LIBRARY

Mlchlgan State

Unlverslty

 

 

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

 

MTE DUE

MTE DUE

DATE DUE

 

"13 35 71.2
JAN 0 6 2003

 

JARW‘WM

 

 

9'“ h "b 7:9”!

'1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 chlWDupfiS—p.“

EFFECTS OF LACTIC ACID BACTERIA
ON THE IMMUNE SYSTEM

By

Maria Victoria Tejada Simon

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

1998

ABSTRACT

EFFECTS OF LACTIC ACID BACTERIA ON THE IMMUNE SYSTEM

By

Maria Victoria Tejada Simon

Lactic acid bacteria are essential for the fermentation of products such as cheese,
buttermilk and yogurt. An increasing number of functional foods and pharmaceutical
preparations are being promoted with health claims based on the potential probiotic
characteristics of some of these bacteria and on their capacity for stimulating the host
immune system. A possible mechanism for these effects is direct stimulation of the
gastrointestinal immune response. The specific objectives of these studies were to
evaluate the effects of in vivo, ex vivo, and in vitro exposure to viable, non-viable strains,
and cell extracts of lactic acid bacteria (Lactobacillus acidophilus, L. bulgaricus, L. casei,
L. gasseri, L. helveticus, L. reuteri, Streptococcus thermophilus and Bifidobacterium) on
leukocyte function. In vivo studies showed that growth rate of mice as well as
immunoglobulin levels were not affected by direct oral administration of lactic acid
bacteria. Although basal cytokine mRNA expression in spleen and Peyer’s patches was
not affected by repeated oral lactic acid bacteria administration (in vivo), single exposures

to certain bacteria altered subsequent mitogen induced cytokine and nitric oxide

production by peritoneal cells (ex vivo). When mice were fed a fermented milk
manufactured with starter cultures containing different species/strains of lactic acid
bacteria for three weeks and after immunizing twice with 10 pg cholera toxin, those mice
responded by producing specific intestinal and serum IgA-anti cholera toxin, isotype that
was significantly increased in mice fed yogurts made with starters containing
Lactobacillus bulgaricus and Streptococcus thermophilus (yogurt bacteria) supplemented
with L. acidophilus and Bifidobacterium spp. These results suggested that lactic acid
bacteria may alter immune function in a strain dependent manner. The effects of in vitro
exposure to heat-killed cells, their cell walls, and their cytoplasmic extracts on
proliferation, cytokine and intermediate production were examined in the RAW 264.7
macrophage cell line, spleen and Peyer’s patch cells as well as in peritoneal .cells from
mice. Lactic acid bacteria as well as their cytoplasmic and cell wall fractions were able to
stimulate cloned macrophages to produce very significant amounts of TNF-oc, IL-6 and
nitric oxide. While similar effects were not noted in spleen and Peyer’s patch cell cultures
from mice, a pronounced enhancement in IL-6 production by peritoneal cells was
observed when cultured with those extracts. The results suggested that as a group, the
lactic acid bacteria are capable of stimulating macrophages and/or other immune cells to

produce cytokines and nitric oxide.

To Klaus, your love and support kept me motivated

iv

ACKNOWLEDGMENTS

I wish to express my gratitude to Dr. James J. Pestka, my major professor, for his
estimable guidance and encouragement throughout this study and in preparation of this
manuscript. Thanks are also given to Dr. John E. Linz for his accessibility and valuable
advice at all times, Dr. Zeynep Ustunol and Dr. Melvin Yokoyama for their assistance
during this research as members of my committee. I also appreciate the help of all my

family and friends who supported me while accomplishing this work.

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES
ABBREVIATIONS

INTRODUCTION

Chapter

1 LITERATURE REVIEW
1.1 Lactic acid bacteria

Historical background

Description and classification

Metabolism

Structure of bacteria cell wall

Industrial applications

Lactic acid bacteria as probiotics and animal supplements
Immunomodulation

Side effects and toxicity of lactic acid bacteria

h—ll—lI—il—It-dl—lh-‘r-d
HI—awp—Lal—si—l—r
OO\IO\UI-§UJNr-‘

1.2 Gastrointestinal Immune System
1.3 Rationale for this research

2 EFFECTS OF LACTOBACILLI, STREPTOCOCCI AND
BIFIDOBACTERIA INGESTION ON MUCOSAL AND
SYSTEMIC CYTOKINE GENE EXPRESSION AND
SECRETION IN A MURINE MODEL

2. 1 Abstract
2.2 Introduction
2.3 Material and Methods

2.3.1 Microorganisms
2.3.2 Experimental design

2.3.2.1 In vivo studies

vi

ix

xi

xiii

10
18
24
29
31
33

43
46

49

50
51
54

54
56

56

2.4

2.5

2.3.2.2 Ex vivo studies
2.3.3. Statistical methods
Results

2.4.1 Viability of lactic acid bacteria

2.4.2 Study I. In vivo effects of lactic acid bacteria on cytokine
mRNA expression

2.4.3 Study 11. Ex vivo effects of lactic acid bacteria on immuno-
globulin and cytokine levels

Discussion

EFFECTS OF LACTIC ACID BACTERIA AND THEIR EXTRACTS
ON CYTOKINE PRODUCTION IN WT R0

3.1
3.2
3.3

3.4

3.5
3.6

Abstract
Introduction
Material and Methods

3.3.1 Lactic acid bacteria fractionation
3.3.2 Sugar determination

3.3.3. Protein determination

3.3.4. DNA determination

3.3.5. RAW 267.4 macrophage culture
3.3.6. Leukocyte preparation

3.3.7. MTT assay

3.3.8. Cytokine quantification

3.3.9. Nitric oxide determination
3.3.10. Statistical methods

Results

3.4.1 Bacteria fractionation process
3.4.2 Cell fraction constituents
3.4.3 Effects of lactic acid bacteria fractions on RAW 264.7 cells

