
Salmonella reduction, respectively (Federal Register, 1996). Therefore, establishments
already employing any of the above mentioned time-temperature schedules would produce
cooked beef or poultry products that comply with the proposed lethality standards.

On January 6, 1999, the F SIS formalized the lethality standards (Federal Register,
1999a). It was required that manufacturers use a combination of thermal and non-thermal
processes that achieve a 6.5-log Salmonella reduction in ready-tO-eat, cooked beef, roast beef
and cooked corned beef; and a 7-log reduction in ready-to-eat poultry products. Currently,
FSIS has not issued a ﬁnal rule on lethality standards for ready-to-eat meat patties, and, in the

meantime, current thermal processing requirements (Federal Register, 1996) will remain in

effect (Federal Register, 1999a).

Although an increase of thermal processing enforcement can help reduce the
incidence of foodborne disease, there is still the problem of conﬁrming adequate cooking
once the product has been processed. The USDA-F SIS currently employs three different
assays to verify proper thermal processing of meat products. The residual Acid Phosphatase
test (USDA-FSIS, 1986a) is used for canned hams, picnics and luncheon meats. This assay is
based on the residual enzyme activity in water-soluble protein extracts of cooked samples.
Positive reactions show a blue color that can be read at 610 nm. One problem with this assay
is that it has been demonstrated that there is a large decrease in acid phosphatase activity
during frozen storage (Townsend, 1989). This should be taken into consideration when
analyzing samples that have been kept under freezing conditions.

The Coagulation test (USDA-F SIS, 1986b) is used for beef and pork products. The
method is based on loss of protein solubility during heat processing of the meat product.
Protein extracts of cooked samples are heated until cloudiness appears. The temperature at
which the turbidity occurs is considered the internal temperature the product reached during
cooking. For products cooked to the 63-71°C the assay can differ 8-10°C from the real
temperature achieved during processing (Townsend and Blankenship, 1989). This test is thus
very subjective and only gives a gross estimation of the thermal process applied.

The Bovine Catalase test (USDA—FSIS, 1989) is used to verify roast beef and cooked
beef. The assay is based on the production of foam when catalase reacts with oxygen.
Bovine catalase is destroyed in the 60.5 - 61 .1°C temperature range. Thus, at temperatures
above this range this assay becomes useless. In addition, the visual detection of foam is very

subjective.

None of these assays are sensitive and reliable. In addition, the thermal inactivation
kinetics Of these markers is not known. Moreover, with the introduction of new processing
techniques, it would be desirable to identify an indicator whose inactivation depends on
time—temperature only (i.e. TTIs) and not on the technique applied. There is urgent need for
objective and accurate assays to verify adequate thermal processing.

4.2. Salmonella

Salmonella is a microorganism commonly found in the intestinal tract of humans and
animals. Although most Salmonella are non-host speciﬁc, some strains are known to be
host-adapted. For example, S. typhi and S. paratyphi A and C are most typically found in
humans; S. pullorum and S. gallinarum are associated with poultry; S. typhisuis and S.
cholerasuis are found in pigs; S. dublin, S. abortusequi and S. abortusovis are associated with
cattle, horses and sheep, respectively (Doyle and Cliver, 1990; Bauer and Hordmansdorfer,
1996).

More than 2000 strains have already been identiﬁed. Of those, approximately 150 are
associated with disease outbreaks. Although Salmonella can be transmitted ﬁom person to
person, it is believed that the major source is contaminated foods of animal origin.

4.2.1. Characteristics of the microorganism.

Microorganisms of the genus Salmonella are part of the Enterobacteriacea family.
They are Gram negative, facultative anaerobic, non-spore forming bacilli, generally showing
motility. Most strains have shown the presence of ﬂagellae or ﬁmbriae.

There are several biochemical characteristics shared by the members of the genus
Salmonella. They use citrate as the only source of carbon and rarely ferment lactose or

sucrose. Most of them ferment glucose with the subsequent formation of gas.

10

The Salmonellae have been classiﬁed based on O or somatic antigen, H or ﬂagellar
antigen and Vi or capsular antigen, according to the Kauffman—White scheme. In addition,
they can be divided into ﬁve subgenera (I to V) based on their biochemical characteristics.

Salmonella can grow at temperatures ranging from 5 to 47°C, with an optimum
growth temperature of 37°C. They also can tolerate pHs between 4.0 and 9.0, with an
optimum pH range of 6.5-7.5 (Doyle and Cliver, 1990; Bauer and Hormansdorfer, 1996).
Salmonella are susceptible to freezing and thawing. A reduction of 1-2 log can be achieved
with one freeze-thaw cycle (Doyle and Cliver, 1990).

One method to eliminate Salmonella from foods is by cooking (CDC, 1998b). This
microorganism is mildly resistant to high temperatures. Several factors inﬂuence the
resistance of Salmonella to heat. Some strains are more resistant than others. For example, S.
senftenberg is unusually heat-resistant. It is considered to be 10-20 times more resistant than
the average Salmonella spp (Doyle and Cliver, 1990). Another factor is the composition of
the food or medium the Salmonella is in. For example, the D value at 60°C of S. typhimurium
in whole eggs was 0.27 min. When 10% sucrose was added to these eggs, the D60 value
increased to 0.6 min. The water activity and pH of the food or medium also plays an
important role in the heat resistance of Salmonella. The microorganism is much more
resistant to dry than moist heat and shows more heat susceptibility at extreme pHs (Schuman
and Sheldon, 1997).

The thermal inactivation of Salmonella in foods has been the subject of numerous
studies. Schuman and Sheldon (1997) studied the heat destruction of Salmonella in egg
products. D values ranged from 0.087 min at 622°C to 0.28 min at 60°C in egg yolk, and

from 1.0 min at 583°C to 7.99 min at 551°C in egg white. The 2 values ranged from 3.54 to

11

4.33°C. Goodfellow and Brown (1978) calculated D values for Salmonella spp in ground
beef. The D values determined at 51.6, 57.2, and 627°C were 61-62, 3.8-4.2, and 0.6-0.7
min, respectively, depending on the recovery method used. The 2 value was found to be
56°C. The results of Goodfellow and Brown were used to design the time-temperature
combinations for the cooking of roast beef and hamburger products listed in the USDA
thermal processing regulations (U SDA-F SIS, 1990).
4.2.2. Clinical presentations.

Salmonellosis is the generic disease caused by bacteria of the genus Salmonella.
There are three different syndromes associated with this microorganism. The most severe is
typhoid fever, which is caused by S. typhi. This strain is host-adapted to humans, therefore
transmission is due to fecal contamination of water and food. Symptoms include septicemia,
high fever, headache, constipation, vomiting and diarrhea (Doyle and Cliver, 1990). Typhoid
fever was cause for concern from late 1800s until 1950's in the US and Europe. Due to
improvement in sewage disposal, water treatment and implementation of public health
measures, this syndrome has become very rare and only occasional outbreaks are seen mostly
in developing countries (Tauxe, 1991).

Another type of syndrome associated with Salmonella is enteric fever. Symptoms are
similar to those of typhoid fever, but less severe. The enteric fever is caused by
S. paratyphi A, B and C (Doyle and Cliver, 1990).

The most common type of salmonellosis is a gastroenteritis syndrome or food
poisoning, caused by all the other types of Salmonella. Symptoms include diarrhea,
abdominal pain, fever, vomiting, dehydration, chills and headache. The incubation period

ranges from 5 to 72 h and the length of the illness is usually from 1 to 4 days (Doyle and

12

Cliver, 1990; Tauxe, 1991; Eley, 1992; Centers for Disease Control and Prevention, 1998b).
4.2.3. Pathogenesis.

The mode of transmission of Salmonella is mostly by ingestion of food contaminated
with the microorganism. The most common foods implicated in outbreaks are meat products
(especially poultry), eggs, and non-processed milk and dairy products (Eley, 1992; Centers
for Disease Control and Prevention, 1998b). After ingestion of contaminated foods, the
microorganisms attach to the epithelial cells of the villi in the terminal small intestine and
colon. Once attached, they can penetrate the intestinal mucosa and invade the lamina propia
and ultimately the lymphatic system. At the lamina propia the microorganisms multiply
causing an inﬂammatory response. In cases of food poisoning, the infection remains localized
in the intestinal area, but in cases of typhoid or enteric fever the bacteria progress further
producing a systemic infection (Doyle and Cliver, 1990; Eley, 1992; Bauer and
Hormansdorfer, 1996). In rare cases, Salmonella spp may cause urinary tract infections and
septicemia that can lead to meningitis and osteomyelitis (Eley, 1992).

4.2.4. Epidemiology.

Salmonella was ﬁrst discovered in 1885 by US Surgeon Daniel E. Salmon (Centers
for Disease Control and Prevention, 1998b). Currently, over 2300 Salmonella strains have
been identiﬁed. All these strains are pathogenic to human and/or animals, with the degree of
virulence depending on strain and susceptibility of the individual infected. Virulence factors
associated with Salmonella are formation of endotoxins, enterotoxins, cytotoxins,
lipopolysaccharides and siderophores, and presence of ﬂagellae and ﬁmbriae (Gast and
Beard, 1993; Bauer and Hordmansdorfer, 1996; Centers for Disease Control and Prevention,

1998b)

13

Since the 1960s Salmonella food poisoning has been on the rise. According to the
Centers for Disease Control and Prevention (Centers for Disease Control and Prevention,
1998b), there are 40,000 cases of Salmonella reported every year. Todd (1996) calculated
40,000 annual cases in the US and 9,000 in Canada. It is also estimated that for each case
reported there may be 30-100 cases that go unreported (Tauxe, 1991; Centers for Disease
Control and Prevention, 1998b).

Susceptibility to Salmonella varies within different groups of the population. Young
individuals are generally more susceptible than older ones. Infants of less than one month to
six months old Show the highest rate of infection, and about 40% of salmonellosis cases
occur in children less than 5 years of age (Doyle and Cliver, 1990; Centers for Disease
Control and Prevention, 1998b). The incidence of Salmonella also increases in individuals
over 60 years old and in other immunocompromised groups of the population, especially
patients with AIDS. This latter group shows a 20 fold increase in gastroenteritis and
septicemia when compared with normal individuals (Tauxe, 1991; Eley, 1992). In addition,
there seems to be a seasonal variation in the incidence of salmonellosis, with a maximum
peak during summer. This could be attributed to an increase in temperature abuse of foods
during these months (Centers for Disease Control and Prevention, 1998b).

4.2.5. Prevention and control.

An extensive review on methods for the detection of Salmonella in foods has been
published by Blackburn (1993). These include culture techniques, measurements of
metabolism (conductance and radiometry), immunoassays and the use of bacteriophages and
gene probes. General measures for prevention of salmonellosis include thorough cooking of

meats and pasteurization of milk and egg products, avoiding cross-contamination of foods, as

14

well as meticulous cleaning of kitchen utensils and personal hygiene. It is also important to
promote education of food handlers and consumers, and improve hygiene at the farm and
slaughter level to prevent contamination (Eley, 1992; Bauer and Hordmansdorfer, 1996;
Centers for Disease Control and Prevention, 1998b).

A more novel approach is the use of irradiation to control foodborne pathogens,
including Salmonella, and to extend shelf-life of food products. In 1992, the USDA approved
the use of ionizing radiation for fresh or frozen, uncooked, packaged poultry carcasses and
mechanically separated poultry products. The approved dose was a minimum of 1.5 kGy and
a maximum 3.0 kGy (USDA-FSIS, 1992). In December 3, 1997, the Food and Drug
Administration (FDA) approved irradiation for refrigerated or frozen uncooked meat, meat
by-products and certain meat products (Federal Register, 1997). The maximum dose allowed
was 4.5 kGy for refrigerated products and 7.0 kGy for frozen products. More recently, F SIS
has proposed to amend the FDA ﬁnal rule in meat and poultry products to make irradiation of
poultry more consistent with meat regulations (Federal Register, 1999b). FSIS is proposing
to approve irradiation of raw poultry products before packaging, as well as chicken carcasses.
It also proposes to remove the minimum dose allowed. The use of irradiation in meat and
poultry products can be a safe and effective measure to eliminate Salmonella from the food
chain when used either as the only method or as a combination with other treatment such as
heating or curing (Centers for Disease Control and Prevention, 1998b).

According to Todd (1996), an effective surveillance system at the national level is
essential to understand and control foodbome diseases. In the US, government agencies
routinely monitor the incidence of salmonellosis. The CDC collaborates with local and State

Health Departments to research outbreaks and create control strategies. The FDA inspects

15

and regulates food products and processing plants. It also promotes the use of safer
techniques to handle foods at the retail level. The FDA also regulates the use of antibiotics in
animal feeds. The USDA supervises health of animal stocks, egg processing and monitors the
quality and safety of slaughtered and processed meat. The US Environmental Protection
Agency regulates the safety of the water supply (Centers for Disease Control and Prevention,
1998b)

In Europe, several approaches have been attempted to eliminate Salmonella from
poultry and animal feeds. However, even with a very comprehensive program they still have
an incidence of 3-4%, and reduction of Salmonella results in a very expensive product (four
to ﬁve times the price of poultry in the US) (Doyle and Cliver, 1990; Cast and Beard, 1993).
Chander et a1. (1997) prepared a vaccine with irradiated S. typhimurium and S. gallinarum.
lmmunized chicks developed an immunoresponse and no Salmonella was isolated after 35
days, in comparison with control chicks that showed Salmonella contamination. The authors
concluded that the use of vaccines may be used to raise Salmonella-free ﬂocks. However, this
vaccine was not tested under ﬁeld conditions. In addition, the vaccine was administered
subcutaneously. This mode of application would not be feasible in routine use. Current
investigation includes testing of the vaccine under ﬁeld conditions and development of a
more appropriate method of administration.

4.3. TIME-TEMPERATURE INTEGRATORS

Thermal processing is a commonly used method for the preservation of foods
throughout the world. The conditions of a thermal process are typically determined by the
food product (composition, properties, consumer acceptability, and destination) and the

thermal treatment applied. However, even with modern and sophisticated technology there is

16

a need to verify that the proper conditions have been achieved for each process. Methods to
evaluate the accuracy of thermal processes are classiﬁed according to three different
approaches: in situ methods, physical-mathematical methods, and time-temperature
integrators (Hendrickx et al., 1995; Van Loey et al., 1996).

4.3.1. In situ methods.

In the in situ method, a safety or quality parameter is chosen and its concentration is
measured before and after the treatment. Such parameters can be sensory factors, nutritional
quality attributes or microbial contaminants. In practice, however, the determination of these
parameters, especially after the treatment, can be very difﬁcult, even impossible in some
cases, because of the detection limits, analytical techniques and/or sampling requirements.
Moreover, these type of methods are often unsuitable for routine monitoring due to their
complexity (Hendrickx, 1995).

4.3.2. Physical-mathematical methods.

In the physical-mathematical approach, the impact of the heat treatment on the
attribute of interest is estimated based on the time-temperature proﬁle and the known kinetic
parameters for such attribute. Limitations of these methods are due to the need for
time—temperature data. In some cases, direct recording is not possible due to processing
conditions; on the other hand, if the time-temperature history is reconstructed by
mathematical modeling, the accuracy of the estimation by model parameters will limit the
accuracy of the process evaluation. Conservative estimates used in these models oﬁen result
in unrealistically severe heat treatment requirements which in turn lead to overprocessing,
signiﬁcant nutrient loss and reduction of quality of the product. In addition, with the

introduction Of new techniques such as ohmic and microwave heating, and the lack of data

17

for kinetic models for these techniques, the use of physical-mathematical methods becomes
very limited. Sastry (1986) developed a mathematical model for aseptic processing of foods
with particulates treated in a scraped surface heat exchanger. The author studied the inﬂuence
of particle size, residence time distributions and estimated values of convective coefﬁcients.
The model was tested under different conditions but because of the lack of experimental data
it could not be validated.

4.3.3. Time-temperature integrators (TTIS).

Although the concept of TTIS is not new, they were originally intended to monitor
and control shelf life of food products (Taoukis et al., 1991; Singh and Wells, 1985). More
recently TTIs have been introduced as an alternative for the monitoring of thermal processes.
A TTI is deﬁned as a small device that responds to time-temperature history by undergoing
an irreversible and precisely measurable change in a manner that mimics the changes of a
target attibute exposed to the same thermal history (Hendrickx et al., 1995). An important
advantage of TTIS is that they allow fast and reliable validation of a process without the need
for detailed information on the actual time-temperature proﬁle within the product.

A reliable TTI should meet the following characteristics:

0 simple and inexpensive in preparation

0 easy to recover

0 give an accurate and easy detection response

- readily incorporated into the food product without interfering with the heat
transfer

- must quantify the process impact on the target attribute

To accomplish the last statement, a TTI must Show the same time-temperature

18

dependent response as that of the target attribute when the temperature is the only
rate-determining factor. Mathematically, this can be written as:

FTargct = Fm (1)
Assuming an nth order TTI system, the rate equation can be expressed as:

dC/dt = - k C' (2)
where k is the reaction constant and n the reaction order. For a ﬁrst order reaction under
isothermal conditions, the integrated equation is:

In (Co/C) = kt (3)
where C0 is the initial value of the target attribute and C is the value of a target attribute at a
time t. For a non-ﬁrst order reaction (n’l), the integrated expression can be written as:

[Hm-1)] (C"" - C0”) = kt (4)

In many cases, according to the Arrhenius equation the rate constant can be expressed as:

k = k0 exp (-EA /R7) (5)
where k0 is the pre-exponential factor, E A is the activation energy, R is the universal gas
constant and T is the absolute temperature.

In the general case where the reaction order is different than unity, with the reaction

rate constant and independent of the order of reaction, the F value can be written as

(Hendrickx et al., 1995):

(F ) —jex -E—"( 1 inn
W “0 p R Trc/ T (6)

 

which represents the equivalent heating time at a reference temperature resulting in the same
lethality as the time-varying temperature proﬁle. This equation is valid for the target attribute

as well, and therefore only when the activation energies of both TTI and target attribute are

19

the same, equation (1) will be satisﬁed. In other words, for a system to be used as TTI it must
have identical activation energy to that Of the attribute of interest. Mathematically this can be
written as:

EA m = EATurgcl (7)

Since 2 values are equivalent to activation energies, equation (7) can also be
expressed as:

277/ = Z Target (8)

TTIs can be classiﬁed according to their working principle (biological, chemical or
physical), type of response (single or multi), origin (intrinsic or extrinsic to the food product),
application in the food material (dispersed, permeable or isolated) and location in the food
(volume average or single point) (Van Loey et al., 1996). Most currently existing TTIS are
based on microbiological or enzymatic assays. Microbiological TTIS mainly employ
"calibrated" (i.e. they have been validated against a known standard) microorganisms or
spores. A thorough review on bacteriological evaluation for thermal process design has been
published by Yawger (1978).

Microbial spores can be used to monitor thermal processing using either a survivor
curve or an endpoint method. In a survivor curve or "count reduction procedure", the number
of microorganisms surviving the heat treatment are related through a calibration curve to
obtain a sterilizing value. Usually, ten or more cans are inoculated with 30-50 million spores
of a heat-resistant microorganism. A series of thermal treatments of varying time or
temperature are applied. The initial population is calculated from a nonprocessed container
and survivor counts are obtained from each processed can. The D value for the

microorganism is calculated for the series of treatments tested in the assay (Yawger, 1978;

20

Pﬂug et al., 1980).

In the endpoint procedure, also called quantal response, several units are subjected to
each treatment. After incubation, each unit is tested for either growth or no growth, and the
number of survivors is determined from the quantal response. An example of endpoint
method is the inoculated pack system (Yawger, 1978). This system consists of a series of
cans (usually 100) which are inoculated with a deﬁnite number of spores of known
resistance. This method is generally used to validate a calculated food sterilization treatment.

The main disadvantages of microbiological TTIS are assay time, because of the
microorganism incubation period, and the likelihood of contamination, as well as cost, labor
and the need to properly calibrate the Spores. In addition, if the initial population of
microorganisms is completely destroyed during processing, the actual lethality derived from
the heat process cannot be determined. Pﬂug et a1. (1980) studied the performance of Bacillus
stearothermophilus spores in measuring the sterilization process in cans of food heated in a
Steritort process simulator. The spores, packed in a canier system (plastic rods called
biological indicator units or BIUS) and previously calibrated at different temperatures, were
placed inside the food cans. After processing, the BIUS were assayed for surviving spores.
Cans containing thermocouples were processed in the same fashion. F values were calculated
from both the BIUS and thennocouple-equipped cans and compared. The authors found a
good agreement between the F values calculated with both methods. However, because the 2
value of the spores employed in this study (7.8°C) differed from the target 2 value (10°C), a
correction method had to be applied to make both P values comparable. Furthermore, in F
values calculated with the spores, an additional correction had to be made to account for

heating and cooling lags of the BIUS due to differences in temperature across the BIU wall.

21

The authors concluded that the accuracy of the method, even with proper calibration, was
within 15% of the true F value. Sastry et al. (1988) developed a B. stearothermophilus
spores-based bioindicator for veriﬁcation of thermal processes in particulate foods.
Individual mushrooms were infused with alginate gels containing a spore suspension, and
processed. Although the authors concluded that the bioindicator was potentially useful as an
indicator Of sterility in continuous aseptic systems, the main problem was that they could not
provide conclusive proof of negligible leakage of the spores into the product during the
thermal process.

Different enzymes, mainly from microbial origin, have also been proposed as TTIS.
Weng et al. (1991a) investigated the thermal behavior of peroxidase that had been modiﬁed
chemically and physically, and studied its application as a bioindicator for thermal
processing. The 2 value Of peroxidase was modiﬁed from 263°C to 141°C by
immobilization of the enzyme in glass beads. Furthermore, the 2 value was lowered to 11.1°C
by modiﬁcation of the environment with organic solvents. In a different study, Weng et al.
(1991b) veriﬁed the use of the immobilized peroxidase in a dodecane environment as an
indicator for pasteurization processes. The peroxidase system had a 2 value of 101°C. When
evaluated at processing temperatures between 70 and 80°C, the lethalities Of the indicator
agreed with those calculated according to the Bigelow’s General Method. In a related study,
Hendrickx et al. (1992a) studied the thermal denaturation of peroxidase as a ﬁinction of water
activity. Lyophilized horseradish peroxidase was equilibrated over standard salt solutions to
achieve a water activity between 0.11 and 0.88. The enzyme was much more thermostable in
the dry state than in aqueous solution, and both D and 2 values changed with water activity.

Inactivation temperatures were in the range 140-160°C at low water activities (compared to

22

70-85°C in aqueous solution). Later, Hendrickx et al. (1992b) studied the application of
immobilized peroxidase as an indicator in a can model system under pasteurization
conditions. Plastic spheres containing peroxidase were placed in a can and processed at 85°C.
The lethalities calculated by integration of the resulting time-temperature proﬁles coincided
very well with those read from the peroxidase system. The authors suggested that the
validated bioindicator could be encapsulated in a small vial and used to calculate lethality in
the center of the unit.