3.4.4 Effects of lactic acid bacteria fractions on peritoneal cells

Discussion
Future studies

ENHANCED MUCOSAL AND SYSTEMIC IgA RESPONSES TO
CHOLERA TOXIN IN MICE FED YOGURT CONTAINING
LAC T OBA CILL US ACIDOPHIL US AND BIFIDOBA C TERI UM SPP.

vii

60
65
66
66

66

72
91

102

103
104
107

107
111
112
112
114
115
117
117
119
119

120

120
120
128
146

160
172

173

4.1 Abstract
4.2 Introduction
4.3 Material and Methods

4.3.1
4.3.2
4.3.3
4.3.4
4.3.5
4.3.6
4.3.7
4.3.8

4.4 Results

4.4.1
4.4.2
4.4.3
4.4.4

Yogurt preparation

Enumeration of yogurt bacteria

Animal model and diet

CT immunization

Fecal pellet and serum preparation
Lymphocyte culture

Enzyme-linked immunosorbent assay (ELISA)
Statistical analysis

Viability of the refrigerated yogurt during storage
Body weight and feed intake

Specific immunoglobulin production

Specific immunoglobulin production in cell cultures

4.5 Discussion
4.6 Future studies

5 SUMMARY

LIST OF REFERENCES

viii

174
176
179

179
179
181
184
184
185
186
188

189

189
189
189
197

204
216

217

220

LIST OF TABLES

TABLE 1.1 Lactic acid bacteria on fermentation of foods

TABLE 1.2 Types of fermented milks

TABLE 1.3 Types of alcoholic fermented milks

TABLE 1.4 Recent human studies on immunostimulation by lactic acid bacteria

TABLE 1.5 Recent animal studies on immunostimulation by lactic acid bacteria

TABLE 1.6 Recent in vitro studies on immunostimulation by lactic acid bacteria

TABLE 2.1 List of cultures

TABLE 2.2 Primer sequences for cytokine cDNA amplification and probe
sequences for slot-blot hybridization to detect amplified cDNA
product

TABLE 2.3 Effects of freezing (-8OOC) on survival of lactic acid bacteria

TABLE 3.1 Cultures used in study

TABLE 3.2 Relationship between weight and number of bacterial cells

TABLE 3.3 Chemical composition of bacterial fractions

TABLE 3.4 Protein content of the heat-killed fraction of lactic acid bacteria
digested or not with protease, RN Ase, and DNAse

TABLE 3.5 Protein content of the cytoplasmic fraction of lactic acid bacteria
digested or not with protease, RNAse and DNAse

TABLE 3.6 Carbohydrate content of the heat-killed fraction of lactic acid
bacteria digested or not with protease, RNAse and DNAse

TABLE 3.7 Carbohydrate content of the cytoplasmic fraction of lactic acid
bacteria digested or not with protease, RN Ase and DNAse

TABLE 3.8 Effect of heat-killed lactic acid bacteria on proliferation of
RAW 264.7 cell line measured by MTT assay

ix

25

27

28

35

37

41

55

59

67

108

121

122

124

125

126

127

131

TABLE 3.9 Effect of the cytoplasmic fraction of lactic acid bacteria on
proliferation of RAW 264.7 cell line measured by MTT assay

TABLE 3.10 Effect of the cell wall fraction of lactic acid bacteria on proliferation
of RAW 264.7 cell line measured by MTT assay

TABLE 3.11 Effect of heat-killed lactic acid bacteria on TNF-or production by
RAW 264.7 cells

TABLE 3.12 Effect of heat-killed lactic acid bacteria on IL-6 production by
RAW 264.7 cells

TABLE 3.13 Effect of the cytoplasmic fraction of lactic acid bacteria on TNF-a
production by RAW 264.7 cells

TABLE 3.14 Effect of the cytoplasmic fraction of lactic acid bacteria on IL-6
production by RAW 264.7 cells

TABLE 3.15 Effect of the cell wall fraction of lactic acid bacteria on TNF-or
production by RAW 264.7 cells

TABLE 3.16 Effect of the cell wall fraction of lactic acid bacteria on IL-6
production by RAW 264.7 cells

TABLE 3.17 Effect of cytoplasmic fraction of lactic acid bacteria treated with
protease, RNAse or DNAse on IL-6 production by RAW 264.7 cells

TABLE 3.18 Effect of heat-killed lactic acid bacteria on N03' production by
RAW 264.7 cells

TABLE 3.19 Effect of the cytoplasmic fraction of lactic acid bacteria on N03"
production by RAW 264.7 cells