DeCordt et al. (1992a) studied the application of the enzyme (It-amylase as a TTI in
thermal processes. The group studied the thermal inactivation kinetics of oc-amylase from
Bacillus licheniformis in the temperature range 90-108°C. They also looked at the inﬂuence
of immobilization, pH, ionic strength, Ca2+, and concentration of enzyme as effective means
to manipulate the thermal stability of the enzyme. The D values at 95°C differed between free
and covalently immobilized enzyme. Extrinsic Ca2+ conferred stability, but there was a
saturation level after which the enzyme was destabilized. The optimum pH to lower thermal
inactivation rates was found at pH of 8.5. Also, increasing the concentration of the enzyme
resulted in higher therrnoresistance. The authors concluded that with manipulation of the
environment, the immobilized enzyme showed potential to be used as a TTI for monitoring
thermal inactivation of Clostridium botulinum spores in food.

In a related paper, DeCordt et al. (1992b) tested biphasic and nth-order models to ﬁt
experimental inactivation data of B. licheniformis (It-amylase immobilized on glass beads.
Both isothermal and non-isothermal experiments were used to estimate model parameters
(E A, k and n) using a non-linear regression procedure. The authors calculated an EA of 293 kJ

mole'l for the system, which fell within the range of that of C. botulinum (265-340 kJ mole").

23

They concluded that the system had potential to be used as a TTI in thermal processes in the
temperature range of 96-108°C.

The general effects of polyols and carbohydrates on thermal denaturation kinetics of
OL-amylase were also studied (DeCordt et al., 1993). The results showed that all polyols
(glycerol, mannitol and sorbitol) and carbohydrates (starch and sucrose) tested were powerful
thermostabilizers, at least at the temperature range used in the study.

DeCordt et al. (1994) studied the thermostability of (It-amylase and peroxidase using
differential scanning calorimetry (DSC). DSC peak temperature was used as a measure of
protein stability. The thermal inactivation kinetics of a-amylase from two Bacillus spp and
horseradish peroxidase were determined as a function of the concentration of glycerol,
sorbitol and sucrose. B. amyloliquefaciens (It-amylase was very stable in presence of either
polyols and carbohydrate or combination of both, and inactivation temperatures were as high
as 127°C. The authors suggested that the use of DSC peak area could be used as a
TTI-response.

Maesmans et al. (1994a) conducted a theoretical study of the possibilities of
combined mathematical model and TTI to monitor ﬂuid-to—particle convective heat transfer
coefﬁcient (CHTC). The authors concluded that although, theoretically, the combined use of
both could be indeed suitable to determine the CHTC, in practice, however, there is a need to
very carefully examine experimental design considerations. These considerations are speciﬁc
for each processing condition, TTI employed and type of material, therefore making the
combined use of a mathematical model and TTI a rather cumbersome methodology. Also,
Maesmans et al. (1994b) evaluated the efﬁcacy of an OL—amylase-based TTI to monitor the

spatial distribution Of thermal processing values in a food model system using a pilot retort.

24

The TTI in this particular case was used to determine particle-to-particle variation in
processing values. One question that was raised was whether the use of two TTIS,
characterized by different 2 values, could determine the "coldest point" at the same position
when applied in the same heat process. The authors suggested that results from a TTI should
not be extrapolated to another TTI with a different 2 value without experimental veriﬁcation.
Furthermore, the authors concluded that for an enzyme system to be used as a TTI for the
monitoring of heat distribution in foods, it is critical to calibrate the system in terms of the
kinetic inactivation parameters (D-z or k-EA values).

Van Loey et al. (1997a) tested the efﬁcacy Of a B. amyloliquefaciens
OL-amylase-based TTI to evaluate in-pack lethalities in a pasteurized food model system.
Thermal inactivation of the enzyme was measured as a function of reaction enthalpy using
DSC. The reaction enthalpy followed a log-linear reduction at constant temperature.
Therefore, a ﬁrst order reaction was assumed and a 2 value of 76°C was obtained. When
tested in a heat processed food model system, the processing values determined from the TTI
read-out coincided very well with the actual integrated processing values calculated using the
general method, suggesting that this enzyme could be used to monitor pasteurization
treatments (up to 90°C). Van Loey et al. ( 1997b) studied the heat inactivation kinetics of B.
subtilis a-amylase under steady and non-steady conditions by following enthalpy changes
associated with the thermal denaturation of the enzyme. When equilibrated at a water activity
of 0.76, the enzyme had a 2 value of 9.7°C. Because of the similarity to the 2 value (10°C) of
C. botulinum, the authors suggested that this enzyme system could be used to monitor the
efﬁcacy of sterilization treatments in foods.

Only very few methods based on either chemical or physical indexes are currently

25

available. The ﬁrst chemical markers that were studied were compounds in foods for which
analytical methods were already established, such as thiamin, pantothenic acid, vitamin C
and methylmethionine sulfonium salt (Kim and Taub, 1993). The limitations of most of these
methods is the need for additional steps (recovery, post treatment assay) after the thermal
process which results in lengthy procedures and fairly technical manipulations (Van Loey et
al., 1996). Kinetically, other challenges have to be faced when dealing with chemical TTIS;
ﬁrst, the rate constant of a chemical reaction is usually smaller than that of microbial
destruction; secondly, the 2 values are usually greater (the 2 value for thiamin is 48°C while
the one for C. botulinum is 10°C) (Mulley et al., 1975).

According to Kim and Taub (1993), compounds formed during food reactions show
more potential to be used as TTIS, since one could determine the gain in concentration of
such product as the heat process takes place. However, it is not easy to ﬁnd such a thermally
produced compound that can be easily assayed and whose concentration after the process can
be used to verify sterility. Kim and Taub (1993) identiﬁed and characterized three chemicals
that showed potential to be used as TTIS at aseptic processing temperatures. They reasoned
that, since carbohydrates are present in mostly all foods, changes in carbohydrate proﬁle
could give an indication of potential markers. To screen for the compounds, they ﬁrst
monitored formation of chemicals by following changes in UV Spectra in different food
products, including meats, fruits and vegetables. They observed formation of three different
compounds (depending on the food product) which they called M-l (found in heated meats,
fruits and vegetables), M-2 (found primarily in heated meats) and M-3 (found in heated fruits
and vegetables), and identiﬁed (or partially identiﬁed) both the markers and their precursors:

M-l (2,3-dihydro-3,5-dihydroxy-6-methyl-(4H)-pyran-4-one) and M-3

26

(5-hydroxymethylfurfural) are products from the degradation of D-fructose, while M-2 was
associated with protein. Once identiﬁed, the authors proceeded to study the kinetic
parameters of these compounds. They established a linear relationship between M-l
concentration and the sterility index, F, using a non-isothermal process, and developed a
model relating microbial destruction to marker formation. They concluded that "preliminary
experiments conducted with inoculated samples processed in a commercial facility have
indicated that the model is basically valid". However, they acknowledged the need to do
more studies using a more controlled time exposure over a wider range of temperatures to
validate the model and to calculate the actual precision with which the F value of the thermal
process could be determined.

All the methods reviewed above are either very theoretical and need further
validation, or they require expensive equipment, tedious calculations and cannot be applied in
routine testing. There is a need to develop TTIS that are easily recoverable and give a
response that is quickly measured.

Traditionally, the thermal destruction of microorganisms has been assumed to follow
ﬁrst-order kinetics (Kormendy and Konnendy, 1997; Peleg and Cole, 1998). In addition,
methods for estimating the safety of commercial thermal processes are also based on this
assumption (Peleg and Cole, 1998). Upon closer observation, however, some of the survival
curves show a slight curvature. Whiting (1993) explained that, in some cases, the presence of
a subpopulation of more heat-resistant bacteria (i.e. inactivation rate is slower than the rest of
population) could account for the tailing of the survival curve. Peleg and Cole (1998) ﬁtted
previously published survival curves of several pathogenic microorganisms using a

mathematical model. They demonstrated that, in all cases, the heat inactivation rate,

27

previously considered linear, was different from unity. In any case, deviation from linearity
may have an impact on calculation of decimal reduction times, which are determined from
the linear regression of number of survivors vs time at a given temperature (Singh and
Heldman, 1993; Peleg and Cole, 1998). This could result in over- or underestimation of the
actual microbial destruction. In the context of food processing, one is more concerned with
underestimation, which can lead to foodborne diseases. Similar behavior has been observed
in thermal inactivation of certain enzymes. Horseradish peroxidase (Weng et al., 1991a and
b; Hendrickx et al., 1992a and b; Saraiva et al., 1996; Garcia et al., 1998), acid phosphatase
(lncze et al., 1999), and Bacillus spp. Ot-amylase (DeCordt et al, 1992a and b, 1993, 1994a
and b; Van Loey et al., 1995b, 1997a and b), among others, do not follow ﬁrst-order kinetics.
The presence of two or more isoforrns of the enzyme showing different thermostability has
been used to explain the reason of this behavior. The development of a TTI that mimics the
heat destruction of a target attribute (microorganism or enzyme), regardless Of the
inactivation rate of such attribute, could be used for safe monitoring of thermal processing in
foods.
4.4. PHYCOBILIPROTEIN S

Phycobiliproteins are photosynthetic antenna pigments that provide the characteristic
colors to cyanobacteria, red algae and cryptomonads. They are located on the cytoplasmic
face of the thykaloid membrane of the chloroplasts (Glazer, 1989), forming distinctive
macromolecular structures called phycobilisomes. Their function is to harvest solar energy in
regions of the visible spectrum that show low chlorophyll absorption and then transfer this
excitation energy to chlorophyll in the photosynthetic membrane. There are three major

phycobiliproteins: phycoerythrin, phycocyanin and allophycocyanin, each one of them

28

showing speciﬁc absorption and ﬂuorescence patterns (Table 4.1) which will vary with
species (MacColl and Guard-Friar, 1987).

The phycobiliproteins are naturally pigmented due to the presence of linear
tetrapyrrole chromophores, called bilins, which are covalently attached to the apoprotein.
These chromophores are not complexed with metal ions and can be easily manipulated by the

apoprotein to produce the Speciﬁc biological characteristics. There are two major

Table 4.1. Absorption and ﬂuorescence characteristics of phycobiliproteins.

 

 

 

 

Phycobiliprotein Color Absorption bands Fluorescence
(nm) Emission bands (nm)
R-phycoerythrin Red 498-568 570-620
Phycocyanin Blue 550-630 630-650
Allophycocyanin Blue .. 598-680 660-680

 

 

 

 

 

 

chromophores, phycoerythrobilin and phycocyanobilin and at least one of them is present in
all phycobiliproteins. These two chromophores are isomeric, with phycocyanobilin having
one more double bond than phycoerythrobilin (MacCOll and Guard-Friar, 198 7). In addition,
there are three minor bilins: phycourobilin, criptoviolin and the 697-nm bilin, which usually
occur together with one of the major chromophores. The unique Spectral properties of each
phycobiliprotein depend on the native structure of the polypeptide. Alteration of this native
conformation, as by heat or chemical denaturation, will result in partial or complete loss of

the optical properties (Ogawa et al., 1991; Creighton, 1993).

29

R-phycoerythrin (R-PE) contains three different polypeptide chains: or, [3, and 'y or
y'. At pH near neutrality, these subunits are usually organized as (043)” or (OLB)6y'
complexes (D'Agnolo et al., 1994). Each 0: and [3 subunit contains two and three
phycoerythrobilins, respectively, while the 'y subunit has two phycoerythrobilins and two
phycourobilins attached (Glazer, 1981). Thus, each complex contains a total of thirty-four
chromophores. O'hEocha (1965) and F ujimori and Pecci (1967) suggested that these
chromophores are located in hydrophobic pockets of the protein molecule. Two types of
phycoerythrobilin chromophores, donors and acceptors, have been identiﬁed. The donors
absorb at shorter wavelengths (540-555 nm) exciting the acceptors that absorb and emit at
longer wavelengths (565-570 nm) (O'hEocha and O'Carra, 1961; Teale and Dale, 1970).
Teale and Dale (1970) speculated that donor and acceptor chromophores are adjacent to each
other. Studies of the conformation and conﬁguration of phycocyanobilin, a tetrapyrrole found
in phycocyanin, revealed that the preferred conformation of the chromophore in solution is
cyclic-helical, while it is extended in the native protein (Rudiger, 1992). Hence, only when
the apoprotein is in its native conformation does the tetrapyrrole adopt the extended
conﬁguration. Cyclization of the phycocyanobilin results in lower visible and higher near-UV
absorbance than an extended conformation (Burke et al., 1972). The spectroscopic properties
of R-PE depend on the native structure of the molecule and the interaction among
chromophores (O'hEocha and O'Carra, 1961; Teale and Dale, 1970; MacColl and
Guard-Friar, 1987). Alterations of the native conformation will result in partial or complete
loss of the optical properties (O'hEocha and O‘Carra, 1961; MacColl and Guard-Friar, 1987;
Ogawa et al., 1991).

Our source of R-PE is the red alga Porphyra yezoensis, an edible seaweed that is

30

consumed in many Asian countries in the form of "nori", a characteristic ingredient of
"sushi" (Merrill, 1993). Under neutral conditions, R—PE has a characteristic absorption
spectrum showing two maximum peaks at 498 and 565 nm, and a shoulder at about 540 nm.
The extinction coefﬁcient has been estimated to be 8.51 mL mg" cm’1 at 565 nm (Merrill,
1985). In ﬂuorescence measurements and under neutral conditions, R-PE shows excitation
and emission maxima at 493 and 578 nm, respectively.

4.5. PRINCIPLES OF FLUORESCENCE SPECTROSCOPY

The measurement of ﬂuorescence is a potent and useful technique to obtain
information about the structure, dynamics, location and motion of molecules (Lakowicz,
1983; Glazer and Wells, 1998). Fluorescence occurs when a ﬂuorescent molecule
(ﬂuorophore) absorbs light (excitation) and subsequently releases it. The emitted light has
lower energy than the absorbed light because some of the excitation energy is used in such
dynamic processes as vibrations, rotations and collisions within the molecules (Glazer and
Wells, 1998). This process is very fast, usually in the range of nanoseconds.

Depending on the type of ﬂuorescence measurement, one can Obtain different
information on the ﬂuorophore, or the molecule containing such ﬂuorophore. For example,
study of the excitation and emission spectra can be used to identify the ﬂuorophore in
solution (Lakowicz, 1983). The determination of the ﬂuorescence lifetime can provide
information about the ﬂuorophore dynamics in the molecular environment (Glazer and Wells,
1998). Another practical tool is the measurement of ﬂuorescence anisotropy. With the use of
polarized light, one can selectively excite ﬂuorophores that show a particular orientation in
solution. This will result in a partial polarized ﬂuorescence emission. The extent of

polarization can be related to the rotational rate of the ﬂuorophore, and, this, in turn, can be

31

used to study the hydrodynamic properties of macromolecules (Lakowicz, 1983). The
ﬂuorescence efﬁciency or quantum yield is the ratio of photons emitted to the photons
absorbed. The quantum yield is greatly affected by alterations in the microenvironment of the
ﬂuorophore and, thus, can be used to monitor changes within the molecules (Lakowicz,
1983)

The main advantages of ﬂuorescence over other colorimetric methods are its high
sensitivity and speciﬁcity. However, ﬂuorescence is greatly dependent on environmental
factors such as temperature, pH, ionic strength, viscosity, etc. Hence, these factors Should be

carefully monitored and controlled when using ﬂuorescent techniques (Guibault, 1988;

Glazer and Wells, 1998).

32

CHAPTER 5: Effect of pH on the thermal inactivation of R-phycoerythrin from
Porphyra yezoensis
5.1. ABSTRACT
R-phycoerythrin (R-PE), a protein which ﬂuoresces in the visible range, was puriﬁed
from fresh and dried cultures of Porphyra yezoensis by precipitation with ammonium sulfate
and gel permeation chromatography. The isolated R-PE had a purity greater than 5 as
determined by A565/A280. The effect of pH on spectral properties and thermal inactivation of
R-PE was examined. Fluorescence was stable between pH 5.0 and 9.0 but decreased at pH
4.0 and 10.0. Complete loss of ﬂuorescence was observed at pH 3.0 and below, and pH 11.0
and above. Thermal inactivation parameters calculated on the basis of ﬂuorescence loss were
Obtained in the pH range 4.0-10.0, at the 60-90°C temperature range. Maximum
thermostability was Observed in pH 6.0 (D value at 70°C =12,258 min), while R-PE was least
thermostable in pH 10.0 (D value at 70°C =1.10 min). The 2 values ranged from 444°C at
pH 10.0 to 915°C at pH 9.0. These results are of particular signiﬁcance in the area of
protein-based time-temperature integrators for monitoring thermal processes in food
products. By modifying the thermal kinetics of R-PE to match that of a target microorganism
in a thermal food process, it could be used as a time-temperature integrator to predict
destruction of such microorganism during cooking.
5.2. INTRODUCTION
Thermal processing is the most common method used for food preparation and

preservation. The conditions of a heat process are usually determined by the type of food

33

product and the speciﬁc thermal treatment applied. Mathematically, the impact of a thermal
process can be expressed in terms Of D and 2 values (Singh and Heldman, 1993; Hendrickx et
al., 1995). D value, or decimal reduction time, is the time in minutes required to decrease a
quality attribute or microbial population by 90% at a constant temperature. The 2 value is the
temperature increase necessary to reduce the D value by 90%. The determination of D and 2
values for a particular microorganism is very useful in designing a thermal process that
targets that speciﬁc microorganism (Pﬂug, 1997). Sterilization processes of canned foods are
designed to obtain a 12D reduction in Clostridium botulinum (Plug and Odlaug, 1978). The
times and temperatures used for milk pasteurization were based on thermal destruction of
Mycobacterium tuberculosis (Kay and Graham, 1933). Processing schedules required by the
US Department of Agriculture for roast beef and hamburger patties were based on thermal
inactivation of Salmonella (Goodfellow and Brown, 1978) and Escherichia coli 01 57:H7
(Doyle and Schoeni, 1984; Line et al., 1991), respectively.

F oodbome outbreaks associated with underprocessed food products have prompted
the search for methodology to ensure adequate thermal food processing. The evaluation of
thermal processes in foods can be done using in situ methods, physical-mathematical models
or time-temperature integrators. Time-temperature integrators (TTIS) can be used to predict
the time-temperature response of a quality or safety index in a food product. This occurs
when the thermal destruction of both TTI and target index are identical (zTTI = 2%“)
(Hendrickx et al., 1995; Van Loey et al., 1996).

Most currently existing TTIS are based on microbiological or enzymatic assays.
Microbiological TTIS mainly employ vegetative cells or spores from thermophilic bacteria

(Yawger, 1978; Pﬂug et al., 1980; Sastry et a1, 1988). The main disadvantages of

34

microbiological TTIS are the length of assay because of the microorganism incubation period
and the likelihood of contamination, as well as cost, labor and the need to properly calibrate
the spores.

Different enzymes, mainly from microbial origin, have also been proposed to be used
as TTIS. The applicability of these enzymes as TTIS lies on the modiﬁcation of their
inactivation kinetics to match those of speciﬁc microorganims targeted in the thermal process
(Hendrickx et al., 1995).

R-phycoerythrin (R-PE), one of the major phycobiliproteins, is a protein that
ﬂuoresce in the visible range and is found in cyanobacteria, red algae and cryptomonads. The
red pigmentation of R-PE is due to the presence of linear tetrapyrrole chromophores, called
bilins, which are covalently attached to the apoprotein through one or two thioether bonds
(MacColl and Guard-Friar, 1987; Sidler et al., 1989). R-PE contains two different types of
chromophores, phycoerythrobilin and phycourobilin (MacColl and Guard-Friar, 1987). The
apoprotein is formed by three different polypeptide chains: or, B, and y or y'. At pH near
neutrality, these subunits are usually organized as (OLB)6Y- or (aB)6y'-complexess (D'Agnolo
et al., 1994). Each 0c and B subunit contains two and three phycoerythrobilins, respectively,
while the Y subunit has two phycoerythobilins and two phycourobilins (Glazer, 1981). Thus,
each complex contains a total of thirty-four chromophores. O'hEocha (1965) and Fujimori
and Pecci (1967) suggested that these chromophores are located within hydrophobic regions
of the protein molecule. There are two types of phycoerythrobilin chromophores that have
been classiﬁed as donors and acceptors. Teale and Dale (1970) suggested that there must be a
close proximity between donor and acceptor chromophores. The unique spectral properties of

R-PE depend on the native structure of the polypeptide and interaction among the different

35

chromophores (O'hEocha and O’Carra, 1961; Teale and Dale, 1970; MacColl and
Guard-Friar, 1987). Alteration of this native conformation, as by heat or chemical
denaturation, will result in partial or complete loss of the optical properties (O'hEocha and
O'Carra, 1961; MacColl and Guard-Friar, 1987; Ogawa et al., 1991). The amino acid
sequence of R-PE subunits in Porphyridiun cruentum (Sidler et al., 1989), Rhodella violacea
(F icner et al., 1992) and Aglaothamnion neglectum (Aptt et al., 1993) has been elucidated.
No evidence of disulﬁde bonds in R-PE molecules has been found.

The main objective of this project was to study the effect of pH on the thermal
inactivation parameters Of R-PE. By modifying the environmental conditions, the
inactivation kinetics of R-PE could be adjusted to match those of target indicators in thermal
food processes.

5.3. MATERIALS & METHODS
5.3.1. Source of R-phycoerythrin.

R-phycoerythrin (R-PE) was puriﬁed from pure cultures of Porphyra yezoensis Ueda
sporophyte phase (conchocelis), strain U-51 (inoculum kindly provided by Dr. A. Miura,
Tokyo University of Fisheries, Tokyo, Japan). Alternatively, R-PE was also puriﬁed from
dried whole P. yezoensis gametophyte thalli provided by the Tagawa Fisheries Supply
Company and the Shin-Futtsu Nori Cooperative, Shin-Futtsu, Japan. The extraction and
puriﬁcation procedure was based on methods in Merrill (1985) and are summarized in Figure
5.1.

5.3.2. Puriﬁcation of R-phycoerythrin
5.3.2.1. Procedure for fresh algal tissue:

Algal tissues were homogenized with cold 50 mM NaCl, 50 mM phosphate buffer,

36

P. yezoensis culture

l 50 mM NaCl, 20 mM phosphate

 

P. yezoensis dried tissue

 

 

50 mM NaCl, 20 mM
buffer, l phosphate buffer,
pH 7.2 pH 7.2
Blend Blend
20% AS , , 20% AS
saturation saturation I Pellet
Pellet 30% AS saturation
Pel
I let
35% AS 35% AS
saturation saturation
Supernatant Repeat > Supernatant
ODCC
V
l flt t. .
Elfrorhart: 1:: h Gel ﬁltration
g py\‘ / chromatography
Puriﬁed
fractions

Figure 5.1. Flow diagram for the puriﬁcation of R-phycoerythrin from P. yezoensis. All steps
were performed at 4°C.