TABLE 3.20 Effect of the cell wall fraction of lactic acid bacteria on N03'
production by RAW 264.7 cells

TABLE 3.21 Effect of heat-killed, cytoplasmic and cell wall fraction of lactic acid
bacteria on IL-6 production by peritoneal macrophages

TABLE 4.1 Composition of lactic acid bacteria in commercial yogurt starter
cultures

TABLE 4.2 Gravimetric estimates to cholera toxin specific immunoglobulins
after three weeks of oral yogurt administration

132

133

135

136

137

138

139

141

142

143

144

145

157

180

203

LIST OF FIGURES

Figure 1.1 Carbohydrate fermentation pathways in lactic acid bacteria
Figure 1.2 Citrate and malate fermentation pathway in lactic acid bacteria
Figure 1.3 Proteolysis in lactic acid bacteria

Figure 1.4 Gram negative bacterial cell envelope

Figure 1.5 Gram positive bacterial cell envelope

Figure 2.1 Weight gain on mice fed lactic acid bacteria for 7 and 14 days

Figure 2.2 Cytokine mRNA levels in mice after oral exposure to live and
dead Lactobacillus bulgaricus 1489 (NCK 231) for l and 7 days

Figure 2.3 Cytokine mRNA levels in mice after oral exposure to Lbulgaricus
1489 (NCK 231) and S. thermphilus St-133 for 7 and 14 days

Figure 2.4 IL-6 levels in peritoneal macrophage cultures from mice fed a single
dose of lactic acid bacteria

Figure 2.5 IFN-y levels in peritoneal macrophage cultures from mice fed a single
dose of lactic acid bacteria

Figure 2.6 IL-12p40 levels in peritoneal macrophage cultures from mice fed a
single dose of lactic acid bacteria

Figure 2.7 Nitric oxide levels in peritoneal macrophage cultures from mice fed a
single dose of lactic acid bacteria

Figure 2.8 IgA levels in sera collected from mice after feeding
A) 1 and 7 doses, B) 1 dose of several viable lactic acid bacteria

Figure 2.9 IgG levels in sera collected from mice after feeding A) 7 and 14
doses, B) 1 dose of several viable lactic acid bacteria

Figure 2.10 IgA levels in feces collected from mice after feeding 7 and 14 doses

Figure 2.11 Possible effects of mitogens on peritoneal cell cultures extracted from
mice fed one dose of lactic acid bacteria

xi

12

15

17

20

23

69

71

74

77

79

82

84

86

88

90

99

Figure 3.1 Schematic representation of lactic acid bacteria fractionation

Figure 3.2 Effect of incubation with different fractions of L. casei ATCC 39539

on RAW 264.7 macrophage cells

Figure 3.3 Effect of polymyxin B on L. casei ATCC 39539 fractions and LPS
stimulation properties

Figure 3.4 Kinetics of IL-6, TNF-oc and No?” after stimulation of RAW 264.7

cells with L. casei ATCC 39539 fractions for l, 6, 12, 24 and 48 hours

Figure 3.5 Effect of incubation of peritoneal macrophages with different fractions

of L. casei ATCC 39539

Figure 3.6 Effect of incubation of peritoneal macrophages with different fractions

of lactic acid bacteria

Figure 3.7 Kinetics of IL-6 after stimulation of peritoneal macrophages with
L. casei ATCC 39539 fractions for 1, 6, 12, 24 and 48 hours

Figure 3.8 LPS signaling pathways in macrophages

Figure 4.1 Experimental design for assessing effects of yogurt ingestion on
immunoglobulin response to cholera toxin

Figure 4.2 Bacterial counts. Total aerobic, streptococci and bifidobacteria
counts in yogurts manufactured for feeding trial

Figure 4.3 Weight gain for mice fed skim milk or yogurt made with lactic acid
bacteria and bifidobacteria for 3 weeks

Figure 4.4 Specific IgA-anti—cholera toxin in fecal samples (coproantibodies)
Figure 4.5 Specific IgA-anti-cholera toxin in serum samples

Figure 4.6 Specific IgG-anti-cholera toxin in serum samples

Figure 4.7 Gastrointestinal immune response following oral immunization

xii

110

130

148

150

152

154

159

166

183

191

193

196

200

202
209

ABTS
AC
AIN

AN OVA

ATCC
ATP
BMC
BZ-MG
BSA
[cm
CaM
cAMP
CF
CFU
CPM
CT
cw
DAG
DMEM

DMSO

ABBREVIATIONS
2, 2’-azino-bis (3-ethylbenz-thiazoline-6 sulfonic acid
Adenylate cyclase
American Institute of Nutrition
Analysis of variance
Antigen presenting cells
American Type Culture Collection
Adenosine tri-phosphate
Blood mononuclear cells
BZ-microglobulin
Bovine serum albumin
Intracellular calcium concentration
Calmodulin
Cyclic adenosine mono-phosphate
Cytoplasmic fraction
Colony forming unit
Counts per minute
Cholera toxin
Cell wall
Diacylglycerol
Dubelcco’s modified Eagle’s medium