37

pH 7.2 (PBS) in a Bead-Beater® (Biospec Products, Bartlesville, OK). The algal suspension
was homogenized in four bursts of 305, followed by cooling periods of 5 min. The resulting
homogenate was decanted and centrifuged at 6,000 x g for 15 min at 4°C, to eliminate cell
debris. After centrifugation, the extract was brought to 20% saturation with solid ammonium
sulfate. After stirring for 1h at 4°C, the suspension was centrifuged at 10,000 x g for 15 min
at 4°C. The resulting supernatant was collected and brought to 35% ammonium sulfate
saturation. After stirring for 1h at 4°C, the suspension was centriﬁiged at 10,000 x g for 15
min, at 4°C. The resulting pellet was suspended in 25 mL of PBS. This suspension (10 mL)
was loaded onto a 26 x 100 mm Sephacryl S-300 gel ﬁltration chromatography column
(Pharmacia, Piscataway, NJ) equilibrated with PBS. Loaded volumes were calculated SO as
not to exceed 5% of total column bed volume. The column was eluted at a 0.4 mL/min ﬂow
rate. Two-milliliter fractions were collected and absorbances at 280 and 565 nm measured. A
ratio of A565/A280 > 5 was established as the standard for puriﬁed protein (Gantt, 1969).
Fractions showing a A5,,5/A280 > 5 were pooled, brought to 60% ammonium sulfate saturation
and stored at 4°C.

5.3.2.2. Procedure for dry algal tissue:

Dried tissue (5g) was soaked overnight at 4°C in 100 mL of PBS. On the following
day, the tissue was extracted using the Bead-Beater® as explained above. The glass beads
were rinsed with PBS until a total volume of IL homogenate was collected. The homogenate
was centrifuged at 6,000 x g for 15 min at 4°C, to eliminate cell debris. After centrifugation,
the extract was brought to 20% saturation with solid ammonium sulfate. After stirring for 1h
at 4°C, the suspension was centrifuged at 10,000 x g for 15 min at 4°C. The same procedure

was repeated for 25, 30 and 35% ammonium sulfate saturation. The resulting pellet was

38

suspended in 90 mL of PBS. The ﬁnal suspension (15 mL) was loaded onto a 26 x 100 mm
Sephacryl S-300 gel ﬁltration chromatography column (Phannacia) equilibrated with PBS.
The column was eluted at a 0.35 mL/min ﬂow rate. Fractions (2 mL) were collected and
absorbances at 280 and 565 nm measured. Fractions showing a A565/A280 > 5 were pooled,
brought to 60% ammonium sulfate saturation and stored at 4°C.

5.3.2.3. Purity and yield of extracted R-phycoerythrin

Purity was also conﬁrmed by native polyacrylarnide gel electrophoresis (PAGE)
(Davis, 1964) using a 4% stacking and a 10% resolving gel. Gels were developed using
Coomassie Blue R-250 (Bio-Rad, Hercules, CA).

Absorption spectra of R-PE extracted with each procedure were identical. Fractions
were pooled together into what was designated the R-PE stock solution and stored in 60%
ammonium sulfate at 4°C. The ﬁnal solution had a concentration of 0.31 mg/mL and a purity
Of 5.2 as measured by the ratio A565/A280. Average yield was estimated to be about 1% of the
fresh weight.

5.3.3. Determination of molecular weight and isoelectric point.

Sodium dodecyl sulfate (SDS) PAGE (Laemli, 1970) was performed to estimate the
molecular weight of the R-PE subunits using a 4% stacking and 14% resolving gel
(Mini-Protean II, Bio-Rad, Hercules, CA). Molecular weights were determined by comparing
relative mobility of protein bands to those of prestained molecular weight standards
(Kaleidoscope Polypeptide Standard ~6.9 - 205 kDa, Cat. No. 161-0324, Bio-Rad) according
to the method of Weber and Osborn (1969). Gels were developed using Coomassie Blue
R-250 and/or a Silver stain kit (Bio-Rad).

The isoelectric point (pl) of R-PE was determined by isoelectric focusing in a vertical

39

polyacrylamide minigel system (Model SE250/260, Hoefer, Piscataway, PA) using
ampholyte pH range 4.0—7.0 (Sigma Chemical, St. Louis, MO) (Robertson et al., 1987).
Isoelectric focusing was performed at constant voltage (200 V) for 2 h. Gels were developed
using Coomassie Blue R-250 (Bio-Rad). The pI was determined by comparing relative
mobility of protein bands to those of protein standards (pH 3.6-6.6, Sigma) run on the same
gel.

5.3.4. Absorbance and ﬂuorescence measurements.

Absorption spectra was determined using a Lambda 20 UVN is Spectrometer and the
UVWinlab Version 2.0 software (Perkin Elmer Corporation, Norwalk, CT). Fluorescence
spectra of R-PE and ﬂuorescence intensity at the emission maximum were determined using
a computer enhanced SLM-4000 ﬂuorometer (SLM Instruments, Rochester, NY). A scan of
maximum excitation and emission spectra was performed for each solution. For thermal
inactivation experiments, excitation was set at 493 nm for all pH conditions, while emission
was recorded in the 578 and 581 nm, depending on pH.

5.3.5. Effects of pH on thermal inactivation of R-PE.

To test the effect of pH, R-PE from the stock solution was dialyzed against water at
4°C overnight. The dialyzed R-PE was diluted immediately prior to the experiments in 20
mM buffers containing SOmM NaCl, ranging from pH 3.0 to 12.0, to achieve an absorbance
of 0.1 at 565 nm. Buffers employed were: citrate (pH 3.0, 4.0 and 5.0, pKal = 3.13, pKa2 =
4.45), phosphate (pH 6.0, 7.0 and 12.0, pKa2 = 6.88, pKa3 = 12.38),
tris(hydroxymethyl)amino-methane (TRIS) (pH 8.0 and 9.0, pKa = 8.2), and glycine (pH
10.0, pKa = 9.78). The pH of R-PE in TRIS buffer changed no more than 0.3 units at the

temperature used.

40

The dialyzed R-PE solution was diluted in each buffer to achieve an absorbance of
0.1 at 565 nm, and 2 mL transferred to 10 x 75 mm test tubes (14-961-25, Fisher Scientiﬁc,
Pittsburgh, PA). The tubes were sealed with Teﬂon( tape and allowed to equilibrate at 4°C.
The test tubes were placed in a wire rack and immersed in a Polystat circulator bath (Model
1268-52, Cole-Parmer Instrument Company, Chicago, IL) connected to a bath programmer
(Model 1268-62, Cole-Farmer). Temperatures ranged from 60 to 90°C. The water bath
temperature was set 0.2°C above the target temperature to allow R-PE solution to reach target
temperature. Temperature in the test tubes was monitored using a thermocouple (Resistance
Temperature Detector Probe, Pt 100, 0.15 cm diameter, (1°C accuracy) inserted in a control
test tube containing 2 mL of the same solution. The thermocouple was connected to a
Solomat MPM 200 Modumeter (Solomat Partners LP, Stanford, CT). Zero time was deﬁned
as the time when the R-PE solution reached the target temperature. Preliminary studies were
performed to calculate the times and temperature necessary to achieve at least a 1 log
reduction in ﬂuorescence. The tubes were removed at predetermined intervals of time and
immediately placed in an ice bath, then kept at 4°C until ﬂuorescence was read within 12 h.
Preliminary studies were performed to Show that no changes of ﬂuorescence occurred during
this holding period. Each thermal inactivation study was done in triplicate.

5.3.6. Calculation of D and 2 values.

Fluorescence counts were converted to logarithms. D values (in minutes) were
calculated by linear regression of ﬂuorescence vs time using Microsoft Excel Version 5.0a
(Microsoft Corporation, Redmond, WA). Thermal resistance curves were determined by
plotting log of D values vs temperature. The 2 values were determined as the negative

reciprocal of the slope of the thermal resistance curve.

41

Statistical differences among 2 values were tested using one way analysis of variance
(SAS Institute Inc., 1995). Mean 2 values were compared using Tukey-Kramer HSD test with
the mean square error at the 5% level of probability.
5.4. RESULTS & DISCUSSION
5.4.1. Determination of molecular mass and p1.

On SDS-PAGE, R-PE migrated as two bands with molecular masses of21.2 and 38.3
KDa, corresponding to a and B (resolved as a single band) and Y subunits, respectively.
Merrill (1985) reported molecular masses of 20-21 KDa for on and B, and 30 KDa for Y
subunits of R—PE in P. yezoensis. Glazer and Hixson (1977) reported values Of 17.5 KDa for
a and B and 30.2 KDa for Y in R-PE from P. cruentum. D'Agnolo et al. (1994) calculated a
molecular mass of 19-21.5 KDa for a and B, and 30-33 Da for Y subunits of R-PE of
Gracilaria longa. Slight differences in calculated molecular masses could be due to
differences in algal species and/or strains or errors inherent in SDS-PAGE method.

Summation of the individual molecular masses of the R-PE subunits according to the
molecular formula (06B)6Y( (MacColl and Guard-Friar, 1987; D'Agnolo et al., 1994) resulted
in a molecular mass of 292 KDa. This value agrees with published values in the literature.
The molecular mass of R-PE has been reported to range from 220 to 300 KDa, with
variations depending on species (Gantt, 1969; MacColl and Guard-Friar, 1987; D'Agnolo et
aL,1994)

On isoelectric focusing, two distinct naturally colored bands appeared with p15 of 4.2
and 4.1, respectively. The presence of two distinct bands, suggested that two R-PE isomers
could coexist in the isolated protein. A minor band appeared with a pl of 6.1, which might

correspond to subunit Y (Figure 5.2). The pI of R-PE from Rhodella violacea was reported to

42

1 2

6.6 —'

‘—6.1
5.9 —'

5.4 ——>

5.1 —*

4.6—*.

4.2—’ " \41

3.6 —’ "“

Figure 5.2. Isoelectric focusing of R-phycoerythrin from P. yezoensis in a 7.5 % acrylamide
gel covering the pH range 4.0 to 7.0. 1) Isoelectric focusing standard, 2) R-phycoerythrin.
be between 4.4 and 4.2 (MacColl and Guard-Friar, 1987). D'Agnolo et al. (1994) reported
two pairs of major bands at pI 4.35-4.30 and 4.55-4.50, and a minor component at pI
5.75-5.70 in R-PE from G. longa. The authors suggested that the band focused at the more
basic pH could correspond to migration Of the Y subunit that dissociated from the intact
native molecule.
5.4.2. Effect of pH on R-PE Spectral properties.

The absorption spectra of puriﬁed R-PE in 50mM NaCl, 20 mM phosphate buffer, pH
7.0, showed two peaks at 498 and 565 nm, and a shoulder at about 540 nm (Figure 5.3).
Similar spectra has been found for R-PE puriﬁed from Porphyridium cruentum (Gantt, 1969;
Teale and Dale, 1970; Tcheruov et al., 1993), Rhodella violacea (MacCOll and Guard-Friar,
1987) and Gracilaria longa (D'Agnolo et al., 1994).

In ﬂuorescence measurements, while excitation maximum remained the same (493

43

 

 

1.4
1.2“
s 1.0.
C
Q) a
D
a 0.8“
.9 .
‘5.
0 0.6“
0.4‘
,
0.2j
0 _ .

 

235 300 350 400 450 500 550 600 650 700
Wavelength (nm)

Figure 5.3. Absorption spectrum of R-phycoerythrin from P. yezoensis in 50mM NaCl, 20
mM sodium phosphate buffer, pH 7.0.
nm), emission maxima varied Slightly with pH. In 50 mM NaCl, 20 mM citrate buffer, pH
4.0, the emission maximum was at 581 nm. In 50 mM NaCl, 20 mM phosphate buffer, pH
7.0, the emission maximum was 578 nm, and in 50 mM NaCl, 20 mM glycine buffer, pH
10.0, it was 579 nm. Excitation and emission maxima did not differ with pH between 5.0 and
9.0.

Fluorescence intensity of R-PE was similar at pH range 5.0 - 9.0, but decreased at pH
4.0 and 10.0. R-PE did not ﬂuoresce at pH'S below 4.0 and above 10.0. Moreover, the

solution of R-PE under neutral conditions (pH 5.0 - 9.0) showed a bright pink-red color that

44

shifted to purple at pH'S above and below neutrality. The intensity of the purple color
increased with very low and very high pH'S.

Ogawa et al. (1991) studied the behavior of R-PE in the pH range 3.0 - 11.0 using
viscosity and sedimentation measurements and spectroscopic techniques. While they did not
observe changes in R-PE absorbance between pH 5.0 and 9.0, all techniques detected
changes in protein conformation at pH'S below 5.0 and above 9.0. Values of intrinsic
viscosity also remained nearly constant between pH 5.0 to 10.0, but increased in more acidic
or alkaline regions of pH. The color of the R-PE solution changed from red to purple and the
ﬂuorescence disappeared at high and low pH's. The authors concluded that R-PE assumes
three forms depending on pH conditions: an aggregated form in acidic solutions near the pl, a
globular or compacted form in neutral conditions and partially unfolded form at pH above
9.0.

Phycocyanin, a closely related phycobiliprotein, has been shown to follow reversible
association-dissociation of subunits with changes of pH. The pl of phycocyanin is 4.3-4.0
(Hattori et al., 1965). At pH between 5.0 and 7.0, this protein shows an equilibrium of
hexamers (043),, trimers (OLB)3 and monomers (aB) (Hattori et al., 1965; Ogawa et al.,1991).
Hattori et al. (1965) reported changes in phycocyanin absorption spectra with pH. The
authors concluded that at pH near the pI, the reduction Of electrostatic repulsive forces
between molecules led to an association of the monomers, which resulted in alteration of the
optical properties. MacColl and Guard-Friar (1987) have also described changes in
absorption spectra of phycocyanin related tO changes in the aggregation pattern of this
protein. At low pH, when the protein was denatured, it showed much lower visible

absorption. The authors explained that changes in optical properties could be due to alteration

45

of the conformation of the chromophores resulting from denaturation of the polypeptide.
Rudiger (1992) reported that the chromophore phycocyanobilin has a cyclic-helical
conformation in solution, while it is extended in the native protein. This suggested that the
native structure of the apoprotein allows the chromophore to adopt the linear conﬁguration.
Moreover, cyclization of tetrapyrroles resulted in lower visible and higher near-UV
absorbance (Burke et al., 1972). Therefore, it can be concluded that the spectral properties of
phycocyanin and other phycobiliproteins are inﬂuenced by the structure of the protein
molecule.

5.4.3. Effect of pH on thermal inactivation of R-PE.

The effects of pH on the thermal inactivation of R-PE were evaluated. The
inactivation temperatures selected to obtain a one log reduction in R-PE ﬂuorescence ranged
from 60°C to 90°C depending on the pH. Between pH 5.0 and 9.0, thermal inactivation
experiments could be performed at temperatures up to 90°C. Thermal inactivation of R-PE at
pH'S 4.0 and 10.0 could not be performed at temperatures higher than 70°C, due to the
substantial loss of ﬂuorescence intensity during the heating lag time. O'hEocha and O'Carra
(1961) also noted increasing quenching of ﬂuorescence upon acidiﬁcation of R-PE solutions.
Bems et al. (1963) reported that thermal denaturation of R-PE occurred at lower temperatures
at pH 4.7 than at pH 7.0, when measured by ﬂuorescence decay.

The D values were calculated on the basis of ﬂuorescence loss as determined at the
maximum excitation peak. It is to be noted that reduction of ﬂuorescence at this band, could
result from either loss of ﬂuorescence intensity at this particular point or by Shift of the peak
to lower or higher regions of the spectra.

Neutral pH 's. R-PE was thermostable at pH'S between 5.0 and 8.0 where inactivation

46

temperatures ranged from 70°C to 90°C. D value was deﬁned as the time necessary to reduce
R-PE ﬂuorescence by 90% at a given temperature and is indicative of the thermal stability of
R-PE at a constant temperature. Thermal inactivation at pH 6.0 resulted in the highest D
values at 70°C and 90°C (12258.72 min and 1.43 min, respectively), while the highest D

value at 80°C was found at pH 7.0 (461.87 min) (Table 5.1).

Table 5.1. D values for R-phycoerythrin at different pHs.

 

 

 

 

 

 

 

 

 

 

 

 

pH D values (min)

4.0 Dﬁ0 = 303 Dm = 49 D70 = 2
5.0 D1L== 9274 Dan = 91 D90 = 0.4
6.0 Dm = 12259 Dan = 240 D90 = 1.4
7.0 D7() = 2205 Dan = 462 D90 = 1.2
8.0 D70 = 2048 D30 = 124 D9,, = 1.3
9.0 D70 = 94 Dan = 4 D90 = 0.6
10.0 D60 =196 D65 =13 D70 =1.1

 

 

R-PE has been reported to be highly stable near neutral pH (Gantt, 1969; Ogawa et
al., 1991). Ogawa et al. (1991) studied conformational changes of the R-PE molecules and
concluded that at pH'S between 5.0 and 9.0, the protein has a compact globular shape that
confers stability. As the pH increases from the pI, the protein becomes more soluble because
of the increase in electrostatic repulsion that reduces aggregation and subsequent
precipitation of the molecules (Creighton, 1993). Thermal unfolding of the protein will alter
the polypeptide conformation to expose the chromophores that are contained in hydrophobic
regions of the protein (O'hEocha, 1965). Unfolding of the protein may cause separation of
chromophores that, under native condition, stay in close proximity and interact to produce the
speciﬁc Optical properties of the native protein (Teale and Dale, 1970; Glazer and Hixson,

1977). In addition, denaturation of the protein molecule may result in cyclization Of

47

chromophores with the subsequent loss of absorbance and ﬂuorescence (Burke et al., 1972;
Rudiger, 1992).

The 2 value is the temperature increase necessary to reduce the D value by 90% and is
a measure of the temperature dependence of R-PE denaturation. The 2 values of R-PE
increased as pH was increased from 458°C at pH 5.0 to 915°C at pH 9.0. There was no
difference between 2 values at pH 7.0 and 8.0 (p > 0.05). The 2 value at pH 9.0 (9.15°C) was
the highest (p < 0.05) among all pH's examined, although the D values were lower at all
temperatures when compared with pH's 5.0 to 8.0 (Table 5.2). Results suggest that thermal
inactivation of R-PE was less temperature dependent at pH 9.0 at the range of temperatures

tested.

Table 5.2. Summary of 2 values and regression analysis for R-phycoerythrin at different pHs.

 

 

 

 

 

 

 

 

pH 2 value (°C) Constant R squared X coefﬁcient
4.0 4.59“ 15.7 0.98 -022
5.0 4.58“ 19.3 0.99 -022
6.0 5.09b 17.9 0.99 -020
7.0 6.13“ 15.1 0.90 -0.16
8.0 6.25c 14.6 0.98 -O.16
9.0 9.15d 9.5 0.98 -0.11
10.0 4.443 15.8 0.99 -0.23

 

 

 

 

 

 

 

Buffers employed were: 20 mM citrate (4.0 and 5.0), 20 mM phosphate (pH 6.0 and 7.0), 20 mM
tris(hydroxymethyl)amino-methane (pH 8.0 and 9.0), and 20 mM glycine (pH 10.0). All buffers contained 50
mM NaCl.

Acidic and alkaline pH. At pH 4.0, the selected inactivation temperatures ranged from
60°C to 70°C. At 70°C, the D value (2.01 min) of R-PE at pH 4.0 was several orders of

magnitude lower than in neutral pH'S, indicating that R-PE was more heat sensitive at pH 4.0

48

(Table 5.1). At pH 4.0, R-PE is near its isoelectric point (p1 = 4.1-4.2) and may form
aggregates with reduced solubility (Ogawa et al., 1991).

At pH 10.0, the heating temperatures selected to obtain a one log reduction in R-PE
ﬂuorescence ranged from 60°C to 70°C. At 70°C, the D value (2.01 min) was much lower
than in neutral pH'S. Overall, the D values of R-PE at pH 10.0 were lower than those of pH
4.0 at all temperatures, suggesting that R-PE is more thermostable in acidic than in alkaline
conditions (Table 5.1). At pH 10.0, the protein is far from its pI (4.1-4.2) and has a marked
negative net charge. Interactions between ionizing groups in the polypeptide are disrupted at

this extreme pH, resulting in unfolding of the molecule (Ogawa et al., 199). This partial

 

unfolding results in alteration of the spectroscopical properties of the protein when compared
to the neutral solutions. When R-PE is heated at pH 10.0 ﬂuorescence is lost by increasing
the exposure of the chromophores to the solution and by decreasing interactions among the
chromophores (O'hEocha, 1965; F ujimori and Pecci, 1967; Teale and Dale, 1977).

The 2 values of R-PE at pH 4.0 and 10.0 were 4.59 and 444°C, respectively (Table
5.2). The 2 values at pH 4.0, 5.0 and 10.0 were similar (p < 0.05) even though experiments at
pH 5.0 were performed at higher temperatures (70-90°C) than at pH 4.0 or 10.0 (60-70°C).
The 2 values at 4.0, 5.0 and 10.0 were lower (p > 0.05) than at pH between 6.0 and 9.0.
Results suggest that the thermal inactivation of R-PE was more temperature dependent at
acidic and alkaline conditions at the range of temperatures tested.
5.5. CONCLUSIONS

The thermal inactivation behavior of R-PE was successfully modiﬁed by altering the
solution composition. The protein was very thermostable at pH 5.0 to 9.0 as measured by

ﬂuorescence decay at the maximum emission. The 2 values of R-PE ranged from 444°C to

49

9.15°C depending on pH. Changes in spectroscopical properties were due to changes in the
protein conformation that result from chemical (pH) and physical (heat) denaturation. By
manipulation of R-PE in a known chemical environment, the 2 value may be adjusted to
approximate the 2 values of several pathogenic microorganisms used as target indicators in
thermal food processes. The encapsulated R-PE could then be used as a TTI to predict
destruction of such microorganisms during cooking of food products.

The z values obtained at pH 4.0, 5.0 and 10.0 (4.59, 4.58 and 444°C, respectively)
should be investigated further for its possible application as TTIS in milk pasteurization,
where the target microorganism, Coxiella burnetti, has a 2 value of 4°C (Goff, 1999). The 2
value of R-PE obtained at pH 6.0 (509°C) could be used to monitor destruction of
Escherichia coli 0157:H7 in beef (25.60,, 015m7= 5.6°C) (Orta-Ramirez et al., 1995) while at
pH 7.0 and 8.0 (6.13 and 625°C, respectively) the 2 value falls within the range of z values
reported for Salmonella in beef (5.6-6.2°C) (Goodfellow and Brown, 1978; Orta—Ramirez et
al., 1995). Additional investigation of the combined effect of pH and additives to further
modify the thermal inactivation kinetics of R-PE would be desirable to assess its usefulness

in the area of TTIS.

50

CHAPTER 6: Effects of sucrose, SDS, NaCl, urea and B-mercaptoethanol on the
thermal inactivation of R-phycoerythrin
6.1. ABSTRACT

Thermal inactivation kinetics (D and 2 values) of R-phycoerythrin (R-PE), calculated
on the basis of ﬂuorescence loss, were studied under different buffer conditions (pH 4.0, 7.0
and 10.0) and concentrations of sucrose, sodium-dodecyl sulfate (SDS), NaCl, urea and
B-mercaptoethanol (ME). Isothermal experiments were performed using capillary tubes to
minimize heating lag times. R-PE solutions were heated at temperatures between 40°C and
90°C depending on buffer conditions. The 2 values ranged from 590°C in 50 mM NaCl, 20
mM glycine buffer, pH 10.0, to 37.75°C in 60% sucrose, 50 mM NaCl, 20 mM phosphate
buffer, pH 7.0. Thermal inactivation parameters for R-PE were successfully modiﬁed by
addition of chemicals. Overall, sucrose and ME had a thermostabilizing effect, while SDS,
NaCl and urea decreased thermal stability of R-PE.