Dimethylsulfoxide

xiii

DPA
ELISA
EMP
FBS
GALT
GI

HBSS

IEL

IFN

IL

ip
1P3
iv
LAB
LP
LPS 1

LTA

MRSL

Diphenilamine

Enzyme-linked immunosorbent assay
Embden—Meyerhoff-Parnas pathway
Fetal bovine serum

gut-associated lymphoid tssue
Gastrointestine

Hank’s buffer saline solution
Ionomycin

Intraepithelial lymphocytes
Interferon

Irnmunoglobulin

Interleukin

Intraperitoneal

inositol tri-phosphate

Intravenous

Lactic Acid Bacteria

Lamina propria

Lipopolysaccharide

Lipoteichoic acids

De Man-Rogosa—Sharpe medium for lactobacilli

MRS + 5% lactose (w/w)

(3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyltetrazolium bromide)

xiv

NED N-( 1 -naphthyl) ethylenediamine dihydrochloride

NFDM non fat dry milk

NK Natural killer cells

NO Nitric oxide

NOS Nitric oxide synthase

NPNL Neomycin sulfate, paromomycin sulfate, nalidixic acid, lithium chloride

medium for bifidobacteria

PBS Phosphate buffered saline
PBS-T PBS + 0.2% Tween 20

PCR Polymerase chain reaction
PIP; Phosphoinositol bi-phosphate
PKA Protein kinase A

PKC Protein kinase C

PLC Phospholipase C

PMA Phorbol l2-myristate-13 acetate
PP Peyer’s patches

RT Reverse transcriptase

SC Secretory component

S.E.M Standard error of the mean
spp. Species

ssp. Subspecies

TA Teichoic acids

XV

TCA

TdT

Th

WC

Trichloroacetic acid

Terminal deoxyribonucleotide transferase
T helper cells

Tumor necrosis factor

0.01M Tris-Cl with lmM EDTA

Whole cell

xvi

INTRODUCTION

For thousands of years lactic acid bacteria have been used to produce fermentation
of a variety of foods and improve the shelf-life, flavor and texture of these products. The
properties attributed to lactic acid bacteria such as anticholesterol activity, improvements in
intestinal motility, metabolism of drugs, vaginitis, immunological status, and the decreased
gastrointestinal disorders and tumors seem to be due to the metabolic activity of these
bacteria and their interaction with other microflora and/or different cells present in the gut
(Fuller, 1991). There is an extensive body of literature addressing the possible health benefits
associated with the consumption of lactic acid bacteria. The putative mechanisms involved
in the production of these favorable effects include changes in viable counts of
microorganisms in the intestinal flora after ingestion, competition for adhesion sites and
nutrients between the ingested bacteria and potential pathogens, production of antibacterial
substances and the action of these bacteria through stimulation of the immune system.
However, convincing scientific experiments which validate these mechanisms are limited
(Gasson, 1993).

The effect of oral administration of lactic acid bacteria on immunity has not been well
defined yet but it is believed that this group of bacteria target the intestine, resist the
digestion process and establish themselves in the gut. Several cell types present in the
intestinal wall such as macrophages, could be stimulated (Perdigon et al., 1986a, 1987, 1988,
1990, 1992; Saito et al., 1987). Cytokines might be produced as a response (Halpem et al.,
1991; Kishi et al., 1996; Kitazawa et al., 1992, 1994; De Simone etal., 1986; Miettinen et
al., 1996) thereby driving a cascade of events which could eventually lead to activation of

other immune cells, further production of cytokines and mediators as well as

immunoglobulin secretion.

Cells of a large number of microbial species as well as bacterial products have been
shown to possess mitogenic and polyclonal activation properties, and are fully capable of
inducing DNA synthesis, blast formation and ultimately division of lymphocytes. One of the
most widely studied bacterial products with this mitogenic activity is lipopolysaccharide
(LPS), which is present in the cell wall of Gram negative bacteria (Gammage et al., 1996).
Some bacterial species that represent major human pathogens are Gram positive and possess
a basic Gram positive cell wall structure. Gram positive purified cell walls containing
teichoic acid and peptidoglycan also seem to be mitogenic and to induce production of
certain cytokines by human monocytes (Heumann et al., 1994).

Several fimdamental questions need to be answered regarding the use and activity of
lactic acid bacteria to better understand their putative probiotic effects in order to ensure the
rational development of new healthy foods. These include precise knowledge of the specific
strains producing beneficial effects, the active constituent of lactic acid bacteria responsible
for each potential effect, the target site in the body, the ecological conditions necessary for
the activity of the active constituent, the pharmacokinetics of ingested lactic acid bacteria,
the percentage of survival at the target site, the concentrations of bacteria, the duration of
passage and its effects, and the type of cells being altered by these bacteria (Fuller, 1991).
All these studies are relevant to the food industry and could be translated into novel
commercially viable foods.

The underlying hypothesis in this dissertation is that lactic acid bacteria have
immunomodulatory effects in cells of the mucosal and systemic compartment and that the

immunopotentiating activity resides in the cell walls of lactobacilli, streptococci, and

bifidobacteria.
The objectives of this research were as follows:

1. To assess the effects of in vivo oral exposure to viable and non-viable strains
of lactic acid bacteria using a murine model.