These results have relevance in the area of protein-based time-temperature integrators
to verify compliance with USDA thermal processing of meat products. The 2 value obtained
for R-PE in 50 mM NaCl, 20 mM glycine buffer, pH 10.0 (5.90°C) closely matched the 2
value reported for Salmonella in beef. The development of a TTI with R-PE as the detector
component could be used to verify the adequacy Of any thermal treatment in beef products.
6.2. INTRODUCTION

Current methodology to assess proper thermal processing in food products can be

divided into three categories: in situ methods, mathematical models and time-temperature

51

integrators (TTIS). A TTI is a device that can predict the thermal history of a cooked product
because it shows the same time-temperature response of a target attribute when the
temperature is the only rate determining factor. Mathematically, this is accomplished when
the 2 value of the TTI is the same as that of a target attribute (zTTI = 2mg“) (Hendrickx et al.,
1995; Van Loey et al., 1996). The United States Department of Agriculture Food Safety
Inspection Service (USDA-F SIS) has recently proposed a modiﬁcation of the current thermal
processing regulations for meat products based on destruction of the pathogenic
microorganism Salmonella expressed as log reductions. It is required that manufacturers use
a combination of thermal and non-thermal processes to achieve a 6.5-log Salmonella
reduction in ready-tO-eat, cooked beef, roasted beef, cooked corned beef. A 5-log reduction in
ready-tO-eat cooked uncured meat patties has been proposed, although this latter one may be
increased (Federal Register, 1996; Federal Register, 1999a). Reported 2 values Of Salmonella
in beef ranged from 56°C to 62°C (Goodfellow and Brown, 1978; Orta-Rarnirez et al.,
1995). The development of a TTI with the same 2 value as Salmonella in beef products could
be used to determine the adequacy of any thermal process.

R-phycoerythrin (R-PE), one of the major phycobiliproteins, is a protein found in
cyanobacteria, red algae and cryptomonads. The optical properties of R-PE are due to the
presence of linear tetrapyrrole chromophores, called bilins, which are covalently attached to
the apoprotein through one or two thioether bonds (MacColl and Guard-Friar, 1987; Sidler et
al., 1989). R-PE has two different types of chromophores, phycoerythrobilin and
phycourobilin (MacColl and Guard-Friar, 1987). The polypeptide chain contains three
different subunits, 01, B, and Y or Y', usually organized as (aB)6Y- or (aB)6Y'-complexess

(D'Agnolo et al., 1994). Each a and B subunit contains two and three phycoerythrobilins,

52

respectively, while Y has two phycoerythobilins and two phycourobilins (Glazer, 1981).
Thus, each complex contains a total of thirty-four chromophores. O'hEocha (1965) and

F uj imori and Pecci (1967) suggested that these chromophores are located in hydrophobic
pockets of the protein molecule. Two types of phycoerythrobilin chromophores, donors and
acceptors, have been identiﬁed. The donors absorb at shorter wavelengths (565-570 nm)
exciting the acceptors that absorb and emit at longer wavelengths (540-5 55 nm) (O'hEocha
and O'Carra, 1961; Teale and Dale, 1970). Teale and Dale (1970) speculated that donor and
acceptor chromophores are adjacent to each other. Studies of the conformation and
conﬁguration of phycocyanobilin, a tetrapyrrole found in phycocyanin, revealed that the
preferred conformation of the chromophore in solution is cyclic-helical, while it is extended
in the native protein (Rudiger, 1992). Cyclization of the phycocyanobilin results in lower
visible and higher near-UV absorbance than an extended conformation (Burke et al., 1972).
The spectroscopic properties of R-PE depend on the native structure of the molecule and the
interaction among chromophores (O' hEocha and O' Carra, 1961; Teale and Dale, 1970;
MacCOll and Guard-Friar, 1987). Modiﬁcation of this native structure, as by heat or chemical
denaturation, will result in partial or complete loss of the optical properties (O'hEocha and
O'Carra, 1961; MacColl and Guard-Friar, 1987; Ogawa etal., 1991).

Ogawa et a1. (1991) studied the behavior of R-PE in the pH range 3.0-11.0. Although
no changes in R-PE between pH 5.0 and 9.0 were observed, there were alterations in protein
conformation starting at pHs below 5.0 and above 9.0. The intrinsic viscosity remained
nearly constant between pH 5.0 to 10, while it increased in more acidic or alkaline

conditions. The authors concluded that R-PE adopts three forms depending on pH conditions:

53

an aggregated form in acidic solutions, a globular form between pH 5.0 and 9.0 and a
partially unfolded form at pH above 9.0.

Thermal inactivation of proteins results in a partial unfolding of the molecule
(Klibanov, 1983). The compact structure of a native protein is sustained by a combination of
noncovalent interactions between atoms in the molecule, including hydrogen bonds, van der
Waals interactions, electrostatic forces, and, most importantly, hydrophobic interactions
(Klibanov, 1983; Creighton, 1993). The presence of additives during heating may stabilize or
destabilize the native structure of the molecules, thus, affecting the overall thermal
inactivation kinetics of the protein.

In previous studies (Chapter 2), the thermal inactivation parameters of R-PE in the pH
range 4.0-10.0 were calculated on the basis of ﬂuorescence loss at the maximum emission
wavelength. R-PE was very thermostable at pH between 5.0 and 9.0, but ﬂuorescence
decreased at pH 4.0 and 10.0. We selected three pHs, 4.0, 7.0 and 10.0, representing acidic,
neutral and alkaline conditions, to further investigate the inﬂuence Of the solution conditions
on the thermal inactivation of R-PE. The objective of this study was to determine the
inactivation kinetics parameters of the ﬂuorescent protein R-phycoerythrin (R-PE) under
different conditions of pH and additives (sucrose, sodium-dodecyl sulfate, NaCl, urea and
B-mercaptoethanol). Our goal was to ﬁnd speciﬁc solution conditions that would result in a 2
value of R-PE close to the reported 2 value of Salmonella in beef.

6.3. MATERIALS AND METHODS
6.3.]. Materials.
R-phycoerythrin (R-PE) was extracted and puriﬁed as explained in Chapter 2. Pure

R-PE was stored in 60% (w/v) ammonium sulfate and kept at 4°C. Prior to the experiments,

54

R-PE was dialyzed overnight against distilled deionized water at 4°C, then diluted
immediately before use in the solutions described in 6.3.2 to achieve an absorbance of 0.1 at
565 um. All chemicals were reagent grade or better.

6.3.2. Solution conditions.

The effects of pH and additives on the thermal inactivation kinetics of R-PE were
investigated using sucrose (0, 20, 40 and 60% w/v at pH 7.0 and 10.0); sodium dodecyl
sulfate (SDS) (0, 0.5, 1 and 2.5% v/v at pH 7.0 and 10.0);NaC1 (0, 1, 2 and 4M at pH 7.0,
and 0, 0.5, 1 and 2M at pH 10.0); urea (0, 0.1, 0.25 and 0.5M at pH 4.0; 0, 1, 2 and 4M at pH
7.0, and 0, 0.5, l and 2M at pH 10.0); and B-mercaptoethanol (ME) (0, 0.5, 1 and 2.5% v/v,
at pH 7.0 and 10.0). The buffers employed were: 50 mM NaCl, 20 mM citrate, pH 4.0 (pKa=
4.45), 50 mM NaCl, 20 mM phosphate, pH 7.0 (pKa = 6.88), and 50 mM NaCl, 20 mM
glycine, pH 10.0 (pKa = 9.78).

Each compound was dissolved in buffer to achieve predetermined concentrations and
pH was adjusted with 1 M HCl or NaOH if necessary. Preliminary studies were conducted to
determine concentrations of the additives to achieve ﬂuorescence loss with heat, but to avoid
complete denaturation of the protein prior to heating. Control solutions were deﬁned as
buffers (pH 4.0, 7.0 or 10.0) containing R-PE to which no additives had been supplemented.
6.3.3. Thermal inactivation.

Capillary tubes (Accu-ﬁll 90 Micropet, Cat. No. 4624, Becton-Dickinson and Co.,
Parsippany, NJ) were ﬁlled with R-PE solution (200 pL) by capillary action. To allow for
even head space at both ends, 50 BL were removed using a micropipet (i.e. the R-PE solution
was centered between the two sealed ends). One end was sealed using a gas/oxygen ﬂame

and the other with Teﬂon tape. The tubes were placed in a wire rack and immersed in a

55

Polystat circulator bath (Model 1268-52, Cole-Partner Instrument Company, Chicago, IL)
connected to a bath programmer (Model 1268-62, Cole-Farmer). Temperature in the capillary
tubes was monitored using a thermocouple (Resistance Temperature Detector, Pt 100, 0.15
cm diameter, (1°C accuracy) inserted in a control capillary tube containing 150 BL of the
same solution. The thermocouple was connected to a Solomat MPM 200 Modumeter
(Solomat Partners LP, Stanford, CT). Zero time was deﬁned as the time when the R-PE
solution reached the target temperature. The heating lag time in the capillary tubes was 10'1
sec.

Temperatures ranged from 40 to 90°C depending on the solution conditions.
Preliminary studies were performed to determine the times and temperature necessary to
achieve at least a one log reduction in ﬂuorescence. The tubes were removed at the speciﬁc
intervals of time and immediately placed in an ice-water bath, then kept at 4°C until
ﬂuorescence was read within 6 h. Preliminary studies were conducted to Show that no
changes of ﬂuorescence occurred during this period. Each study was done in triplicate.
6.3.4. Fluorescence measurement.

R-PE ﬂuorescence at the maximum emission wavelength was measured using a
CytoFluor II Microwell Fluorescence reader and the CytoFluorII software, Version 2.00
(Biosearch Incorporated, Bedford, MA). Heated and unheated R-PE solutions were placed in
96-well ﬂuorescence reader plates (Cat. No. DG5515, Life Science Products Inc., Denver,
CO). Excitation was set at 485/20 nm and emission at 590/35 nm.

6.3.5. Data analysis.
Fluorescence counts were converted to logarithms. D values (in minutes) were

calculated by linear regression of ﬂuorescence vs time using Microsoft Excel Version 5.0a

56

(Microsoft Corporation, Redmond, WA). Thermal resistance curves were determined by
plotting log of D values vs temperature. The 2 values were determined as the negative
reciprocal of the slope of the thermal resistance curve.

Statistical differences among 2 values within treatments were tested using
two-factorial analysis of variance (ANOVA) (Systats Version 5.0, Systats, Evanston, IL).
Mean 2 values were compared using Tukey's test with the mean square error at the 5% level
of probability.

6.4. RESULTS AND DISCUSSION

The minimum temperature needed to achieve signiﬁcant ﬂuorescence loss of R-PE
was greatly inﬂuenced by the solution conditions. Adjustments of heating time and
temperature combinations, as well as of concentration of the additives had to be made for
each treatment. At pH 4.0, the protein was so unstable that addition of either sucrose, SDS,
NaCl or ME resulted in almost complete loss of ﬂuorescence, even at temperatures as low as
40°C.

6.4.1. Effects of sucrose.

At pH 7.0, the heating temperatures to achieve at least a one log reduction in R-PE
ﬂuorescence were lower for the control (60°C-80°C) than for the R-PE solutions containing
sucrose (70°C—90°C), suggesting that sucrose had a thermostable effect on R-PE (Figure 6.1).
The D values of R-PE in the control at pH 7.0 ranged from 2500 min at 60°C to 31.44 min at
80°C (Figure 6.2). In this study, the D value was deﬁned as the time necessary to reduce the
ﬂuorescence intensity of R-PE by 90% at a given temperature. The D value is used as an
indication of the thermal stability of R-PE at a constant temperature. An increase in the D

value of R-PE in sucrose solutions with respect to the control indicates that addition of

57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

20% Sucrose pH 7.0 20% Sucrose pH 10.0
10000 - 10000
é s
3 a
a N“ ~4\‘\+—-~—. O —._ H—Q
° ° 'B\a
5 1000 5 1000
o o
D Q
2 SW +10c 2 +eoc
o +aoc o +esc
E +906 “—3- +700
100 . . v . . . . 100 . .
0 50 100 150 200 250 300 0 20 40 80 80
Time (min) Time (min)
40% Sucrose pH 7.0 40% Sucrose pH 10.0
10000 - 10000
=5 =2
e . e
. \F_ _ . -4 , —:a;:%
U U
r: 1:
1000 _ 1000-
3 B a 3
I a
. +7oc g +606
3 —a——ooc 3 +05:
E ——.—-—SOC 5 +700
100 . . - 100 - r T
O 100 200 300 o 20 4o 00 00
Time (min) Time (min)
60% Sucrose pH 7.0 60% Sucrose pH 10.0
10000 , 10000
a =i
«i 3 29;...
3 «M4 3
c 1000 5 1000 .
8 0
: ._._7oc 3 +606
3 —a—aoc 3 +0»
:1 .—.—90C :1
100 . . . . a . 100 . 4 .
0 so 100 150 200 250 300 ° 2° ‘0 5° 3°
Time (min) Time (min)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.1. Thermal inactivation curves of phycoerytth in sucrose solutions in 50 mM
NaCl, 20 mM phosphate or glycine buffer, pH 7.0 and 10.0, respectively.

58

 

pH 4.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3‘ 10000
2.
3 1000 a s mg
C
3 +4OC
3 100 +soc
'5 +600
2 10 l T I T Fl' 1
1L
0 100 200 300 400 500 600
Time (min)
pH 7.0
3 A
c u u a v a
o
o
8 +600
; +70C
3 +80C
I 10 I Y T Y I'
0 50 100 150 200 250 300
Time (min)
pH 10.0
0 10000
o
c A
3 ’T 1000 v s 4
g 3. —o—-—-6OC
3 3 100 +650
3 +70C
E 10 I 7 f 1’ 7 ﬂ
0 50 100 150 200 250 300
Time (min)

 

 

 

Figure 6.2. Thermal inactivation curves of phycoerythrin in 50 mM, 20 mM citrate buffer,
pH 4.0, 50 mM, 20 mM phosphate buffer, pH 7.0 and 50 mM, 20 mM glycine buffer, pH
10.0.

59

 

sucrose confers stability towards temperature. When compared at the same temperatures
(70°C and 80°C), the D values for R-PE in sucrose solutions at pH 7.0 were much higher
than those of the control. When heated at 70°C, the D values of R-PE increased as sucrose
concentrations were increased to 40%, but the D value in 60% sucrose was lower than that at
40% sucrose. At 80°C, the D values for R-PE increased with increasing concentrations of
sucrose in solution. At 90°C, the D values of R-PE in 20 and 40% sucrose were similar, but
increased by almost one log in 60% sucrose (Table 6.1). It can be suggested that, at lower
temperatures, sucrose enhances protein stability up to a certain concentration, beyond which
sucrose competes with protein for water of solvation. At higher temperatures this was not
seen as sucrose was more readily solubilized.

At pH 10.0, the heating temperatures were the same for all R-PE solutions (60-70°C)
suggesting that the thermostabilizing effect of sucrose was not as strong as pH 7.0 (Figure
6.1). At pH 10.0, all D values increased as the concentration of sucrose was increased when
R-PE was heated at 65°C and 70°C. At 60°C, the D value of R-PE was lower in 20% sucrose
than in the control, but then increased with increasing concentrations Of sucrose (Table 6.2).

Addition of sucrose to samples of R-PE at pH 7.0 before heating reduced ﬂuorescence
intensity by about 17% at 20 and 40% concentration and 20% at 60% sucrose concentration.
This suggested that sucrose had a protective effect on R-PE against thermal inactivation but
in unheated samples the native conformation was altered in the presence of the sugar. When
sucrose was added to R-PE solutions at pH 10.0 before heating, the ﬂuorescence intensity
was decreased by 17, 10 and 3% at 20, 40 and 60% sucrose, respectively. Results suggest that

at this pH, sucrose had a smaller effect on R-PE native structure than at pH 7.0.

60

Table 6.1. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in sucrose solutions in 50 mM NaCl, 20 mM phosphate buffer, pH 7.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sucrose Conc. T(°C) D value (min) X coefficient Constant R
(% w/v) srmared
60 2500 -0.0004 3.68 0.93
O 70 124 -0.03 3.65 0.93
80 31 -0.03 3.19 0.86
70 855 -0.001 3.52 0.87
20 80 179 -0.006 3.19 0.85
90 50 -0.02 3.22 0.95
70 1408 -0.001 3.48 0.68
40 80 313 -0.003 3.26 0.81
90 44 -0.02 3.25 0.94
70 1250 -0.0008 3.47 0.64
60 80 493 -0.002 3.38 0.93
90 369 -0.003 3.29 0.64

 

 

 

 

 

 

Table 6.2. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in sucrose solutions in 50 mM NaCl, 20 mM glycine buffer, pH 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sucrose Conc. T(°C) D value (min) X coefficient Constant R
(% w/v) squared
0 60 348 -0.003 3.52 0.85
65 36 -0.03 3.44 0.93
70 7.2 -0.14 3.27 0.95
60 285 -0.004 3.53 0.90
20 65 153 -0.007 3.47 0.86
70 34 -0.03 3.29 0.90
60 370 -0.003 3.58 0.77
40 65 240 -0.004 3.56 0.94
70 62 -0.02 3.41 0.92
60 885 -0.001 3.54 0.65
60 65 746 -0.001 3.60 0.56
70 106 -0.002 3.49 0.93

 

 

 

 

 

 

61

 

 

Addition of sugars enhances thermostability of proteins in aqueous solutions because
of a "preferential hydration" effect. The sugar molecules interfere with the cohesive force of
water at the water-protein interface and are preferentially excluded from the protein's
surrounding, thus, favoring the native state over the denatured one (Creighton, 1993;
Timasheff and Arakawa, 1997). This effect has been observed in several proteins. DeCordt et
al. (1994) used sugars to modify the thermal inactivation kinetics of (It-amylase from Bacillus
spp. Addition of 20, 40 or 60% sucrose resulted in higher thermal stability with respect to the
control. These researchers found that, in general, protein stability increased as sucrose
concentration was increased from zero to 40%, but addition of 60% sucrose did not confer
higher stability. Boye et al. (1996) studied the thermal stability Of (-lactoglobulin using
differential scanning calorimetry. Addition of either glucose or sucrose (at 50%
concentrations) stabilized partly denatured protein, inhibiting its aggregation. Sucrose
showed a stronger effect than glucose. Raj eshwara and Prakash (1996) observed that addition
Of sucrose (at 25 and 50% concentrations) to heat-treated wheat germ increased the thermal
stability Of the enzyme with respect to the sample in buffer alone.

In our study, the 2 value was deﬁned as the temperature increase necessary to reduce
the D value by 90%. The 2 value is indicative of the temperature dependence of R-PE
denaturation. The 2 values for R-PE in 20 and 40% sucrose at pH 7.0 (16.12°C and 13.27,
respectively) were not different (p>0.05) from the control (10.52°C). The 2 value for the 60%
solution (37.75°C) was higher (p<0.05) than the other solutions, suggesting that the
denaturation of R-PE was less temperature dependent. At pH 10.0, the 2 values of R-PE in
20, 40 and 60% sucrose (10.92°C, 12.87°C and 10.87°C, respectively) did not differ, but

were higher (p<0.05) than that of the control (590°C) (Table 6.3).

62

Table 6.3. The 2 values and regression parameters for thermal inactivation curves of
R-phycoerythrin in sucrose solutions in 50 mM NaCl, 20 mM phosphate buffer, pH 7.0 and
and 50 mM NaCl, 20 mM glycine buffer, 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pH Sucrose 2 value (°C) X coefficient Constant R squared
Conc. (% w/v)

7.0 O 10.53 -0.10 8.98 0.96
20 16.12‘ -0.06 7.26 0.99
40 13.3%l -0.08 8.46 0.99
60 37.8b -0.03 4.90 0.91

10.0 0 5.9a1 -0.17 12.61 0.99
20 10.9b -0.09 8.01 0.95
40 12.9b -0.08 7.30 0.92
60 10.9b -009 8.60 0.81

 

Values with different superscripts within the same pH are statistically different (p<0.05)

Table 6.4. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in sodium-dodecyl sulfate (SDS) solutions in 50 mM NaCl, 20 mM
phosphate buffer, pH 7.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SDS Conc. T(°C) D value X coefﬁcient Constant R squared
(°/o v/v) (min)

60 2500 -0.0004 3.68 0.93

0 70 124 -0.03 3.65 0.93

80 31 -0.03 3.19 0.86

50 87 —0.01 3.37 0.98

0.5 55 36 -0.03 3.09 0.98
60 10 -O.10 3.10 0.95

50 81 -0.01 3.37 0.99

1.0 55 35 -0.03 2.99 0.93
60 9.1 -0.11 3.09 0.95

50 73 -0.01 3.31 0.99

2.5 55 33 -0.03 3.01 0.94
60 9.1 -0.11 3.00 0.92

 

63

 

 

In view of the results, R-PE in 50 mM NaCl, 20 mM glycine buffer, pH 10.0 (control)
should be investigated further for its applicability as a TTI in thermal processing of beef
products. The calculated 2 value (590°C) for R-PE under these conditions fell within the
range of Salmonella 2 values (5.6-6.2°C) reported in the literature (Goodfellow and Brown,
1978; Orta-Ramirez, 1995). It is worth noticing that, although not within the scope of this
study, the calculated 2 values for R-PE in 50 mM NaCl, 20 mM phosphate buffer, pH 7.0
(10.52°C), 20% sucrose, pH 10.0 (10.92°C) and 60% sucrose, pH 10.0 (10.87°C), Should be

investigated further for possible application as a TTI in commercial sterilization processes,

- -1 “mm

where the target microorganism, Clostridium botulinum, has a 2 value of 10°C (Pﬂug and

 

Odlaugh, 1978).
6.4.2. Effects of SDS.

Addition of SDS to R-PE solutions before heating resulted in loss of ﬂuorescence
intensity by 62% at 0.5% concentration and by 75% at 2.5% concentration at pH 10.0. R-PE
ﬂuorescence intensity before heating was reduced by 35% at pH 7.0, regardless of SDS
concentration, indicating that SDS has a strong denaturant effect even without heating.

At pH 7.0 and 10.0, the addition of SDS to R-PE resulted in a lower range of heating
temperatures (SO-60°C and 40-50°C, respectively) necessary to decrease R-PE ﬂuorescence
by a one log, when compared to the controls (60-80°C and 60-70°C, respectively), indicating
that SDS had a therrnodestabiling effect (Figure 6.3). At pH 7.0 and 10.0, the D values at
each temperature were very Similar for all concentrations of SDS, but were much lower than
the control (Table 6.4, Table 6.5). Results suggest that addition of 0.5% SDS decreased
thermal stability of R-PE when compared to the control, and addition of more SDS had little

effect at either pH.