2. To assess the effects of ex vivo exposure of primary leukocytes to lactic acid
bacteria.

3. To determine the in vitro effects of lactic acid bacteria and their extracts
obtained by fractionation on proliferation, cytokine and intermediate
production by exposing a macrophage cell line.

4. To determine the in vitro effects of lactic acid bacteria and their extracts on
cytokine and intermediate production by exposing primary leukocytes from
mice.

5. To compare the effects of ingestion of yogurts manufactured with different
starter cultures on the gastrointestinal immune system of the mouse.

The aforementioned issues were addressed in this research. The dissertation presented
here is composed of five chapters. Chapter 1 reviews recent and classical literature on lactic
acid bacteria, characteristics and structure, use in industry and probiotic research. Chapter
2 describes the effects of lactic acid bacteria ingestion on mucosal and systemic cytokine
gene expression and secretion in a murine model. Chapter 3 investigates the in vitro effects
of lactic acid bacteria and their extracts on cytokine production in a macrophage cell line, and
in primary leukocytes from mice. Chapter 4 determines the effects of feeding a fermented
milk manufactured with different strains of lactic acid bacteria on the immunoglobulin

response to bacterial toxin. Finally, Chapter 5 summarizes these inter-related studies.

CHAPTER 1

LITERATURE REVIEW

1.1 Lactic acid bacteria

1.1.1 Historical background.- F or thousands of years, fermentation with lactic acid
bacteria has been used to produce a variety of foods with improved shelf-life, different
flavors and textures. These fermented products may contain a variety of strains belonging to
different genera and species but each having the major characteristic of producing lactic acid.
The special metabolism of these microorganisms may contribute to possible nutritional and
health benefits. Selection and development of new strains of lactic acid bacteria is currently
research of the highest priority.

Humans have used fermentation as one of the more effective and oldest techniques
for the preservation of foods. Fermented products were an important aspect of survival in
mountainous or desertic areas particularly during seasons where the ability to obtain fresh
food was impossible. Old European civilizations believed that these foods were produced by
their gods and the technology delivered to humans by them. Such a belief motivated man to
name these products with words such as “life”, “long life” and “health” (Benoit, 1981).
Lactic acid bacteria have an essential role in the majority of these food fermentation
processes. The traditional elaboration of these products consisted of leaving raw whole milk
at room temperature, therefore allowing development of the natural bacteria to initiate
fermentation. It is believed that when husbandry practice was initiated, men utilized animal
products such as milk and when this milk was not consumed fast enough, it would become
acidified thereby originating the first fermented milks. Often, a small amount of the old

product was added as inoculum.

Western Europe and Mediterranean countries have classically been the biggest
consumers of fermented milk products. This fact is mentioned in the Bible where Moses
classified them among the foods that Jehovah procured (Deuteronomy, XXXII, 14). Also
Abraham ‘8 life is linked to these type of foods (Genesis, XVIII, 8). Marco Polo, a Venetian
traveler who explored Asia from 1271 to 1295, wrote in his book “Travels of Marco Polo”:
“...they are prepared with mare’s milk and a pleasure to drink, being similar to white wine...”.
Therapeutical characteristics were recognized and reported by Arab writers such as
Abumahomet Abdullah who said that “leben” (a fermented milk with its origin in Egypt)
fortified the stomach, cured diarrhea, stabilize the body temperature, purified the blood and
improved the color of the skin, mucosa surfaces and lips (Mateos Garcia, 1984).

Elie Metchnikoff (1907) wrote in his book “The Prolongation of Life” that the long
life span attributed to eastern European populations was due in part to the consumption of
great amounts of fermented milks. He believed that overgrth of some organisms in the
gastrointestinal tract produced damaging substances resulting in aging due to the stimulation
of autoimmune reactions. It was then when he established the controversial theory about
antagonism between lactic acid bacteria and intestinal pathogens, a theory that today is still
accepted with some modifications. He evidenced Lactobacillus bulgaricus as the main agent
causing milk fermentation. It was due to these first investigations that lactic acid bacteria
became very popular. As a result of his writings, acidophilus milk and koumiss were
introduced in the United States and Soviet Union, respectively, as beneficial products for the
treatment of tuberculosis and some other diseases of the time. Kopeloff(1926) and Rettger

et a1. (1935) conducted probably the first studies describing the use of these microorganisms

in humans to cure intestinal illnesses. The study of the intestinal microflora flourished after
World War II and along with it the theory of possible beneficial effects on health by these
microorganisms and their metabolic functions, notably in infants and the elderly (Speck,
1980)

Today, there is an extensive body of literature addressing the possible health benefits
associated with the consumption of lactic acid bacteria. One of the theories to explain these
effects include stimulation of the immune system, but convincing scientific experiments
which prove this are limited. Modern molecular and immunological techniques do, however,

provide a powerful strategy with which to address some of these issues (Gasson, 1993).

1.1.2 Description yd classification,- The lactic acid bacteria produce lactic acid as
the major product from the energy-yielding fermentation of sugars. They are gram positive
rods or cocci, anaerobic, micro-aerophilic or aero-tolerant, catalase negative, without
cytochromes. They do not form spores (except for Sporolactobacillus) and are non-motile
(except for Vagococcus) (Wood and Holzapfel, 1995). Much has happened regarding
classification of lactic acid bacteria. Recently, several species have been stripped of their
dairy and enteric species (i.e. Streptococcus of Lactococcus and Enterococcus), newly
identified (i.e. Sporolactobacillus) or characterized (i.e. Vagococcus).