64

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.5% SDS pH 7.0 0.5% SDS pH10.0 1
10000 10000
3 1 z 1" \
E 1000!!\g : 1000 I \.\.\
0 Q 0 i \o\
c \ C ‘0
3 .‘e 8‘11 3 Raiser
3 100 “\\B +soc 3 '00 k +40c
‘5 +650 3 +450
5 +60C E +50c
10 10
0 20 40 so so 0 20 40 so 00
Time (mln) Time (min)
1% SDS pH 7.0 1% SDS pH 10.0
10000 10000
37 ? 4k
5 3 \
01 ‘°°° a 1000i \.
o u
c : ~\\.
0 o
o u
: 100 = 100 X "owioc
'6 3 +tsc
:1 3
IT. I: “OwM‘JC
10 10
o 20 40 so an o 20 40 00 In
Time (min) Tlme(mir1)
2.5% SDS pH 7.0 2.5% SDS [1” 10.0
mm 10000
3 it 3 1: \,
: 1000 \_ : 1000 It \kc‘
u \ \Q\\_ o . ‘C
5 ‘1 ER ""W 5 In“ \K’
0 El\ 0 BE]
2 100 & \EL\\ : 100 ‘0 +40c
5 B ‘5 -—a—45C
:1 =
E E +soc
lo 10
0 20 10 so so 0 20 40 so 00
Time (min) Time (min)

 

 

 

 

 

 

Figure 6.3. Thermal inactivation curves of phycoerythrin in sodium dodecyl sulfate solutions
in 50 mM NaCl, 20 mM phosphate or glycine buffer, pH 7.0 and 10.0, respectively.

65

Table 6.5. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in sodium-dodecyl sulfate (SDS) solutions in 50 mM NaCl, 20 mM glycine
buffer, pH 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SDS Conc. T(°C) D value X coefﬁcient Constant R squared
(% v/v) (min)

0 60 348 -0.003 3.52 0.85

65 36 -0.03 3.44 0.93

70 7.2 -0.14 3.27 0.95

0.5 40 70 -0.01 3.41 0.96
45 24 -0.04 2.90 0.96

50 10 -0.10 2.75 0.83

1.0 40 71 -0.01 3.39 0.96
45 24 -0.04 2.94 0.96

50 10 -0.10 2.80 0.88

2.5 40 66 -0.02 3.41 0.97
45 24 -0.04 2.99 0.98

50 9.5 -0.11 2.89 0.91

 

Table 6.6. The 2 values and regression parameters for thermal inactivation curves of
R-phycoerythrin in sodium-dodecyl sulfate (SDS) solutions in 50 mM NaCl, 20 mM
phosphate buffer, pH 7.0 and 50 mM NaCl, 20 mM glycine buffer, 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pH SDS Conc. 2 value X coefficient Constant R
(% v/v) (°C) squared

7.0 0 10.53 -0.10 8.98 0.96
0.5 14.021 0.07 7.55 0.96
1.0 15.93 -0.06 6.89 0.98
2.5 21.83 -0.05 5.48 0.99

10.0 0 5.9a -0.17 12.61 0.99
0.5 20.5b -0.05 5.54 0.96
1.0 26.1c -0.04 4.82 0.93
2.5 33.4bc -0.03 4.18 0.97

 

'Values with different superscripts within the same pH are statistically different (p<0.05)

66

 

 

SDS is a well known protein denaturant used in electrophoretic separations of
proteins (Laemli, 1970). As many other detergents, SDS interacts with the nonpolar amino
acid residues of the polypeptide chain interfering with the hydrophobic forces that maintain
the native structure (Voet and Voet, 1990; Creighton, 1993). The chromophores in R-PE are
located within hydrophobic regions of the protein molecule (O'hEocha, 1965; F uj imori and
Pecci, 1967). The unique spectral properties of R-PE depend on the native structure of the
polypeptide and interaction among the different chromophores (O'hEocha and O'Carra, 1961;
Teale and Dale, 1970; MacColl and Guard-Friar, 1987). Alteration of the hydrophobic
interactions that maintain the chromophore native environment will result in loss of the
spectroscopical properties (i.e. absorbance and ﬂuorescence). In addition, SDS may cause
dissociation of subunits which could also result in loss of color and ﬂuorescence. Moreover,
denaturation of the molecule can cause cyclization of the chromophores resulting in a
decrease of spectral properties (Rudiger, 1992).

Treatment of R-PE with 1% SDS for 3 h at 37°C resulted in loss of ﬂuorescence
(Gantt, 1969). The color of the solution changed from pink to purple. Absorbance
measurements revealed the disappearance of the peak at 563 nm. The author concluded that
addition of SDS caused dissociation of the protein into monomers, which resulted in changes
of the optical properties. MacColl and Guard-Friar (1987) reported a decrease in absorption
of C-phycocyanin, another phycobiliprotein, upon addition of SDS.

No differences (p>0.05) were found in 2 values when R-PE was heated in 0.5, 1.0 and
2.5% SDS (13.98°C, 15.85°C and 21 .79°C, respectively) and the control (10.52°C) at pH 7.0
(Table 6.6), indicating that SDS did not inﬂuence the temperature dependence of the

inactivation rate.

67

At pH 10.0, the 2 value of the control (590°C) was lower (p<0.05) than that for R-PE
in SDS solutions. The 2 value for R-PE in 0.5 (20.52°C) and 1.0% (26.08°C) solutions were
different (p<0.05) from each other, but not different from 2.5% SDS (33.38°C) (Table 6.6).
Large 2 values indicate that a large increase in temperature is needed to reduce the D values
by 90%. The large 2 values observed when R-PE was heated in SDS solutions indicate that
the denaturation of R-PE was not highly temperature dependent.

The 2 values of PE in 2.5 % SDS, pH 7.0 (21.79°C), and 0.5 %, 1.0 and 2.5% SDS,
pH 1.0 (20.52, 26.08 and 33.38°C, respectively) should be explored further for its possible
application in monitoring destruction of vitamins A (z=23-25°C), Bl (z=22.0-31.3°C) and
B6 (z=22.0-31.0°C) during thermal processing of several food products (Holdsworth, 1997).
6.4.3. Effects of NaCl.

Addition of NaCl to R-PE before heating reduced ﬂuorescence intensity by about
25% at pH 7.0, and by 33% at pH 10.0, regardless of the concentration, suggesting that NaCl
had a destabilizing effect on R-PE native structure.

At pH 7.0, the range of heating temperatures used in the inactivation experiments
(60°C -70°C) was lower than that of the control (60°C -80°C), suggesting that NaCl had a
therrnodestabilizing effect on R-PE (Figure 6.4). Preliminary studies had been conducted to
determine the maximum temperature at which the R-PE ﬂuorescence would not decrease
markedly during the heating lag time. For R-PE in NaCl solutions at pH 7.0 the maximum
temperature was 70°C. At higher temperatures, R-PE ﬂuorescence was lost during the
heating lag period. However, even after heating for 225 min, at 60, 65 and 70°C, no log
reduction in ﬂuorescence was achieved. The D values were then estimated from the

ﬂuorescence reduction during this time period. The D values for R-PE at 60°C were lower,

68

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M NaCl pH 7.0
10000
=1
" Harm?“
v ﬂ._ ‘\
o
u
c
9 1000
o
O
0
I-
o
2
IL
too
0 50 100 150 200 250
Time (min)
2M NaCi pH 7.0
10000 ._____,.__ -2 .__
‘7 i
=!
. ,._n
. 3W H“. i
0
° I
5 1000 ,
U l
I
0 —O——SOC
I-
0 +65C ‘
:i .
E +70c i
100
0 50 100 150 200 250
Time (min)
4M NaCi pH 7.0
10000
‘7
3.
5 “W H
3 ‘R
E 1000
o
‘ I
Q .. — SOC
I-
O a— sec
2 :
IL . 706 i
100
0 50 100 ‘50 200 250
Time (min)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.5M NaCI pH 10.0
10000
=2
3 at
‘\
8 ‘i\\““~+ ..... +
c \ B. k“.
0 1000 ‘ B“
a 13.-.]
2 .o. 7+55c
O ﬁn... sec
3
LI. +650
too
0 20 40 60 00 100 120
Time (min)
1M NaCI pH 10.0
10000
=3
3
3 ~ *
\ a
5 EL N‘ we
3 RE
0 ~._550
O —-a-— 600
:i
E 47065::
I00
0 20 40 60 lo 100 120
Time (min)
2M NaCI pH 10.0
10000
=2
5 in
E if\\\e\_‘\
o 1000 \ h \‘9x—4
3 \ \‘EL\
2 E \f] +550
0 +600
3 O
E .4»~650
‘00
0 20 40 60 00 100 120
Time (min)

 

 

Figure 6.4. Thermal inactivation curves of phycoerythrin in NaCl solutions in 50 mM NaCl,
20 mM phosphate or glycine buffer, pH 7.0 and 10.0, respectively.

69

but at 70°C were higher than those of the control, for all the NaCl concentrations. No
differences in D values for R-PE were observed with increasing concentrations of NaCl
(Table 6.7).

At pH 10.0, the range of heating temperatures (55°C -65°C) was also lower than that
of the control (60°C -70°C) (Figure 6.4). Overall, the D values at the same temperature
decreased with increasing NaCl concentration (Table 6.8). At this pH, the protein was already
destabilized by pH, and was more sensitive to presence of ions in the solution. In addition, at
this pH, the protein net charge is markedly negative (pI of R-PE = 4.1-4.2). Interactions with
the Na+ in solution may interfere with charge-charge interactions within the protein structure
resulting in further unfolding.

Addition of salts to proteins in aqueous solutions can either stabilize or destabilize
their native structure (Voet and Voet, 1990; Creighton, 1993; Timasheff and Arakawa, 1997).
During thermal denaturation, partial unfolding of the protein occurs resulting in the exposure
of amino acid side chains (Klibanov, 1983; Creighton, 1993). This exposure facilitates
interactions with the free ions in solution. Folawiyo and Owusu-Apenten (1996) studied the
heat stability of cruciferin under different NaCl concentrations. Raising the concentration of
NaCl from 0.1 to 1M, had a stabilization effect at temperatures of 70-80°C, but not at
temperatures below 50°C. The Tm (temperature at the midpoint of the thermal unfolding
transition) also increased linearly as a function of NaCl concentration. The authors concluded
that the NaCl stabilization effect could be related to non-speciﬁc salt effects by interfering
with charge-charge interactions within the native conformation of cruciferin.

At pH 7.0, no differences (p>0.05) were found between the 2 values of control

(10.52°C) and any NaCl solutions (15.28, 16.72 and 994°C, respectively), suggesting that

70

Table 6.7. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in NaCl solutions in 20 mM phosphate buffer, pH 7.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N aCl T D value X coefﬁcient Constant R
Conc. (M) (°C) grin) squared
0.05 60 2500 -0.0004 3.68 0.93
70 123 -0.03 3.65 0.93
80 31 -0.03 3.19 0.86
1.05 60 990 -0.001 3.56 0.74
65 806 -0.001 3.50 0.73
70 219 -0.005 3.53 0.90
2.05 60 1818 -0.0006 3.54 0.89
65 559 -0.002 3.50 0.92
70 459 -0.002 3.51 0.30
4.05 60 1 190 -0.0008 3.49 0.90
65 725 -0.001 3.43 0.84
70 1 17 -0.009 3.49 0.92
Table 6.8. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in NaCl solutions in 20 mM glycine buffer, pH 10.0
NaCl T D value X coefﬁcient Constant R
Conc. (M) (°C) (min) squared
0.05 60 348 -0.003 3.52 0.85
65 36 -0.03 3.44 0.93
70 7.2 -0.14 3.27 0.95
0.55 55 308 -0.003 3.38 0.87
60 109 -0.009 3.32 0.88
65 31 -0.03 3.11 0.84
1.05 55 245 -0.004 3.41 0.95
60 101 -0.01 3.31 0.90
65 27 -0.04 3.15 0.90
2.05 55 251 -0.004 3.36 0.91
60 86 -0.01 3.26 0.90
65 24 -0.04 3.07 0.87

 

 

 

 

 

 

 

71

 

 

addition of NaCl to R-PE solutions did not affect the temperature dependence of R-PE
denaturation. On the other hand, no differences (p>0.05) were found in 2 values of R-PE
among the three NaCl solutions (10.09, 10.47 and 978°C) at pH 10.0, and they all had
greater 2 values than the control (p<0.05) (Table 6.9). The calculated 2 values for R-PE in
NaCl at pH 10.0 could be further investigated for their possible application as a TTI for food
sterilization processes (2 value of C. botulinum = 10°C).

6.4.4. Effects of urea.

Urea was the only additive in which R-PE ﬂuorescence was not completely lost
before heating at pH 4.0. Addition of urea to samples before heating reduced ﬂuorescence
intensity of R-PE by 17% at pH 4.0, regardless of the concentration. At pH 7 .0, addition of
urea at 1 and 2 M concentration did not affect R-PE ﬂuorescence intensity before heating, but
at 4M concentration resulted in a decrease of R-PE ﬂuorescence by about 25%. When urea
was added to unheated samples at pH 10.0 at 0.5 and 1 M concentrations, it did not reduce
ﬂuorescence intensity of R-PE. At 2 M concentration, addition of urea at pH 10.0 decreased
initial ﬂuorescence intensity of R-PE by about 50%. These observations suggest that the
denaturing effect of urea was enhanced by heating at lower concentrations, but at 4 M for pH
7.0 and 2 M for pH 10.0 there was a noticeable effect even before heating. Addition of urea
at pH 4.0 allowed for a higher range of heating temperatures (55°C - 65°C) than that for the
control (40°C - 60°C), suggesting that urea had a thermostabilizing effect at this pH (Figure
6.5). At 60°C, the D value for R-PE in urea solutions was lower than that of the control. The
D values did not change as concentration of urea was increased (Table 6.10). The D values
calculated at pH 4.0, however, should be considered carefully. Upon examination of Figure

6.5 it should be noted that values of ﬂuorescence at time zero for pH 4.0 solutions

72

Table 6.9. The 2 values and regression parameters for thermal inactivation curves of

R-phycoerythrin in NaCl solutions in 20 mM phosphate buffer, pH 7.0 and 50 mM NaCl, 20

mM glycine buffer, pH 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pH NaCl Cone. 2 value X coefﬁcient Constant R
(M) (°C) squared

7.0 0.05 10.5“ -0.1 8.98 0.96
1.05 15.3“ -0.07 7.00 0.85
2.05 16.7“ -0.06 6.78 0.85
4.05 9.9“ -0.1 9.21 0.90

10.0 0.05 5.9“ -0.2 12.61 0.99
0.55 10.1b -0.1 7.95 0.99
1.05 10.5b —0.1 7.67 0.99
2.05 9.8“ 0.1 8.04 0.99

 

Values with different superscripts within the same pH are statistically different (p<0.05)

Table 6.10. D values and regression parameters for thermal inactivation curves of R-

phycoerythrin in urea solutions in 50 mM NaCl, 20 mM citrate buffer, pH 4.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Urea Conc. T D value X coefﬁcient Constant R

(M) (°C) (min) squared

0 40 5000 -0.0002 3.04 0.90

50 2500 -0.0004 3.04 0.96

60 270 -0.004 2.70 0.76

0.1 55 222 -0.005 2.60 0.72

60 40 -0.03 2.98 0.94

65 7.5 -0.1 2.91 0.99

0.25 55 169 -0.006 2.74 0.96

60 35 -0.03 3.12 0.94

65 7.6 -0.1 2.99 0.89

0.50 55 235 -0.004 2.92 0.94

60 36 -0.03 3.18 0.91

65 5.9 -0.2 2.81 0.97

 

73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.5M Urea pH 4.0 1M Urea pH 7,0 0.5M Urea pH 10.0
10000 , 10000 10000 ,
3" =3 =3
5' 3 3
0 8 3 a
g C 1:
o 1000 o 1000 1 q, 1000
u 0 o
10 W a
e 2 a
3 o “*— °°° o +60C
: a
2 E +65C E E 550
u. “‘9‘“ 7°C + 700
100 . . 1oo . . .
100
0 100 200 300 0 20 40 so no
0 2O 40 00
Time (min) Time (min) Time (min)
0.25M Urea pH 4.0 2M Urea pH 7.0 1M Urea pH 10.0
10000 , 10000 10000 ,
+550 A
‘3’ —a—soc :i H :3
2., Mom. 65C 3 y. 3
o 3 8
o
c
5 1000 0 1000 1 s 1000 “‘0
u ° 0
a " "
e 2 2
3 0‘“ 3 +500 0 ——o—soc
3 E +650 2 +050
u u.
-4...— 70C +70C
100 100 . j 100 . . . .
0 2° 40 6° 0 100 200 300 o 20 40 00 so
Time (min) Time (min) Time (m in)
0.5M Urea pH 4.0 MI Urea pH 7.0 2" Urea pH10.0
1oooo , 10000 7 10000 7
+550
’3? +600 3 3'
3; +656 3 3
e o
3 3 e 3
5 1000 3 1000 1 \O o 1000
3 H-* 2 2
3 0 +6“ 0 +600
3 ”3- ~----a»—— 65C 5 65C
11.
+7“: +700
100 0 . 100 . . r
100 . f .
o 100 200 300 o 10 20 so
0 20 40 50
Time (min) Time (min) Time (min)

 

 

Figure 6.5. Thermal inactivation curves of phycoerythrin in urea solutions in 50 mM NaCl,
20 mM citrate, phosphate or glycine buffer, pH 4.0, 7.0 and 10.0, respectively.

74

 

 

 

are very low compared to the zero values under other conditions. It can be seen that the
protein is undergoing structural changes during the heating lag time resulting in a signiﬁcant
loss of ﬂuorescence. Thus, the calculated D values are overestimating the length of time to
reduce the ﬂuorescence by one log.

In contrast to the effect of sucrose, SDS, NaCl and MB on R-PE at pH 4.0, the
combination of urea and pH 4.0 did not cause complete loss of R-PE ﬂuorescence before
heating. This could be the result of an increase of solubility of the protein in the urea
solutions. At pH 4.0, R-PE is at pH near the isoelectric point (p1 = 4.1-4.2). Proteins at their
pl often have low solubility because the molecule has no net charge, and tend to form
aggregates (Voet and Voet, 1990; Creighton, 1993). Addition of urea at pH 4.0 may have
help to partially solubilize the protein inhibiting aggregation and allowing the protein to
adopt a more native-like conformation.

At pH 7.0, the range of heating temperatures was lower for R-PE in urea solutions
(60-70°C) than for the control (60-80°C), indicating that urea had a thennodestabiling effect
(Figure 6.5). At all temperatures, the D values for the control were greater than the D values
of R-PE in urea solutions at both pH 7.0 and 10.0. In addition, D values of R-PE at pH 7.0
decreased with increasing concentrations of urea at all temperatures (Table 6.11). At pH
10.0,the range of heating temperatures was same for all solutions (60-70°C)(Figure 6.5). All
D values for R-PE in the urea solutions were lower when compared with the control at pH
10.0. D values decreased as the concentration of urea was increased (Table 6.12).

When comparing 2 values, no differences (p>0.05) were found among R-PE in urea
solutions (6.79, 7.43 and 623°C) and control (10.70°C) at pH 4.0, indicating that addition of

urea at this pH does not affect the thermal resistance of R-PE. At pH 7.0, the 2 values of

75

Table 6.11. D values and regression parameters for thermal inactivation curves of
R—phycoerythrin in urea solutions in 50 mM NaCl, 20 mM phosphate buffer, pH 7.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Urea Cone. T D value X coefficient Constant R

(M) (°C) (min) squared

0 60 2500 -0.0004 3.68 0.93

70 124 —0.03 3.65 0.93

80 31 -0.03 3.19 0.86

1.0 60 1515 -0.0007 3.69 0.92

65 286 -0.003 3.74 0.77

70 40 -0.03 3.59 0.74

2.0 60 775 -0.001 3.68 0.94

65 146 -0.007 3.70 0.94

70 23 -0.04 3.74 0.99

4.0 60 357 -0.002 3.47 0.92

65 62 -0.02 3.37 0.99

70 9.2 -O.1 3.41 0.78

 

 

Table 6.12. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in urea solutions in 50 mM NaCl, 20 mM glycine buffer, pH 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Urea Cone. T D value X coefﬁcient Constant R

(M) (°C) (min) squared

0 60 348 -0.003 3.52 0.85

65 36 -0.03 3.44 0.93

70 7.2 -O.l 3.27 0.95

0.5 60 125 -0.008 3.58 0.91

65 18 -0.05 3.45 0.84

70 4.0 -0.3 3.21 0.81

1.0 60 114 -0.009 3.53 0.89

65 11 -0.09 3.35 0.96

70 2.5 -0.4 3.16 0.98

2.0 60 38 -0.03 3.50 0.91

65 13 -0.08 3.05 0.94

70 2.1 -0.5 2.84 0.99

 

 

 

 

R-PE in solutions containing urea (6.33, 6.56 and 630°C) did not differ (p>0.05), but were
lower (p<0.05) than that of the control (10.52°C). These results suggest that urea increases
the temperature dependence of R-PE denaturation at pH 7.0 At pH 10.0, the 2 value of the
control (590°C) and R-PE in 1 M urea solution (602°C) were not different (p>0.05). In
addition, 2 values of R-PE in 0.5 (666°C) and 2 M (798°C) solutions were not different
(p>0.05), but these latter ones were higher (p<0.05) from the former ones, suggesting that
addition of 0.5 and 2 M urea decreased the temperature dependence of R-PE denaturation at
pH 10.0 (Table 6.13). The 2 value of R-PE in 1 M urea at pH 10.0 (602°C) should be further
investigated for its possible application as a TTI to predict destruction of Salmonella in beef
products (250mm,, = 5.6-6.2°C)

Urea is a very common protein denaturant. It can reduce the magnitude of
hydrophobic bonding by up to one-third causing an increase in protein solubility (Creighton,
1993). Urea also affects the structure of water because of its hydrogen-bonding capabilities
(Voet and Voet, 1990; Creighton, 1993). In addition, urea molecules can penetrate the inside
of the proteins disrupting the packed interior of the polypeptide (Creighton, 1993). Addition
of urea resulted in a decrease of thermal stability of R-PE at pH 7.0 and pH 10.0. By
promoting the unfolding of R-PE molecules, the chromophores that, under normal conditions
are closely packed in the hydrophobic interior of the molecule, are exposed to the
environment resulting in a decrease of ﬂuorescence (O'hEocha, 1965; Fujimori and Pecci,
1967). In addition, denaturation of the protein molecule may promote cyclization of the
chromophores, which results in reduced ﬂuorescence intensity (Burke et al., 1972; Rudiger,
1992). Jones and Fujimori (1961) reported reduction in R-PE absorbance at 540 and 565 nm

with addition of 6 M urea, although this effect was mild when compared to treatment with

77

Table 6.13. The 2 values and regression parameters for thermal inactivation curves of
R-phycoerythn'n in urea solutions in 50 mM NaCl, 20 mM citrate buffer, 4.0, 50 mM NaCl,
20 mM phosphate buffer, pH 7.0 and 50 mM NaCl, 20 mM glycine buffer, 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pH Urea Conc. 2 value X coefficient Constant R

(M) (°C) squared

4.0 0 10.7“ -0.1 8.12 0.98

0.1 6.8“ -0.2 10.44 0.99

0.25 7.4“ -O.1 9.62 0.99

0.5 6.2“ -O.2 11.19 0.99

7.0 0 10.5“ -0.1 8.98 0.96

1 6.3b -02 12.68 0.99

2 6.6b -0.2 12.04 0.99

4 6.3b .02 12.09 0.99

10.0 0 5.9“ -0.2 12.61 0.99

0.5 6.7b -0.2 11.07 0.99

1 6.0“ -O.2 11.96 0.98

2 8.0“ -0.1 9.16 0.98

 

 

 

 

Values with different superscripts within the same pH are statistically different (p<0.05)

Table 6.14. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in B-mercaptoethanol (ME) solutions in 50 mM NaCl, 20 mM phosphate
buffer, pH 7.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ME Conc. T D value X coefficient Constant R
(”/0 v/v) (°C) (min) squared

0 60 2500 -0.0004 3.68 0.93

70 124 -0.03 3.65 0.93

80 31.44 -0.03 3.19 0.86

70 385 -0.003 3.67 0.82

0.5 80 51 -0.02 3.18 0.86
85 36 -0.03 3.40 0.95

70 315 -0.003 3.63 0.87

1.0 80 57 -0.02 3.15 0.85
85 38 -0.03 3.46 0.95

70 183 -0.005 3.54 0.94

2.5 80 68 -0.01 3.19 0.91
85 37 -0.03 3.47 0.98

 

 

78

 

4 M guanidine HCl. O'hEocha and O'Carra (1961) reported ahnost complete quenching of
R-PE ﬂuorescence in the presence of 8 M urea for 24 h at 3-5°C, while phycocyanins lost
ﬂuorescence immediately under the same conditions.

6.4.5. Effects of B-mercaptoethanol (ME).

Addition of ME to samples before heating increased the ﬂuorescence intensity of
R-PE by 5 and 8% at pH 7.0 and 10.0, respectively, regardless of the concentration,
suggesting that ME has a protective effect on the native conformation of R-PE. The range of
heating temperatures at pH 7.0 was slightly higher for R-PE in ME-containing solutions
(70°C-85°C) than the control (60°C-80°C), suggesting that ME had a thermostabilizing effect
on R-PE (Figure 6.6). All D values were higher for the three ME concentrations when
compared to those of the control at the same temperature. At 70°C, the D values for R-PE at
pH 7.0 decreased with increasing concentration of ME, while at 80°C the opposite
relationship was observed. D values were not affected by ME concentration at 85°C (Table
6.14).

At pH 10.0, the heating temperature range for R-PE in solutions containing ME
(70-85°C) was higher than the control (60-70°C), indicating that ME had a thermostabilizing
effect on R-PE at this pH (Figure 6.6). All D values were higher for R-PE in ME solutions
when compared to the control. Overall, the D values did not change as concentration of ME
was increased (Table 6.15).

The 2 value of R-PE at pH 7.0 (10.52°C) was not affected (p>0.05) by addition of 0.5

or 1% MB (10.71 and 10.51°C, respectively). The 2 value of R-PE in 2.5% MB (11.04°C) at

79

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.5% B-ME pH 7.0 0.5% B-ME pH 10.0
10000 10000
,7 .7 «.3- 700
z 0 \K :2 0\ +000
v I *3... - 4 5 [a \ 050
0 o l \. °
2 .. x-
o 1000 g 1000 \\‘
2 a ,4... . a .. \«\
a \ .8750(; l s ‘
O B\ o
3 x e . 350 H 3
E s r E
100 l 100
0 20 40 60 50 100 o 50 100 150
Time (min) Time (min)
1% B-ME pH 7.0 1% B-ME pH 10.0
10000 1 10000
a: ‘ :i ‘ «“400
u a +000
. E * “.7 I o 9 +350
0 o
5 1°00 \ g 1000
3 +700 3
3 3,. soc :
O O
+7550
5 m l e
100 i 100
0 20 40 so 00 100 0 50 100 150
Time (min) Tlm- (min)
2.5% B-ME pH 7.0 2.5% B-ME pH 10.0
10000 10000
.g ‘3‘ +700
10' I d wEiwewc
3.. * v
o E \O“ 3 +050
2 *w 2 \
8 1000 \ 8 \O\
o E\\B- +100 0 \‘\
E “\B.\ 771377.000 5 \.
2 ﬂ 4“,”; 3
1.1. lb
‘00 100
o 20 40 so 50 100 o 50 100 150
Time (min) Time (min)

 

 

 

 

 

Figure 6.6. Thermal inactivation curves of phycoerythrin in B-mercaptoethanol solutions in
50 mM NaCl, 20 mM phosphate and glycine buffer, pH 7.0 and 10.0, respectively.

80

Table 6.15. D values and regression parameters for thermal inactivation curves of
R-phycoerythrin in B-mercaptoethanol (ME) solutions in 50 mM NaCl, 20 mM glycine

buffer, pH 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ME Conc. T D value X coefficient Constant R
(% v/v) (°CL min) squared

0 60 348 -0.003 3.52 0.85

65 36 -0.03 3.44 0.93

70 7.2 -0.1 3.27 0.95

70 127 -0.008 3.54 0.91

0.5 80 54 -0.02 3.41 0.94
85 22 -0.05 3.30 0.98

70 127 -0.008 3.55 0.93

1.0 80 69 -0.02 3.40 0.89
85 31 -0.03 3.30 0.94

70 117 -0.009 3.58 0.93

2.5 80 67 -0.02 3.47 0.94
85 40 -0.03 3.31 0.98

 

Table 6.16. The 2 values and regression parameters for thermal inactivation curves of

R-phycoerythrin in B-mercaptoethanol (ME) solutions in 50 mM NaCl, 20 mM phosphate

buffer, pH 7.0 and 50 mM NaCl, 20 mM glycine buffer, 10.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

pH ME Conc. 2 value (°C) X coefficient Constant R
(%) squared
7.0 0 10.5a -0.1 8.98 0.96
0.5 10.72‘ -0.1 6.64 0.99
1.0 10.5ab -01 6.71 0.98
2.5 11.0b -0.1 6.43 0.98
10.0 0 5.921 -0.2 12.61 0.99
0.5 12.0b -0.1 5.17 0.99
1.0 11.6bc .01 5.28 0.99
2.5 11.8c -0.1 5.19 0.99

 

Values with different superscripts within the same pH are statistically different (p<0.05)

81

 

 

 

 

C31

dis

pH 7.0 was higher (p<0.05) than the control and 0.5% ME. At pH 10.0, the 2 value of the
control (590°C) was lower (p<0.05) than all ME solutions (12, 11.62 and 11.84°C,
respectively) (Table 6.16). No differences were found (p>0.05) between the 2 value of R—PE
in 1% ME and 0.5% or 2.5% ME at pH 10.0. These results suggest that at pH 7.0 only 2.5%
ME concentration affected the temperature dependence of R-PE denaturation, but addition of
ME at 0.5% concentration or above increased the thermal resistance of R-PE at pH 10.0. The
2 values of R-PE in 0.5 and 1% ME at pH 7.0 could be explored further for its possible use as
TTIS in the area of sterilization of canned foods (2C bommm = 10°C).

In our experiments, the addition of ME had a marked protective effect on R-PE's
ﬂuorescence at both pHs when compared to the control. At this point, no speciﬁc conclusions
can be made with respect of the mechanism of action for this effect since no evidence of
disulﬂde bonds in the native structure have been observed (Sidler et al., 1989; Ficner et al.,
1992). We can speculate that addition of ME may have prevented photobleaching of the
chromophores. In addition, ME, being a reducing agent, may protect the protein against
oxidation. Jones and Fujimori (1961) reported that the 565 nm absorption peak of R-PE was
highly unstable in the presence of 1 % hydrogen peroxide. The authors concluded that the
chromophore and apoprotein were susceptible to changes in oxidation-reduction potential.
6.4.6. Implications for food processing.

As mentioned before, a reliable TTI should show the same time-temperature
dependent response as that of the target attribute when the temperature is the only
rate-determining factor. This happens when the 2 value (E A) of both TTI and target attribute
are the same (Hendrickx et al., 1995). Therefore, appropriate selection of TTI and target

attribute is of utmost importance. Deviations between 2 values (E A) of TTI and target

82

attribute can result in over- or underestimation of the actual lethality of the process.
According to Van Loey et al. (1995b), a TTI with a 2 value larger than the one for the target
attribute will result in underestimation of the process lethality if the processing temperatures
are above the reference temperature. The process lethality will also be underestimated if the 2
value of the TTI is smaller than the one for the target and the processing temperature is below
the reference temperature. However, if the 2 value for TTI is smaller than the one for the
target attribute, the process lethality will be overestimated when the processing temperature
is above the reference temperature. It will also be overestimated if the 2 value for TTI is
larger than the one for the target and the processing temperature is below the reference
temperature (since 2 value and E A are inversely proportional, the complete opposite will be
applied for the Arrhenius model) (Table 6.17). This phenomenon has also been described by
Pﬂug (1997) and Incze et al. (1999). In practice, however, and as seen in certain foodborne
outbreaks, overestimation of the process lethality is much more dangerous than
underestimation.

Table 6.17. Effects of the holding temperature and 2 value (E A) of a time-temperature
integrator on the actual processing value reported (Source: Van Loey et al., 1995b)

 

 

 

 

 

 

 

 

 

 

 

TDT'mOdel zl'l'l<z’l‘arget zrri>zrarget
TTI Target TTI Target
TH<Tref FTref< F Tref TTIFTrei> T hf
TTI Target arget
TH > Tref FTref> FTref FTref< FTref
Arrhenius-model E m1<E “m E ATTI>E Am
TTI Target TTI Target
TH<rrd TTIFTmiT FTmf TTIFIIEE T FTL
arget arget
TH > Tm FTrcf< F Tref F Tref> FTI2[_

 

TH = holding temperature, Tref = reference temperature, TTIFTref ﬁarocess lethality calculated from the TTI,
TargetFTref =actual process lethality observed in the target attribute, TTIFTref<TargetFTref =underestimation,
TTIFTref>TargetFTref =overestimation.

83

6.5. CONCLUSIONS

The thermal inactivation parameters (D and 2 values) of R-phycoerythrin as measured
by ﬂuorescence loss were modiﬁed by addition of sugars or denaturants and compared with
control samples where no chemicals had been added. Overall, sucrose and ME had a
stabilizing effect, while SDS, NaCl and urea caused protein denaturation. The conformation
of the protein in solution played a major part in the denaturation. At pH 4.0, near its pI, R-PE
exists in the form of aggregates; at pH 7.0, it has a compact, native conformation, with its
chomophores tightly packed in the hydrophobic interior, and at pH 10.0, the protein is in a
partially unfolded form (Ogawa et al., 1991). Addition of sucrose, SDS, NaCl or ME at pH
4.0 resulted in complete loss of R-PE ﬂuorescence before heating. Comparatively, at pH
10.0, the protein, being already partially unfolded was more sensitive to addition of the
cosolvents than at pH 7.0, where the protein exists in its native conformation.

These results are of particular signiﬁcance in the area of protein-based
time-temperature integrators for monitoring thermal processes in meat products. By
modifying the 2 value of R-PE, it can be adjusted to that of a target microorganism found in
meats. Reported 2 values for Salmonella, the target microorganism in cooked beef products,
ranged from 56°C (Goodfellow and Brown, 1978) to 62°C (Orta-Ramirez et al., 1996). The
2 value obtained for R—PE in 50 mM NaCl, 20 mM glycine buffer, pH 10.0 was 590°C,
which falls in the middle of this range. The development of a TTI with R-PE as the detector

component could be used to monitor thermal processes in beef products.

84

CHAPTER 7: Validation of R-phycoerythrin as a time-temperature integrator to
monitor thermal processes in beef products

7.1. ABSTRACT

Time-temperature integrators (TTIS) can be used to rapidly and accurately verify
thermal processing of foods because they respond predictably to thermal history. When they
function properly, TTIs predict the same lethality as that of a target microorganism or quality
attribute of interest, when the two systems are subjected to the same thermal process.
Salmonella was identiﬁed as the target microorganism in this study. Linear and non-linear
models were used to ﬁt experimental heat inactivation data of the protein R-phycoerythrin. D
and 2 values were calculated on the basis of ﬂuorescence loss using linear regression
analysis. Kinetic parameters (11, k and E A) were estimated using non-linear regression
analysis. The reaction orders (n) calculated for R-PE and S. senftenberg were 0.55 and 1.2,
respectively. The calculated E A for R-PE and S. senftenberg were 347.4 and 220.84 kJ mole",
respectively. The lethalities calculated for R-PE and S. senftenberg assuming non-linear
kinetics were highly predictable (R2 of 0.98 for R-PE and 0.93 for S. senftenberg). Results
strongly suggest that lethality of Salmonella can be predicted from R-PE ﬂuorescence loss.
7.2. INTRODUCTION

Thermal processing is the most common method used for food preparation and
preservation. The principal effects of the heat treatment are: a) modiﬁcation of the

organoleptic properties to give the desirable characteristics that will make the product

85

acceptable to the consumer; b) inactivation of enzymes and spoilage microorganisms, with
the consequent increase in shelf-life of the product; and, c) destruction of pathogenic bacteria
to ensure the safety of the cooked product. Mathematically, the impact of a thermal process
can be expressed in terms of D and 2 values (Singh and Heldman, 1993; Hendrickx et al.,
1995). The D value, or thermal reduction time, is the time in minutes required to reduce a
microbial population by 90% or one log cycle. The 2 value is deﬁned as the temperature
increase necessary to reduce a D value by 90% or 1 log. The determination of the D and 2
values for a particular microorganism is very useful in designing a thermal process that
destroys that speciﬁc microorganism. However, even in products that are thermally
processed, contamination can be a problem. Inadequate cooking and improper storage and
holding temperatures are the most common errors that lead to foodborne outbreaks (Todd,
1989)

Recently, time-temperature integrators (TTIs) have been introduced as an alternative
for the monitoring of thermal processes. By deﬁnition, a TTI is a small device that responds
to time-temperature history by undergoing an irreversible precisely measurable change in a
manner that mimics the changes of a target attribute exposed to the same thermal history
(Hendrickx et al., 1995). An important advantage of TTIS is that they allow fast and reliable
validation of a process without the need for detailed information on the actual
time-temperature proﬁle within the product.

The United States Department of Agriculture Food Safety Inspection Service
(USDA-F SIS) has recently proposed a modiﬁcation of the current thermal processing
regulations for meat products based on destruction of the pathogenic microorganism

Salmonella expressed as log reduction. The USDA requires that manufacturers use a

86

combination of thermal and non-thermal processes to achieve a 6.5-log Salmonella reduction
in ready-to-eat, cooked beef, roasted beef and cooked corned beef and a 7-log reduction in
ready-to-eat poultry products. A 5-log reduction in ready-to-eat, cooked uncured meat patties
has been proposed, but this may be increased (Federal Register, 1999a). The development of
a TTI with the same 2 value as Salmonella in beef and poultry products could be used to
determine the adequacy of any log reduction process.

The main objective of this project was to evaluate the applicability of a ﬂuorescent
protein, R-phycoerythrin, to be used as a TTI for predicting Salmonella destruction in cooked
beef products.

7.2.1. Theoretical considerations.

The thermal death time (F value) is the time required to achieve a particular reduction
in a target attribute (quality or safety index) value (Hendrickx et al., 1995). The F value can
be computed from:

F = D, log (Co/C) (1)
where F has units of time, DT is the decimal reduction time at a given temperature T, and C0
and C are the initial and ﬁnal values, respectively, of the target attribute. Thus, C0 and C may
refer to the concentration of a microorganism, activity of an enzyme, or other measurable
parameter speciﬁc to the target attribute chosen.

A TTI must show the same time-temperature dependent response as that of the target
attribute when the temperature is the only rate-determining factor. Mathematically, this can

be written as:

F Target = F 77] (2)

87

Assuming an nth order TTI system, the rate equation can be expressed as:
dC/dt = - k C" (3)
where k is the reaction constant and n the reaction order. For a ﬁrst order reaction under
isothermal conditions, the equation can be integrated to give:
In (C, /C) = kt (4)
where C0 is the initial value of the target attribute and C is the value of a target attribute at a
time t, and k is the reaction rate constant.
For a non-ﬁrst order reaction (n"), the integrated expression can be written as:
[Mn-1)] (CM - Co”) = kt (5)
In many cases, the rate constant can be expressed by the Arrhenius equation:

k = k0 exp (~EA /RT) (6)
where k0 is the pre-exponential factor, EA is the activation energy, R is the universal gas
constant and T is the absolute temperature.

In the general case where the reaction order is other than unity, and the reaction rate
constant is independent of the order of reaction, the F value for an isothermal process can be
written as (Hendrickx et al., 1995):

1
T ref

 

t EA 1
(Fire/)1: = E[exp-"IF( - ?) dt (7)

where F m, represents the equivalent heating time at a reference temperature resulting in the
same lethality as the time-varying temperature proﬁle. This equation is valid for the target
attribute as well. Therefore equation (2) will only be satisﬁed when the activation energies of

both the TTI and the target attribute are the same. In other words, for a parameter to be used

88

as a TTI, it must have identical activation energy to that of the attribute of interest.
Mathematically this can be written as:
E... =E... (8)
Since 2 values are equivalent to activation energies, equation (6) can also be

expressed as:

277/ = ZTarget (9)

7.3. MATERIALS AND METHODS
7.3.1. Source and preparation of R-phycoerythrin.

R-phycoerythrin (R-PE) was extracted and puriﬁed as explained in Chapter 2. Pure
R-PE was stored in 60% (w/v) ammonium sulfate and kept at 4°C. Prior to the experiments,
R-PE was dialyzed overnight against distilled deionized water at 4°C, then diluted
immediately before use in 0.012 M borate buffer, pH 9.0 to achieve an absorbance of 0.1 at
565 nm.

7.3.2. Isothermal experiments.

Isothermal inactivation experiments were conducted as described in Chapter 3.
Capillary tubes containing 150 tiL of R-PE in 0.012 M borate buffer, pH 9.0 were placed in a
wire rack and immersed in a Polystat circulator bath (Model 1268-52, Cole-Farmer
Instrument Company, Chicago, IL) connected to a bath programmer (Model 126862,
Cole-Panner). Temperature in the capillary tubes was monitored using a thermocouple
(Resistance Temperature Detector probe, Pt 100, 0.15 cm diameter, (1°C accuracy) inserted
in a control capillary tube containing 150 pL of the same solution. The thermocouple was

connected to a Solomat MPM 200 Modumeter (Solomat Partners LP). Zero time was deﬁned

89

as the time when the R-PE solution reached the target temperature. The heating lag time in
the capillary tubes was 10'l sec.

Preliminary studies were performed to calculate the times and temperature necessary
to achieve at least a one log reduction in ﬂuorescence. Temperatures ranged from 60°C to
70°C. The tubes were removed at predetermined intervals of time and immediately placed in
an ice-water bath, then kept at 4°C until ﬂuorescence was read within 6 h. Preliminary
studies were conducted to show that no changes in ﬂuorescence occurred during the holding
period. Each experiment was performed in triplicate.

7.3.3. Non-isothermal experiments.

Non-isothermal inactivation experiments were performed in a temperature controlled,
programmable thennocycler (GeneAmp PCR System, Model 9600, Perkin Elmer, Norwalk,
CT). R-phycoerythrin solutions (125 |J.L) were placed in 0.2 mL thin-wall microtubes with
attached cap (Cat. No PCR-02A, United Scientiﬁc Products, San Leandro, CA). Heating
protocols representing roast beef and hamburger cooking requirements were performed based
on USDA requirements (USDA-F SIS, 1997) to achieve properly cooked, undercooked and
overcooked samples (Table 7.1). During the experiments, temperature was increased
gradually (A = 11°C/min) to the target temperature, followed by a holding period, then
decreased at the same rate. Samples were allowed to equilibrate at 25°C inside the
thennocycler immediately before and after the heating process. The tubes were cooled in an
ice-water bath immediately after removal from the thennocycler.

Predicted ﬂuorescence values corresponding to the heating protocols were calculated from
the temperature proﬁles according to the General Method (Pﬂug, 1997) using the following

expression:

90

Cp = C, ’0 ""0“” (10)
where Cp is the predicted ﬂuorescence and C0 is the initial ﬂuorescence, t is the lethality (in
min) calculated for each process, and Dref is the D value at the reference temperature (65°C).
Each experiment was performed in triplicate.

The residual ﬂuorescence of the R-PE solutions after heating was analyzed for
statistical differences. One way analysis of variance (AN OVA) for the time—temperature
combinations was conducted using JMP Version 3.2.2. (SAS Institute Inc., 1995). Mean
ﬂuorescence values were compared using Tukey-Kramer HSD test using the mean square
error at the 5% level of probability.

7.3.4. Fluorescence measurement.

R-PE ﬂuorescence at the maximum emission was measured using a CytoFluor II
Microwell Fluorescence reader and the CytoFluorII software, Version 2.00 (Biosearch
Incorporated, Bedford, MA). Heated and unheated R-PE solutions were placed in 96-well
ﬂuorescence reader plates (Cat. No. DGSSIS, Life Science Products Inc., Denver, CO).
Excitation was set at 485/20 nm and emission at 590/35 nm.

7.3.5. Experimental heat inactivation data of S. senftenberg.

Previously published isothermal inactivation data of S. senftenberg (Orta-Ramirez et
al., 1995) was used to test the R-PE model. Brieﬂy, sterile ground beef was inoculated with
S. senftenberg (ATTC 43485) to achieve levels of 107 microorganisms/ g meat. Isothermal

inactivation at 53, 58, 63 and 68°C was performed in the same fashion as above except that

91

Table 7.1. Time-temperature schedules for non-isothermal heating of R-phycoerythrin in

0.012 M borate buffer, pH 9.0.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample Target T Holding Heating/ Cooling Cooking
No. (°C) Time Lag Time Status“

1 60.0 0 3 min 10 sec Undercooked

2 60.0 9 min 3 min 10 sec Undercooked

3 60.0 12 min 3 min 10 sec Properchooked
4 60.0 24 min 3 min 10 sec Overcooked

S 62.7 0 3 min 25 sec Undercooked

6 62.7 1.5 min 3 min 25 sec Undercooked

7 62.7 3 min 3 min 25 sec Properly cooked
8 62.7 6 min 3 min 25 sec Overcooked

9 66.1 0 3 min 40 sec Undercooked
10 66.1 41 sec 3 min 40 sec Properly cooked
11 66.1 82 sec 3 min 40 sec Overcooked

12 69.4 0 4 min Undercooked

13 69.4 10 sec 4 min Properly cooked
14 69.4 20 sec 4 min Overcooked

15 69.4 40 sec 4 min Overcooked

 

“Based on USDA roast beef and hamburger cooking requirements (USDA-FSIS, 1997).