Traditionally, the formation of lactic acid as a endproduct of carbohydrate
metabolism defined the classification of lactic acid bacteria. But some other microorganisms
(such as Actinomyces) under appropriate conditions are also able to convert carbohydrates

to lactic acid. The development of new molecular biological techniques made possible to

more accurately study taxonomy and phylogeny of lactic acid bacteria (Lortal et al., 1997).
The current classification is mainly based on comparative sequence analysis of 168 rRNA,
DNA:DNA homology, DNAzRNA and G+C content. Gram positive bacteria are divided in
two groups, one with a low G+C content ranging between 33-50 G+C mol % (Clostridium
branch) and another with a high G+C content ranging between 55-67 G+C mol % . Typical
lactic acid bacteria have a G+C < 50 mol%. The genera used as starters by the food industry
consist of Lactobacillus, Streptococcus, Lactococcus, Leuconostoc and Pediococcus.
Lactobacillus delbrueckii, L. acidophilus, L. gasseri, L. helveticus and some other species
are mainly homofermenters (only lactic acid is produced). Lactobacillus casei and
Pediococcus could be included in another group where facultative heterofermenters (lactate,
acetate, ethanol and CO2 are produced), obligate heterofermenters or homofermenters are
found. Leuconostoc is an obligate heteroferrnenter. The genus Streptococcus can be divided
into oral, pyogenic and viridans groups and also it can be differentiated by the composition
of the cell wall, classifying it into the so-called Lancefield groups (groups A, B, C, E, F,). S.
thermophilus belong to the oral streptococci (S. salivarius spp. thermophilus). It is closely
related to the genus Lactococcus (group N Lancefield group), and Enterococcus (group D
Lancefield group).

Other genera of lactic acid bacteria include Enterococcus, Carnobacterium,
Vagococcus and T etragenococcus. Enterococcus appears to be also closer to
Carnobacterium ( unique because of its ability to grow at pH between 8.5-9.0) and
Vagococcus (the only motile group N streptococci ) when 168 rRNAs are compared.

Pediococcus halophilus was also reclassified as a new genus called Tetragenococcus, and

10

it is capable of growing at very high salt concentrations.

Originally, Bifidobacterium was considered to be a typical lactic acid bacteria, but
its G+C content (>55 mol%) and 16S rRNA showed to be a distinct group (Woods and
Holzapfel, 1995b). Nevertheless, members of the genus Bifidobacterium are still considered
to as genuine lactic acid bacteria with saccharoclastic fermentation producing lactate and
acetate, but the pathway used for hexose fermentation is different or special as compared to

typical lactic acid bacteria (Woods and Holzapfel, 1995).

1.1.3 Megbglisms Lactic acid bacteria use carbohydrates as a primary energy source.
Hexoses are degraded to lactate mainly and possibly to additional products such acetate,
ethanol, formate, succinate or CO2 (Gilliland, 1985). Lactic acid bacteria can be classified
as homolactic or homofermenters and heterolactics or heterofermenters, depending on the
pathway followed for carbohydrate fermentation and the end-products obtained.
Homofermenters follow the Embden-Meyerhoff—Parnas (EMP) pathway for glycolysis
producing lactate. Heteroferrnenters produce lactate as well as acetate, ethanol and C02.
Figure 1.1 shows a general schematic representation of lactose and galactose uptake and
conversion to lactate and other products. Pentoses also can be used, incorporating them to
this pathway as xylulose SF.

The organic acids fermented most frequently by lactic acid bacteria are malate and
citrate. Citrate is metabolized by some lactic acid bacteria to acetate, acetoin, diacetyl, 2,3-
butylene glycol and C02, when there is a fermentable carbohydrate in the media (Gilliland,

1985). Malic acid is actively fermented by most strains of Leuconostoc producing lactate and

11

Figure 1.1. Carbohydrate fermentation pathways in lactic acid bacteria.

12

 

 

 
  
  

 

     

 

 

 

 

 

 

GALACTOSE LACTOSE LACTOSE GALACTOSE
outside
cell membrane . .‘ Permease Pcrmease
inside
Lactose P Lactose
Galactose 6-P ‘ l b 1:.‘CLUCOSE ' ‘ F Galactose

 

 

 

 

Tagatose 6-P Galactose l-P

  

Tagatose 1,6-P ___ __ .— — Glucose l-P

 

 

 

 

 

    
 

Acetate
(Ethanol)

Gluccie 6-P " Fructose 6-P Glucosz
Fructose 1:6-1’ Acetyl-P + Erythrose 4-P 6'? G'“°°"at°
lAldolase -. Fructose 6-P
—> Triose 3-P Heptose-P Xylulose 5-P + C02
+ 3
Pentose-P 3
'5
‘_____ g _____,
Pyruvate 35
8
— ' _ +
Acetyl-P+ Triose 3-P E Triose 3 P Acetyl P
I l
j V Pyruvate
HOMOFERMENTATION Pyruva‘e l +

   

. l

BIFIDUS PATHWAY 6P GLUCONATE PATHWAY
HETEROFERMENTATION

13

C02, besides acetate, acetoin, diacetyl and 2,3-butyleneglycol (Figure 1.2). Citrate is present
in many foods such as fruit, vegetables and milk. Its degradation results in the formation of
aroma compounds that have a different effect depending on the type of fermented food. For
instance, the presence of diacetyl is desirable for butter, buttermilk, cottage cheese, but it is
not wanted for products such as beer, wine or sausages (Hugenholtz, 1993).