 

one-gram samples were ﬁlled in sterile 10 x 75 thermal death time tubes (14-961-25, Fisher
Scientiﬁc, Pittsburgh, PA). The tubes were heated to internal target temperatures and held for
precalculated intervals of time to achieve at least a one-log reduction in Salmonella. Bacterial
counts were determined by decimal dilution of the ground beef samples in sterile 0.1%
buffered peptone water and plated on Petriﬁlm(Coliform Count Plates (3M, St. Paul, MN).
All samples were incubated at 37°C for 24 h. Plate counts (CFU/mL) were converted to
logarithms. Plate counts less than 10 were entered as zeros. Calculation of D and 2 values

was conducted as explained in Section 7.3.6.

92

7.3.6. Data analysis.

Fluorescence counts were converted to logarithms. D and z values were initially
estimated assuming ﬁrst-order kinetics (Pﬂug, 1997). D values (in minutes) were calculated
by linear regression of ﬂuorescence vs time data, using Microsoft Excel Version 5.0a
(Microsoﬁ Corporation, Redmond, WA). Thermal resistance curves were determined by
plotting log D values vs temperature. The 2 value was determined as the negative reciprocal
of the slope of the thermal resistance curve.

Kinetic parameters (It and k) were calculated using non-linear regression analysis
(Sigma Plot v. 4.01, SPSS Inc., Chicago, IL). First, the D values from the isothermal
experiments were used to estimate the rate constant (k) at each temperature according to the
relationship:

k=2.303/D (11)

The estimated k values for each temperature were used to calculate the reaction order
(11) using the following nth-order model, which results from rearranging Equation (5):

(C/CO) = [(n-1)ket +1] MM” (12)
where C/C0 is the normalized ﬂuorescent intensity (ratio of ﬂuorescence at time t to
ﬂuorescence at time zero), kc is the rate constant estimated from the D value at the same
temperature, and t is time in minutes. The average n obtained from the ﬁve temperatures was,
in turn, used to re-calculate the rate constant (k) using the model above and solving for k at
each temperature. Rate constants were assumed to follow the Arrhenius law:

k = k0 exp[-EA /(RT)] (13)
where k0 is the pre-exponential factor, E A is the activation energy of the process, R is the

universal gas constant (8.314 J mole K“), and T is the temperature.

93

7.4. RESULTS AND DISCUSSION
7.4.1. Isothermal treatment.

The D values of R-PE in 0.012 M borate buffer, pH 9.0 ranged from 512.82 at 60°C
to 10.57 min at 70°C (Table 7.2). A 2 value of 599°C was calculated from the linear
regression of log D vs temperature and fell within the range of 2 values reported for
Salmonella, the target microorganism in thermal processing of beef products (USDA-F SIS,
1997). Published 2 values for this microorganism in ground beef ranged from 56°C to 62°C
(Goodfellow and Brown, 1978; Orta-Ramirez et al., 1995). Variations in 2 values could be
due to different bacterial strains, physiological condition of the cells, fat content and pH of
the meat, and the method employed for enumeration of survivors. Goodfellow and Brown
(1978) used a mixture of Salmonella strains and two enumeration methods: Plate count agar
overlayed with supplemented XL agar base, and the most probable number technique. The
authors did not mention the fat content of the meat employed in their experiments.
Orta-Ramirez et al. (1995) used S. senftenberg enumerated on Petriﬁlm Coliform Count
Plates. The fat content of the meat in the study was 3.3% and the pH was 6.3.

Table 7.2. Summary of D and 2 values and regression analysis for R-phycoerythrin in 0.012
M borate buffer, pH 9.0.

 

 

 

 

 

 

 

T (°C) D value (min) X coefﬁcient Constant R squared
60 513 -0.002 3.6 0.85
62.5 203 -0.005 3.4 0.85
65 65 -0.02 3.3 0.87
67.5 33 -0.03 3.1 0.87
70 11 -0.1 3.3 0.94
z value 5.99 -0.2 12.7 0.99

 

 

 

 

 

 

 

94

Traditionally, the thermal inactivation of microorganisms and proteins has been
assumed to follow ﬁrst-order kinetics (Kormendy and Kormendy, 1997; Peleg and Cole,
1998). In addition, methods for estimating the safety of commercial thermal processes are
also based on this assumption (Peleg and Cole, 1998). Even though at lower temperatures
regression analysis of R-PE seemed to deviate slightly from linearity, the high square
coefﬁcients (R2 > 0.85 for D values at all temperatures and 0.99 for the 2 value) seemed
adequate to assume ﬁrst-order kinetics to initially and simply ﬁt the inactivation of R-PE.
After determining the thermal inactivation parameters, R-PE was veriﬁed using
non-isothermal heating experiments in an attempt to simulate real thermal treatments used in
the processing of beef products.

7.4.2. Non-isothermal treatment.

The non-isothermal experiments were conducted in the 60-69.4°C temperature range
using published requirements (USDA-FSIS, 1997). During the experiments, temperature was
increased gradually (AT = 11°C/min) to the target temperature, followed by a holding period
that had been selected to obtain properly cooked, undercooked and overcooked samples
based on USDA thermal processing regulations in beef products (Table 7.1).

Residual ﬂuorescence of R-PE decreased for all heat treatments when compared to
the unheated protein (p < 0.01) (Figure 7.1). In addition, residual ﬂuorescence decreased as
holding time was increased at each temperature Q) < 0.01). In other words, at a given
temperature, residual ﬂuorescence of R-PE could be used to distinguish between

undercooked, properly cooked and overcooked samples.

95

 

 

 

 

 

 

 
  

 

7000 a
:3 6000 .Undercooked( )
15'. Pro erl cooked *‘
3 5000 b [j P Y ( l
§ 4000. IOveroooked (*)
m 3000. c d g C
2 e
3 20001 HT 11
U- 1000 ’1
0, 111 1111
$6 9 0 -0 9
6° 5‘“ 6““ «.690 459‘ 6‘ o’ Ne
65” @009 '0’ ’1“ .31 0 05° 66‘ o"
0 Q C) 0’0 O ’\ ’\
‘3 6" Q90 6" cake be” be-
Time-temperature combination

 

 

 

Figure 7.1. Residual ﬂuorescence in arbitrary units (a.u.) of R-phycoerythrin in 0.012 M
borate buffer, pH 9.0, after non-isothermal heating

96

Table 7.3. Residual ﬂuorescence of R-phycoerythrin after thermal processing under USDA
cooking regulations.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Temperature Holding time Cooking Status Fluorescence "'
(°C) (a.u.)

0 Undercooked 4349i92 b

60.0 9 min Undercooked 2791i90°
12 min Properly cooked 2523i88 d
24 min Overcooked 2031i223 '3

0 Undercooked 3889i62 f

62.7 1.5 min Undercooked 2910i81g
3 min Properly cooked 2529i30d
6 min Overcooked 20853:]05 °

0 Undercooked 2723i73 c

66.1 41 s Properly cooked 1929i25 ‘
82 s Overcooked 1636i43 h

0 Undercooked 1212i50i

69.4 10 s Properly cooked 11112t26j
20 s Overcooked 1007i47 "

40 s Overcooked 752i108'

 

 

 

 

 

 

lFluorescence values expressed as mean i standard deviation
2'Values with different superscripts are statistically different (p<0.05)

Comparing residual ﬂuorescence at equivalent treatments, the residual ﬂuorescence
decreased as cooking temperatures increased (Table 7.3). This suggests that, for the safe use
of this TTI, the person handling this device would need to know at least the actual
temperature of the thermal process. By establishing a cut-off value of residual ﬂuorescence
for each temperature, one can know if the product has been undercooked or properly thermal
processed. When comparing properly cooked samples, R-PE ﬂuorescence was not different
(p>0.05) between samples cooked to 60 or 62.7°C (p<0.05), but ﬂuorescence was higher than
in samples cooked to 66.1 or 69.4°C. In addition, ﬂuorescence was higher (p<0.05) in
samples properly cooked at 66.1°C than at 69.4°C. Ideally, ﬂuorescence values should have

been the same among equivalent processes (i.e. among either undercooked, properly cooked

97

or overcooked), which, theoretically, result in similar lethalities (Goodfellow and Brown,
1978; USDA-FSIS, 1997). Results suggest that R-PE ﬂuorescence did not respond accurately
to the combination of time and temperature.

Predicted ﬂuorescence values corresponding to the heating protocols were calculated
from the temperature proﬁles using the General Method (Pﬂug, 1997) using 65°C as the
reference temperature. Predicted ﬂuorescence values of R-PE were compared to the observed
ﬂuorescence values (Figure 7.2). Predicted values were much higher than the observed
values, suggesting that, in practice, the heat treatment had a stronger effect on R-PE kinetics
than theoretically expected. When observed and predicted ﬂuorescence values were
compared, there was weak correspondence (R2 < 0.80), indicating that the non-isothermal
inactivation of R-PE did not follow ﬁrst-order kinetics. When residual ﬂuorescence
(expressed as C/CO, where C is ﬂuorescence at time t and CO is initial ﬂuorescence) was
compared to the calculated lethality for each of the processes, an exponential relationship was
found, which a square coefﬁcient of 0.88 (Figure 7.3). This indicated that ﬂuorescence loss
of R-PE upon cooking, could be used as an indicator of thermal process lethality, regardless
of its kinetic order. Several enzymes have thermal inactivation kinetics that do not follow
ﬁrst-order, including: horseradish peroxidase (Weng et al., 1991a and b; Hendrickx et al.,
1992a and b; Saraiva et al., 1996; Garcia et al., 1998); acid phosphatase (Incze et al., 1999);
and Bacillus spp oc-amylase (DeCordt et al., 1992a and b, 1993, 1994a and b; Van Loey et
al., 1995b, 1997a and b), and have been used to monitor process lethality. This non-linear
behavior has been explained by the presence of two or more isoforms of the enzyme showing
different (Chapter 2) have reported the presence of two or more isoforms of R-PE, as

determined on isoelectric focusing, that could account for our ﬁndings.

98

 

 

7000
6000 1 ,
5000 . .

4000 -

3000 ~
2000 < y = 0.2881 x 4' 5056.8

2..
1000 1 R -0.7986

Predicted Fluorescence

 

 

 

0 500 1000 1 500 2000 2500 3000 3500 4000 4500 5000

Observed Fluorescence

 

 

Figure 7.2. Comparison of observed residual ﬂuorescence and predicted ﬂuorescence as
calculated by the General Method of R-phycoerythrin in 0.012 M borate buffer, pH 9.0 after
non-isothermal experiments.

 

 

 

 

 

 

 

3
7 -4
6 .. O
.5 5 y = 16.8348-6'7047)‘
g l . R2 = 0.8794
31‘ 4‘
3 2 .
1 .
0 V V 7 I I T
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.3 0.9 1
FxlFo

 

 

 

Figure 7.3. Comparison of ﬂuorescence of R-phycoerythrin in 0.012 M borate buffer, pH
9.0, after non-isothermal heating, and experimental lethality of the thermal process.

99

In view of these results, experimental data of R-PE obtained with isothermal studies was used
to develop a non-linear model that could explain the thermal inactivation kinetics of R-PE.
7.4.3. The nth-order model.

The D values obtained in the isothermal experiments of R-PE were used to estimate
the reaction constant (kc) for R-PE in borate buffer, pH 9.0, at 60°C, 625°C, 65°C, 675°C
and 70°C. For each temperature, ke was used to calculate the reaction order (n). The average
n obtained from the data sets was, in turn, used to calculate the rate constant (k) at each
temperature. The same procedure was applied to experimental data of S. senftenberg
(Orta-Ramirez et al., 1995) to calculate the rate constant of this microorganism in ground
beef. Using the D values obtained in isothermal experiments, we calculated a reaction order
of n = 0.55 for R-PE (Table 7.4) while S. senftenberg had a reaction order of n = 1.2 (Table
7.5). Peleg and Cole (1998) analyzed previously published survival curves of several
microorganisms, including S. typhimurium, and demonstrated that they did not follow
ﬁrst-order kinetics, and, in fact, the reaction orders for all cases were substantially different
from unity. Van Loey et a1. (1995) determined fractional reaction orders for quality factors in
green peas and white beans. Calculated rate constants were lower for R-PE (Table 7.6) than
for S. senftenberg (Table 7.7) at similar temperatures (62.5-63°C and 67.5-68°C), indicating
that S. senftenberg was more heat sensitive than the protein.

The rate constants (k) for each temperature obtained with the nth-order model were
transformed into D values and compared to those obtained assuming ﬁrst-order kinetics for
both R-PE and S. senftenberg. Although both D-z and k-EA models can be applied with the

same accuracy over a certain temperature range, the D-z model is more generally adopted in

100

Table 7.4. Estimation of reaction order (n) for R-phycoerythrin in 0.012 M borate buffer, pH

9.0, using an nth-order model.

Table 7.5. Estimation of reaction order (n) for Salmonella senftenberg in ground beef using
an nth-order model.

Table 7.6. Estimation of reaction constant (k) for R-phycoerythrin in 0.012 M borate buffer,
pH 9.0, using an nth-order model.

Table 7.7. Estimation of reaction constant (k) for Salmonella senftenberg using an nth-order

model.

 

 

 

 

 

 

 

T (°C) n Std. Error CV (%)
60 0.52 0.14 27.5
62.5 0.58 0.13 22.8
65 0.45 0.13 29.2
67.5 0.56 0.16 21.5
70 0.61 0.09 14.9

 

 

 

 

 

 

 

 

 

 

T (°C) 11 Std. Error CV (%)
53 0.94 0.08 8.75
58 0.93 0.04 4.33
63 1.38 0.09 6.62
68 1.54 0.06 4.05

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T (°C) k Std. Error CV (%)
60 0.006 0.0005 8.5
62.5 0.01 0.001 10.0
65 0.04 0.004 8.7
67.5 0.07 0.007 9.7
70 0.2 0.02 6.6

 

 

 

 

 

 

 

T (°C) k Std. Error CV (%)
53 0.18 0.004 2.4
58 0.30 0.01 3.5
63 0.98 0.07 7.1
68 6.59 0.03 4.5

 

 

 

 

101

 

 

 

 

the area of food science (Hendrickx et al., 1995). The D values for R- PE calculated either

by linear or nth-order kinetics were very similar except at 60°C At this temperature, the D
value calculated assuming first-order kinetics was 512.8 min while when assuming non-linear
kinetics the resulting D value was 388.8 min (Figure 7.4). For S. senftenberg, the D values at
the 63°C and 68°C were very similar when calculated assuming linear or non-linear kinetics.
At 53°C and 58°C the D values calculated assuming ﬁrst-order (53.5 and 15.2 min,
respectively) were noticeably higher than when the D values were calculated assuming non-
linear kinetics (12.8 and 7.7 min, respectively) (Figure 7.5).

Overestimation of D values can result in an over- or underestimation of the thermal
processing value (F value). The P value is the time required to achieve a stated reduction in a
population of microorganisms, and is usually expressed as a multiple of the D value (Singh
and Helman, 1993; Hendrickx et a1, 1995). Let's assume we want to design a thermal process
that results in a 5D reduction in Salmonella at 58°C: if we use the D value (15.2 min)
calculated using linear regression, 5 x 15.2 min = 76 min, while if we use the D value (7.7
min) obtained using non-linear kinetics, 5 x 7.7 min = 38.5 min, which is half of the previous
one. Obviously, this has serious implications for both consumers and meat processors.
Underestimation may pose a hazard to the consumer because of the possible foodborne
contamination. But, also, overestimation and subsequently, overcooking, may result in loss of
quality and nutritional value of the ﬁnal product, while increasing the cost of production due
to unnecessary higher energy consumption.

Isothermal inactivation of S. senftenberg in ground beef (Orta-Ramirez et al., 1995)
was conducted in thermal death time tubes (10 mm diameter) while for R-PE it was

performed in capillary tubes. The advantage of using the latter is that the time required by the

102

 

 

 

 

 

 

 

 

 

 

 

600

500 + + 1st order
3‘ 400 - +non-1st order
5
g 300 .0.
E
a 200 +

100 «—

O I. . 1 4. 1 ‘1
58 6O 62 64 66 68 7O 72

 

Figure 7.4. Thermal resistance curves of R-phycoerythrin in 0.012 M borate buffer, pH 9.0,
calculated using linear (n=1) and non-linear kinetics (n=0.55).

 

 

D value (min)

 

 

60
50 4»- l\

40 1»

30 «- \

20 T \\

 

 

 

—|—— 1st order
-—9—— non-lst order

 

 

 

 

 

 

T("C)

64 66 68 70

 

Figure 7.5. Thermal resistance curves of Salmonella senftenberg in ground beef calculated
using linear (n=1) and non-linear kinetics (n=1.2).

103

 

sample to reach the target temperature (heating lag time) can be neglected. In the case of S.
senftenberg, the use of a meat model system did not allow for the utilization of capillary
tubes. This could have interfered with the precision in calculating D values. At the lower
temperatures, long heating lag times may result in an actual increase of microbial numbers.
At high temperatures, when the microorganism is most sensitive, a considerable destruction
of Salmonella can occur in the surface of the product while the center has not yet reached the
target temperature.

7.4.4. Validation of parameter estimates for the nth-order model to predict thermal
inactivation of S. senftenberg.

In order to compare the inactivation kinetics of R-PE and S. senftenberg, the
activation energies (E A) for both were calculated by a linear regression of 1n k vs l/T. For
R-PE we calculated an E A of 347.41 kJ/mole over the 60-70°C temperature range (Figure 7.6)
and for S. senftenberg an E A of 220.84 kJ/mole over the 53-68°C range (Figure 7.7). In both
cases, the kinetics were very predictable, showing squared coefficients of 0.93 for S.
senftenberg and 0.98 for R-PE (Table 7.8). This suggests that, regardless of the discrepancies

in model parameters (E at E”), a mathematical expression could be derived to predict

Salmonella

Salmonella lethality from R-PE ﬂuorescence loss.

Table 7.8. Kinetic parameters (k0, E A and n) of R-phycoerythrin and Salmonella senftenberg.

 

n Lnko(min") EA(KJ mole") R2

 

R-phycoerythrin 0.55 120 347 0.98
S. senﬂenberg 1.20 80 221 0.93

 

 

 

 

 

 

 

 

104

 

 

 

 

   
 

0 1 1 1 1 1 1 1 1
_1 q 0.00292 0.00294 0.00296 0.00298 0.003
"E -2
4 1 y = 41783): + 120.26
_5 q_ R’ = 0.9834
-6

 

 

1IT (1IK)

 

Figure 7.6. Arrhenius plot of R-phycoerythrin in 0.012 M borate buffer, pH 9.0.

 

 

y - -26561x + 79.4
R“ - 0.9295

 

2.96E-03

 

In It (min 1)
o .5

O
1/

 

 

 

1IT(1/K)

 

Figure 7.7. Arrhenius plot of Salmonella senftenberg in ground beef.

105

 

 

 

 

7.5. CONCLUSIONS

In this study, R-phycoerythrin has been proposed to be used as a single component,
extrinsic, ﬂuorescence—based TTI for the monitoring of thermal processes in beef products.
Assuming ﬁrst-order kinetics, the 2 value calculated for R-PE (599°C) fell within the range
of the 2 value reported for Salmonella in beef products (56°C - 62°C). Under non-isothermal
heating experiments, however, R-PE showed thermal inactivation kinetics that were not
ﬁrst-order. A non-linear model was developed to fit heat inactivation of both, R-PE and S.
senftenberg. In both cases, lethality was very predictable (R2 = 0.98 for R-PE and 0.93 for S.
senftenberg), strongly suggesting that a mathematical relationship can be established between
lethality of S. senftenberg and ﬂuorescence loss of R-PE. By developing such an expression,
R-PE could be used to predict destruction of this microorganism during thermal processing of
beef products.

7.6. NOMENCLATURE AND UNITS

concentration of a target attribute at time t
CO concentration of a target attribute at time zero
predicted concentration of a target attribute

decimal reduction time, time necessary to reduce a quality
D min
attribute by 90% at a constant temperature

m. D value at the reference temperature (65°C in this study) min
E A activation energy of a process k] mole!l
E Marge, activation energy corresponding to a target attribute kJ “1016.1
activation energy corresponding to a time-temperature 1
EATTI k] mole
integrator
F processing value min

106

TargctFT f
re

TTI
F Trc f

ZTargct

ZTTl

processing value as read from the target attribute

processing value as read from the time-temperature

integrator
reaction rate constant

pre-exponential factor

reaction rate constant estimated from D values of isothermal

experiments

reaction order

universal gas constant

time

temperature (TDT-model)
temperature (Arrhenius model)
thermal death time

holding temperature

reference temperature
time-temperature integrator

thermal resistance constant, temperature increase necessary

to reduce the D value by 90%
2 value corresponding to a target attribute

2 value corresponding to a time-temperature integrator

107

min

min

min

min

min

J mole K'l
min

°C

K

°C
°C

°C
°C

CHAPTER 8: Conclusions

Adequate thermal processing of meat products destroys pathogenic microorganisms
that may be present in the raw meat. In this study, the ﬂuorescent protein R-phycoerythrin
(R-PE) was investigated for its application as a time-temperature integrator (TTI) to monitor
adequate thermal processing in beef products. R-PE was selected as an extrinsic, single
component, ﬂuorescent—based TTI based on the calculated z value of 599°C determined
when the protein was dissolved in 0.012 M borate buffer, pH 9.0. This 2 value was found to
fall within the range of the reported 2 value for Salmonella in beef (5.6-6.2°C). This
microorganism has been identiﬁed by the USDA as the target microbe in thermal processing
of beef products.

R-PE was isolated from algal tissues of Porphyra yezoensis with a high purity
(A565/A280>5). Isothermal experiments were conducted to determine the thermal inactivation
parameters (D and 2 values) of R-PE under different conditions of pH and additives. Thermal
inactivation parameters were calculated on the basis of ﬂuorescence loss at the maximum
emission wavelength. The inactivation parameters of this protein were altered by modifying
the R-PE solution conditions. R-PE was most heat resistant between pH 5.0 and 9.0, but
became more heat sensitive at pH 4.0 and 10.0. Addition of sucrose or B-mercaptoethanol
increased the thermostability of R-PE, while sodium-dodecyl sulfate, NaCl and urea had a
destabilizing effect, when compared to solutions of R—PE where no chemicals had been

added. The wide range of 2 values calculated (4.44-37.75°C) suggests that R-PE could also

108

be used as a TTI in other thermal processes, such as sterilization of canned foods or milk
pasteurization.

When R-PE was dissolved in 0.012 M borate buffer, pH 9.0, the resulting 2 value
(599°C) closely matched that of Salmonella in beef (5.6-6.2°C). When R-PE was tested in a
buffer system under non-isothermal conditions, changes in R-PE ﬂuorescence could
distinguish between adequate and inadequate heat processes as required by the USDA beef
processing schedules, but the resulting inactivation kinetics were not ﬁrst-order.

A non-linear mathematical model was developed to ﬁt the heat inactivation kinetics
of R-PE and Salmonella senftenberg in ground beef. The reaction orders (11) calculated for
R-PE and S. senftenberg were 0.55 and 1.2, respectively. The activation energies (E A) were
estimated using the Arrhenius model. The calculated E A for R-PE and S. senftenberg were
347.4 and 220.84 kJ mole“, respectively. Despite the differences in model parameters, the
lethalities determined using the non-linear model were highly predictable (R2=0.98 for for
R-PE and 0.93 for S. senftenberg), suggesting that a mathematical relationship could be
establish between lethality of Salmonella and R-PE ﬂuorescence loss. Thus, it was concluded
that R-PE could be used as the detector component of a TTI to predict destruction of this

microorganism during thermal processing of beef products.

109

CHAPTER 9: Future research

This study is the ﬁrst attempt to develop a ﬂuorescence-based time-temperature
integrator (TTI) to monitor adequate thermal processing of beef products. The following are
recommendations for future research in this area:

Non-isothermal heating experiments were conducted in a buffer model system.
Non-isothermal experiments should also be performed in a meat model system to verify that
heat transfer within the meat does not affect the rate of R-PE inactivation. The effect of fat
content of the meat product on heat transfer should be investigated as well.

It would also be useful to perform non-isothermal experiments of R-PE in
Salmonella-inoculated beef samples to be able to compare actual destruction of the
microorganism and R-PE ﬂuorescence loss during the same heating process.

Experimental inactivation data of S. senftenberg was collected during isothermal
experiments of ground beef in 10 x 75 mm test tubes. The heating lag time (time necessary
for the center of the sample to reach the target temperature) can be considerable at the low
temperatures. On the other hand, at high temperatures, when the microorganism is most
sensitive, a substantial destruction of Salmonella can occur in the outer layers of product
while the center has not yet reached the target temperature. Obviously, this would have a
great signiﬁcance in the proper calculation of D and 2 values and, thus, the actual lethality of

Salmonella by heat. It would be adequate to calculate the kinetics parameters of Salmonella

110

in such a vessel that allows for instantaneous heating of meat samples to avoid the effect of
the heating lag time.

The experiments described here were performed using capillary glass tubes. In a real
situation, this would be unacceptable for the obvious safety reasons. Additional research is
needed to ﬁnd a most suitable material to encapsulate R-PE.

With the introduction of new techniques, such as microwave and ohmic heating,
additional research is needed to investigate possible application of R-PE as a TTI in these
processes.

R-PE belongs to a family of proteins called phycobiliproteins, which also include
phycocyanin and allophycocyanin. It would be of interest to test these and other proteins
which are ﬂuorescent in the visible range, for their applicability as TTIS in monitoring
thermal processes.

Although we studied the use of R-PE in thermal processing of beef products, it would
be beneﬁcial to explore the applicability of R-PE as a TTI in other thermal processes of
foods. The wide range of 2 values obtained by modiﬁcation of solution conditions using pH
and additives, suggest that the z value R-PE can be manipulated to monitor adequate
commercial sterilization of canned foods (2 value for C lostridium botulinum= 10°C) and milk

pasteurization (2 value for C oxiella burnetti= 4°C).

111

BIBLIOGRAPHY

112

BIBLIOGRAPHY

Apt, K.E., Hoffman, NE. and Grossman, AR. 1993. The ’y subunit of R-phycoerythrin and
its possible mode of transport into the plastid of red algae. J. Biol. Chem.
268(22): 16208-1621 5.

Bauer, J. and Hormansdorfer, S. 1996. Salmonellosis in farm animals. Fleischwirtsschaﬂ Intl.
3:40-43.

Bean, NH. and Grifﬁn, RM. 1990. Foodborne disease outbreaks in the United States,
1973-1987. Pathogens, vehicles and trends. J. Food Protect. 53:804-817.

Bems, D.S., Crespi, H.L. and Katz, J .J . 1963. Isolation, amino acid composition and some
physico-chemical properties of the protein deuterio-phycocyanin. J. Am. Chem. Soc. 85:8-14.

Blackburn, C. de B. 1993. Rapid and alternative methods for the detection of salmonellas in
foods. A review. J. Appl. Bact. 75:199-214.

Boye, J .I., Ismail, A.A. and Alli, I. 1996. Effects of physicochemical factors on the secondary
structure of (-1actoglobulin. J. Dairy Res. 63297-109.

Burke, M.J., Pratt, DC. and Moskowitz, A. 1972. Low-temperature absorption and circular
dichroism studies of cytochrome. Biochem. 11:4025-4031.

Chander, R., Rajmane, B.V. and Thomas, P. 1997. Control of Salmonella in poultry by
radiovaccine. J. Food Safety 17:223-228.

CDC. 1998a. "Summary of notiﬁable diseases, United States, 1995". <http://www.cdc.gov/
Epo/mmwr/preview/mmwrhtml/00044418.html> (14 April, 1999)

CDC. 1998b. "Preventing foodborne diseases: Salmonellosis". <http://www.cdc.gov/
ncidod/diseases/foodbome/Salmonella.htm> (4 April, 1999)

Creighton, TE. 1993. Proteins: Structures and Molecular Properties, 2nd ed. W.H. Freeman,
New York, NY.

D'Agnolo, E., Rizzo, R., Paoletti, S. and Murano, E. 1994. R-Phycoerythrin from the red alga
Gracilaria longa. Phytochem. 35:693-696.

113

Davis, 8.]. 1964. Disc electrophoresis. II. Method and application to human serum proteins.
Ann. NY. Acad. Sci. 121 :404-427.

DeCordt, S., Vanhoof, K., Hu, J., Hendrickx, M., Maesmans, G. and Tobback, P. 1992a.
Therrnostability of soluble and immobilized a-amylase from Bacillus licheniformis. Biotech.
Bioeng. 40:396-402.

DeCordt, S., Hendrickx, M., Maesmans, G. and Tobback, P. 1992b. Immobilized oc-amylase
from Bacillus licheniformis: A potential enzymic time-temperature integrator for thermal
processing. Int]. J. Food Sci. Tech. 27:661-673.

DeCordt, S., Saraiva, J., Hendrickx, M., Maesmans, G and Tobback, P. 1993. Changing the
thermostability of Bacillus licheniformis OL-amylase. In Stabilization of Enzymes,
Proceedings of an International Symposium held in Maastricht, The Nederlands. W.J.J. van
den Tweel, A. Harder and RM. Buitelaar (Eds) Elsevier Science Publishers B.V.

DeCordt, S., Avila, I., Hendrickx, M. and Tobback, P. 1994. DSC and protein-based
time-temperature integrators: Case study of (IL-amylase stabilized by polyols and/or sugar.
Biotech. Bioeng. 44:859-865.

Doyle, MP and Cliver, DO. 1990. Salmonella. Ch. 11 in Foodbome Diseases, D.O. Cliver
(Ed.), p. 186-204. Academic Press, Inc., New York.

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

Eley, AR. 1992. Infective bacterial food poisoning. Ch. 2 in Microbial Food Poisoning, A.R.
Eley (Ed.), p. 16-21. Chapman & Hall, London.

FDA. 1993. Recommendations of the US Public Health Service Food and Drug
Administration, Food Code. Section 3-401.11. Food and Drug Administration, Washington,
DC.

Federal Register, 1996. Performance standards for the production of certain meat and poultry
products. Vol. 61 No. 86, p. 19564. Food Safety Inspection Service, US. Dept. of
Agriculture, Washington, DC.

Federal Register, 1997. Irradiation in the production, processing and handling of food. Vol.
62, No. 232, p. 64107—64121.
<http://frwebgate5.access.gpo.gov/cgi-bin/waisgate.cgi?WAISdocID=0759416956+0+0+0&
WAISaction=retrieve> (4 April, 1999)

Federal Register, 1999a. Performance standards for the production of certain meat and
poultry products. Vol. 64, No. 3, p.732-749.
<http://frwebgate5.access.gpo.gov/cgi-bin/waisgate.cgi?WAISdocID=0703428121+2+0+0&
WAISaction=retrieve> (4 April, 1999)

114

Federal Register, 1999b. Irradiation of meat and meat products. Vol. 64, No. 36,
p.9089-9105.
<http://frwebgate5.access.gpo.gov/cgi-bin/waisgate?WAISdocID=0703428121+0+0+0
&WAISaction=retrieve> (4 April, 1999)

F icner, R. and Huber, R. 1993. Reﬁned crystal structure of phycoerythrin from Porphyridium
cruentum at 0.23-nm resolution and localization of the ( subunit. Eur. J. Biochem.
218:103-106.

Folawiyo, Y.L. and Owusu-Apenten, R.K. 1996. Effect of pH and ionic strength on the heat
stability of rapeseed 128 globulin (cruciferin) by the ANS ﬂuorescence method. J. Sci. Food
Agric. 70:241-246.

Fujimori, E. and Pecci, J. 1967. Distinct subunits of phycoerythrin from Porphyridium
cruentum and their spectral characteristics. Arch. Biochem. Biophys. 118:448-455.

Gantt, E. 1969. Properties and ultrastructure of phycoerythrin from Porphyridium cruentum.
Plant Physiol. 44: 1629-1638.

Garcia, D., Ortega, F. and Marty, J .L. 1998. Kinetics of thermal inactivation of horseradish
peroxidase: stabilizing effect of methoxypoly(ethy1ene glycol) Biotechnol. Appl. Biochem.
27:49-54.

Gast, R.K. and Beard, CW. 1993. Research to understand and control Salmonella enteritidis
in chickens and eggs. Poultry Sci. 72:1157-1163.

Glazer, A.N. 1981. Photosynthetic accessory proteins with bilin prosthetic groups. In The
Biochemistry of Plants, Vol. 8. MD. Hatch and N.K. Boardman (Eds), p. 51-96. Academic

Press, Inc., New York.

Glazer, A.N. 1989. Light guides: Directional energy transfer in a photosynthetic antenna. J.
Biol. Chem. 264:1-4.

Glazer, AN. and Hixson, CS. 1977. Subunit structure and chromophore composition of
Rhodophytan phycoerythrins. J. Biol. Chem. 252:32-42.

Glazer, AN. and Wells, S. 1998. Recent advances in ﬂuorescence: Labeling, detection and
visualization. BioRadiations. 98:4-8.

Goff, H.D. 1999. Thermal destruction of microorganisms. <http://www.foodsci.uoguelph.
ca/dairyedu?TDT.html> (19 April, 1999)

Goodfellow, SJ. and W.L. Brown. 1978. Fate of Salmonella inoculated into beef for
cooking. J. Food Protect. 41(8):598-605.

115

Guibault, G.G. 1988. Principles of ﬂuorescence spectroscopy in the assay of food products.
In Fluorescence Analysis in Foods, L. Munck (Ed). p. 33-58. Longman Scientiﬁc and
Technical, New York.

Hattori, A., Crespi, H.L. and Katz, J .J . 1965. Association and dissociation of phycocyanin
and the effects of deuterium substitution on the process. Biochem. 4(7):]225—1238.

Hendrickx. M., Saraiva, J ., Lyssens, J ., Oliveira, J. and Tobback, P. 1992a. The inﬂuence of
water activity on thermal stability of horseradish peroxidase. Intl. J. Food Sci. Tech.
27:33-40.

Hendrickx, M., Weng, Z., Maesmans, G. and Tobback, P. 1992b. Validation of a
time-temperature-integrator for thermal processing of foods under pasteurization conditions.
Intl. J. Food Sci. Tech. 27:21-31.

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

Holdsworth, SD. 1997. Thermal Processing of Packaged Foods, 1st ed. Blackie Academic &
Professional, New York.

Incze, K., Konnendy, L., Konnendy, I. and Zsamoczay, G. 1999. Considerations of critical
microorganisms and indicator enzymes in connection with the pasteurization of meat
products. Meat Sci. 51:115-121.

Jones, RF. and Fujimori, E. 1961. Interactions between chromophore and protein in
phycoerythrin from the red alga Ceramium rubrum. Physiol. Plant. 14:253-259.

Kay, H.D. and Graham, W.R. 1933. Phosphorus compounds of milk: the effect of heat on
milk phosphatase. J. Dairy Res. 5:65-74.

Kim, H.J. and Taub, LA. 1993. Intrinsic chemical markers for aseptic processing of
particulate foods. Food Tech. 47(1):91-99.

Klivanov, AM. 1983. Stabilization of enzymes against thermal inactivation. Adv. Appl.
Microb. 29: 1-28.

Kormendy, I. and Konnendy, L. 1997. Considerations for calculating heat inactivation
processes when semilogarithmic thermal inactivation models are non-linear. J. Food Eng.

34:33-40.

Laemli, UK. 1970. Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature 227:680-685.

Lakowicz, J .R. 1983. Principles of Fluorescence Spectroscopy. Plenum Press, New York.

116

Line, J .E., Fain, A.R. JR, 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.

MacColl, R. and Guard-Friar, D. 1987. Phycobiliproteins. CRC Press, Boca Raton, FL.

Maesmans, G., Hendrickx, M., DeCordt, S. and Tobback, P. 1994a. Feasibility of the use of a
time-temperature integrator and a mathematical model to determine ﬂuid-to-particle heat
transfer coefﬁcients. Food Res. Int. 27:39-41.

Maesmans, G., Hendrickx, M., DeCordt, S., Van Loey, A., Noronha, J. and Tobback, P.
1994b. Evaluation of process value distribution with time-temperature integrators. Food Res.
Int. 27:413-423.

Merrill, J .E. 1985. Photosynthetic Pigment Dynamics in Red Algae. Ph.D. dissertation, Univ.
of Washington, Seattle, WA.

Merrill, J .E. 1993. Development of nori markets in the western world. J. Appl. Phycol.
5:149-154.

Mulley, E.A., Stumbo, CR. and Hunting, W.M. 1975. Thiamine: A chemical index of the
sterilization efﬁcacy of thermal processing. J. Food Sci. 40:993-996.

Nolan, D.N. and O'hEocha, C. 1967. Determination of molecular weights of algal biliproteins
by gel ﬁltration. Biochem. J. 103:39P

Ogawa, H., Mizuno, H., Saito, T., Yamada, Y., Oohusa, T. and Iso, N. 1991. Effects of pH on
the conformation of phycoerythrin from Nori Porphyra sp. Nippon Suisan Gakkaishi.
57(5):899-903.

O'hEocha, C. 1965. Biliproteins of algae. Ann. Rev. Pl. Physiol. 16:415-434.

O'hEocha, C. and O'Carra, P. 1961. Spectral studies of denatured phycoerythrins. J .Am.
Chem. Soc. 83:1091-1093.

Orta-Ramirez, A. 1994. Lactate Dehydrogenase as Indicator of Proper Heat Processing and
Death of Escherichia coli 0157:H7 and Salmonella in Ground Beef Patties. MS. Thesis.
Michigan State University, East Lansing, MI.

Orta-Ramirez, A., Price, J .F ., Hsu, Y-.C., Veeramuthu, G.J, Cherry-Merritt, J .S. and Smith.
D.M. 1995. Thermal inactivation of Escherichia coli 0157:H7, Salmonella and enzymes with
potential as time-temperature indicators in ground beef. J. Food Protect. 60:471-475.

Orta-Ramirez, A., Smith, D.M., Wang, C.H., Veeramuthu, G.J., Abouzied, M.M., Price, J .F.

and Pestka, J .J . 1996. Lactate dehydrogenase monoclonal antibody sandwich ELISA to
determine cooking temperature of ground beef. J. Agric. Food. Chem. 44:4048-4051.

117

Peleg, M. and Cole, MB. 1998. Reinterpretation of microbial survival curves. Crit. Rev.
Food Sci. 38:353-380.

Pﬂug, U. 1997. Evaluating a ground-beef patty cooking process using the general method of
process calculation. J. Food Prot. 60:1215-1223.

Pﬂug, M., Jones, AT. and Blanchett, R. 1980. Performance of bacterial spores in a carrier
system in measuring the F0-value delivered to cans of food heated in a steritort. J. Food Sci.
45:940-945.

Pﬂug, U, and Odlaugh, TE. 1978. A review of z and F values used to ensure the safety of
low-acid canned food. Food Tech. 32(6):63-70.

Rajeshwara, AN. and Prakash, V. 1996. Effect of denaturant and cosolvents on the stability
of wheat germ lipase. J. Agric. Food Chem. 44:736-740.

Robertson, E.F., Dannelly, H.K., Malloy, RI. and Reeves, HQ 1987. Rapid isoelectric
focusing in a vertical polyacrylamide minigel system. Anal. Biochem. 167:290-294.

Rudiger, W. 1992. Events in the phytochrome molecule aﬂer irradiation. Photochem.
Photobiol. 56:803-809.

Sadeghi, F. and Swartzel. KR. 1990. Time-temperature equivalence of discrete particles
during thermal processing. J. Food Sci. 55:1696—1698, 1739.

Saraiva, J ., Oliveira, J .C., Oliveira, S. and Hendrickx, M. 1996. Inactivation kinetics of
horseradish peroxidase in organic solvents of different hydrophobicity at different water
contents. Intl. J. Food Sci. Tech. 31:233-240.

SAS Institute Inc. 1995. JMP( Introductory Guide, Version 3. SAS Institute Inc., Cary, NC.
Sastry, SK. 1986. Mathematical evaluation of process schedules for asseptic processing of
low-acid foods containing discrete particulates. J. Food Sci. 51:1323.

Sastry, S.K., Li, S.F., Patel, P., Konanayakarn, M., Bafna, P., Doores, S. and Beelman, RB.
1988. A bioindicator for veriﬁcation of thermal processes for particulate foods. J. Food Sci.
53:1528-1531, 1536.

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

Sidler, W., Kumpf, B., Suter, F., Klotz, A.V., Glazer, AN. and Zuber, H. 1989. The complete
amino acid sequence of the or and [3 subunits of B-Phycoerythrin from the Rhodophytan alga
Porphyridium cruentum. Biol. Chem. Hoppe-Seyler. 370:115-124.

Singh, RP. and Wells, J .H. 1985. Use of time-temperature indicators to monitor quality of
frozen hamburger. Food Technol. 39(12):42-50.

118

Singh, RP. and Heldman, DR. 1993. Introduction to Food Engineering, 2nd ed. Academic
Press Inc. New York, NY.

Smith. D.M. and Orta-Ramirez, A. 1995. Enzyme-linked immunosorbent assay technology to
verify endpoint cooking temperatures of meat products. J. Clin. Ligand Assay. 18:161-165.

Taoukis, P.S., F u, B. and Labuza, TR 1991. Time-temperature indicators. Food Technol.
45(10):70-82.

Tauxe, R.V. 1991. Salmonella: A post-modem pathogen. J. Food Prot. 54:563-568.

Tcheruov, A.A., Minkova, K.M., Georgiev, DI. and Houbavenska, NE. 1993. Method for
B-phycoerythrin puriﬁcation from Phorphyridium cruentum. Biotech. Tech. 72853-858.

Teale, F.W. and Dale, RE. 1970. Isolation and spectral characterization of phycobiliproteins.
Biochem. J. 116:161-169.

Timasheff, S. N. and Arakawa, T. 1997. Stabilization of protein structure by solvents. In
Protein Structure: A Practical Approach, T.E. Creighton (Ed.), p.349-364. Oxford University
Press, New York.

Todd, E.C.D. 1989. Preliminary estimates of costs of foodborne disease in the United States.
J Food Prot. 52:595-601.

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

Todd, E.C.D. 1996. Worldwide surveillance of foodborne disease: the need to improve. J.
Food Protect. 59:82-92.

Townsend, W.JE. 1989. Stability of residual acid phosphatase activity in cured/caned picnic
samples stored at -34°C for 15 and 36 months. J. Food Sci. 54:752-753.

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

USDA-F SIS. 1986a. Determination of internal cooking temperature (Acid

phosphatase activity). In Revised Basic Chemistry Laboratory Guidebook. No. 3.018z3-49.
Science Chemistry Div., Food Safety and Inspection Service, US. Dept. of Agriculture,
Washington, DC.

USDA-FSIS. 1986b. Determination of cooked temperature (Coagulation). In Revised Basic

Chemistry Laboratory Guidebook. No. 3.019, p. 3-55. Science Chemistry Div., Food Safety
and Inspection Service, US. Dept. of Agriculture, Washington, DC.

119

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

USDA-F SIS. 1990. Code of Federal Regulations, Title 9. Ofﬁce of the Federal Register,
National Archives and Records, GSA: Washington, DC. Chapter 3, 491-492.

USDA-F SIS. 1992. Poultry irradiation and preventing foodborne illness. FSIS Backgrounder,
pp. 1-6. Food Safety and Inspection Service, US. Dept. of Agriculture, Washington, DC.

USDA-FSIS. 1993. Instructions for verifying internal temperature and holding time of meat
patties. F SIS Directive 7370.1, Food Safety Inspection Service, US. Dept. of Agriculture,
Washington, DC.

USDA-FSIS. 1997. Heat-processing procedures, cooking instructions, and cooling, handling
and storage requirements for uncured meat patties. In Code of Federal Regulations, Food
Safety Inspection Service (Meat and Poultry), Ch. III, Part 318.23. Ofﬁce of the Federal
Register, National Archives and Records, Washington, DC.

Van Loey, A. F ransis, A., Hendrickx, M., Maesmans, G. and Tobback, P. 1995a. Kinetics of
quality changes in green peas and white beans duing thermal processing. J. Food Eng.
24:361-377.

Van Loey, A., Ludikhuyze, L., Hendrickx, M., DeCordt, S. and Tobback, P. 1995b.
Theoretical consideration on the inﬂuence of the z-value of a single component
time-temperature integrator on the thermal process impact evaluation. J. Food Prot. 58:39-48.

Van Loey, A., Hendrickx, M., De Cordt, S., Haentjens, T. and Tobback, P. 1996.
Quantitative evaluation of thermal processes using time-temperature integrators. Trends
Food. Sci. Tech. 7(1):1-33.

Van Loey, A., Arthawan, A., Hendrickx, M., Haentjens, T. and Tobback, P. 1997a. The
development and use of an a-amylase-based time-temperature integrator to evaluate in-pack
pasteurization processes. Lebensm.-Wis. u.-Technol. 30:94-100.

Van Loey, A.M., Haentjens, T.H., Hendrickx, ME. and Tobback, P.P.1997b. The
development of an enzymic time-temperature integrator to assess thermal efﬁcacy of

sterilization of low-acid canned foods. Food Biotechnol. 11:147-168.

Voet, D. and Voet, J .G. 1990. Biochemistry, lst ed. John Wiley & Sons, New York.

Weber, K. and Osborn, M. 1969.The reliability of molecular weight determination by
dodecyl sulfate-polyacrylamide gel electrophoresis. J. Biol. Chem. 244:4406-4412.

Weng, Z., Hendrickx, M. and Tobback, P. 1991a. Thermostability of soluble and
immobilized horseradish peroxidase. J. Food Sci. 56:574-578.

120

Weng, Z., Hendrickx, M., Maesmans, G. and Tobback, P. 1991b. Immobilized peroxidase: A
potential bioindicator for evaluation of thermal processes. J. Food Sci. 56:567-570.

Whiting, RC. 1993. Modeling bacterial survival in unfavorable environments. J. Ind.
Microb. 12:240-246.

Yawger, ES. 1978. Bacteriological evaluation for thermal process design. Food Tech.
32(6):59-62.

121

nrcurcnn STATE UNIV. LIBRnRIEs
11111111111111 1111111”11111111111111” 11111 1111111 111111111
31293020589978