Lactic acid bacteria cannot synthesize several amino acids, and this is one of the
reasons for these microorganisms being fastidious regarding their nutritional requirements.
Lysis of proteins from the substrate where they are growing is necessary for these bacteria
to proliferate. Lactic acid bacteria possess a limited proteolytic capability, having some
proteolytic enzymes which are located in the cell wall, cell membrane and cytoplasm (Figure
1.3). Lactic acid bacteria are able to degrade proteins and use them as a nitrogen source. This
potential is exploited by using this bacteria in dairy technology as starters or as ripening
agents to contribute to the development of flavor and texture of the product (Pritchard and
Coolbear, 1993).

The properties attributed to lactic acid bacteria such as anticholesterol activity,
improvements in intestinal motility, metabolism of drugs, vaginitis, immunological status,
and the decreased gastrointestinal disorders and tumors might be due, in part, to the
metabolic activities of these bacteria. Specifically they produce a great amount of acid very
rapidly and also some other metabolic products such as H202, C02, diacetyl and the so-called
bacteriocins (Davidson and Hoover, 1993). Acetic and lactic acids can inhibit other bacteria
and act in foods as a preservative extending the shelf-life of these products. Lactic acid

bacteria produce large amounts of H202 through pyruvate metabolism. They do not produce

14

Figure 1.2. Citrate and malate fermentation pathway in lactic acid bacteria.

15

CITRATE

Citrate lyase > ACETATE

 

 

 

 

 

 

V

OXALACETATE

Malate Dehydrogenase

Malolactic

enzyme
LACTATE
V\(:)2

 

 

 

 

 

FUMARATE
PYRUVATE
Acetate
CO SUCCINATE
2
ILACTATEI
[ Diacetyl l
: PROPIONATE
Leuconostoc :
S. lactis spp. diacetylactis
I
l
I
V

 

2,3 butyleneglycol

 

 

 

16

.3583 Box 362 E £93805 .mA 95»;

l7

monaEEoneeam
"~an eEE< All 33an =«Em

:

EmuESAU

 

 

 

u :33 5»: ~39

 

     

ESE—w. 53923.:
uEEmEthQ

  

ESSA.” remnant
Eve 3: E v.

    

 

 

 

 

 

@352

 

m3: cEE< 3339a =«Em 33an owned Eomau

 

18

catalase but they are resistant to H202, while other microorganisms are very sensitive and
die in its presence. Diacetyl is also synthesized from pyruvate by fermentation of citrate. It
is known to be able to inhibit yeast and gram negative bacteria but at very high
concentrations (Salminen and von Wright, 1993). There are some other antimicrobial
substances produced by lactic acid bacteria which have a peptide/protein nature known as
bacteriocins. Klaenhammer (1988) defined bacteriocins as “proteins or protein complexes
with bactericidal activity directed against species that are usually closely related to the
producer microorganism”. Bacteriocins are heterogeneous compounds which vary in
molecular weight and many other biochemical characteristics. An all comprehensive
coverage on bacteriocins can be found in the reviews cited here (Davidson and Hoover,

1993; Lindgren and Dobrogosz, 1990; Klaenhammer, 1993; Nes et al., 1996).

1.1.4 SEW: The cell wall of bacteria is a rigid structure that
confers the characteristic shape. This structure prevents the cell from expanding, bursting
because of water uptake, and in general it protects the cell from adverse physical conditions
(Pelczar et al., 1993). Usually this structure accounts for 10 - 4O % of the dry weight of the
entire cell. Cell walls from different bacteria vary in thickness as well as composition.
Among bacteria, the Gram negative species possess a thinner wall than those of Gram
positive bacteria.

For Gram negative bacteria (Figure 1.4) the main distinctive characteristic is the
presence of lipopolysaccharide (LPS). The basic molecule of LPS consists of a lipid

component, the lipid A, covalently bound to a heteropolysaccharide of two distinct regions:

l9

duo—02$ :8 6:823 gunmen 880 .94 usurp

20

=S=e£ :3 "we
e233: 2.5:.an
gunman".
mEnzogBHEm

en . .. t t.. “a
accenEwE «Emuinim
can. 5 :aumeEEmnuOm
mafia—:2: EEOHEO
muE=ocachnnE
9:305an

 

 

guano—52m...— W a

         

.

.....

O
O
O
O
0

g 0000'.

~ ~ ~

engenai?
a.
.2

_

O .00.
.000...

~—.

econ

E

 

’o.‘....090000

@g
e

21

the core oligosaccharide and the O-specific chain (Stewart-Tull et al. 198 5). It is also liable

to contain polysaccharide chains of variable lengths. The O-specific chain is composed of
a polymer of oligosaccharide molecules in repeating units, the nature of which is
characteristic and unique for a given LPS. The O-specific chain is linked to a subterrninal
glucose residue in the core which in turn is covalently linked through a 2-keto-3-deoxy-D-
manno-octonate to an unusual lipid region, lipid A. Lipid A consists of a phosphorylated [3(1-
6)-linked D-glucosamine disaccharide backbone to which long-chain fatty acids are attached.
LPS has been recognized as a macrophage stimulator.

Gram positive purified cell walls (Figure 1.5) contain teichoic acid and
peptidoglycan. Peptidoglycan (40-90% of the cell wall) is an essential cell wall polymer
consisting of a polysaccharide backbone (glycan strand) which is cross-linked through
oligopeptides (Stewart-Tull et a1. 1985). It consist of 0(1-4)-glycosidically linked N-
acetylglucosamine residues (which are believed to play an immunodominant role). Each
alternate N-acetylglucosamine residue is substituted by a D-lactic acid ether in its C-3
hydroxyl group. This derivative of glucosamine is called murarnic acid. The carboxyl group
of murarnic acid is substituted by an oligopeptide which contains alternating L- and D- amino
acids. Adjacent peptide subunits are cross-linked either directly or via an interpeptide bridge.
This gives rise to a huge macromolecule encompassing the bacterial cell. There are more
than 100 different primary structures of peptidoglycan or murein types. It is believed that
peptidoglycan has an immunoadjuvant effect and it is capable of B-cell activation among
other biological activities. The teichoic acids are found in most gram positive bacteria,

occurring either in the cell wall or associated with the cell membrane (Stewart-Tull et al.

22

duo—3:0 :8 3583 266cm EEO .md charm

23

 

1

3

J

wcflnEwE «Emmi.

 

2

 

 

 

 

. u .3;
__ _ O.
: o.
: H .. 5.305.ng
Z W o-
: A. ..
___”___ ..
m=mmm_..._ 52..."? u.
_ _ : s
"UTE"... 9.590.55—
: :2
w_3mnm0__._.“.:__

_ “:06 U_OSO_OHOQ_I_ eon
_ . . I .

__...:..

 

 

 

24

1985). They are basically stable polymers which usually consist of a glycerophosphate or
ribitol phosphate backbone upon which glycosyl or D-alanyl groups may be substituted. The
glycerol variety may contain fatty acids. It is not known if fatty acids or other esters are
removed during cell wall purification procedures. Chain lengths of those polymers may vary
upon extraction procedures. Lipoteichoic acid (LTA) appears to be associated with the
membrane surface. Teichoic acids (TA) dissociate readily from bacterial cells in the culture

medium and by washing cells with saline solutions.

1.1.5 Industrial application .- Lactic acid bacteria have been used to produce
fermented foods with improved preservation, flavors and textures as compared to the original
food. Initially this was achieved without understanding the process and the scientific basis
for the fermentation. Now, a broad variety of foods contain lactic acid bacteria such a
sausages, ham, wine, cider, beer, pickles, milk products, olives and bakery products (Table
1.1). Notably, fermentation of milk with lactic acid bacteria yields more than a thousand
products with specific organoleptic characteristics of taste, aroma and texture.

“Starter cultures” are the specific strains of lactic acid bacteria used to inoculate a
food resulting in fermentation of that product by the metabolic activity of bacteria. There are
two main sets of criteria for selection of starters. One set is based on the rate of acid
production, capacity for polysaccharide production, ability of proteolysis and production of
flavor compounds. The second set of criteria is based on their ability to increase nutritional
value of the food, producing perhaps a beneficial health effect (Gilliland, 1985). In addition

new methods being considered for strain selection include adhesion, potential for immune

25

TABLE 1.1. Lactic acid bacteria used in the fermentation of foods (McKay et al., 1990)

 

 

 

 

 

Foods Microorganisms
Vegetable ferrnentations Leuconostoc mesenteroides
Pediococcus pentosaceous
Lactobacillus plantarum
Meat and fish fermentation Lactobacillus plantarum
Pediococcus acidilactici
Alcoholic beverages Leuconostoc oenos

Coffee and cocoa

Soy sauce

Silage

Bakery products
Sourdough

Soda crackers

Fermented dairy products

Lactobacillus delbrueckz'i

Various LAB

Lactobacillus delbrueckii
Pediococcus soyae

Lactobacillus plantarum

Lactobacillus sanfrancisco
Lactobacillus brevis
Lactobacillus plantarum
Lactobacillusfermentum

Lactobacillus plantarum
Lactobacillus delbrueckz'i
Lactobacillus let'chmaniz’
Lactobacillus casei
Lactobacillus brevis

Lactococcus lactis spp. lactis

Lactococcus lactis spp. cremoris

Lactococcus lactis spp. lactis var. diacetylactis
Leuconostoc mesenteroides spp. cremoris
Leuconostoc lactis

Streptococcus salivarius spp. thermophilus
Lactobacillus delbruecki'i spp. bulgaricus
Lactobacillus helveticus

Lactobacillus acidophilus

Lactobacillus casei

 

26

effects, and gastrointestinal colonization (Salminen et al., 1996a, 1996b, 1996c).

The dairy industry is interested in mesophilic starters capable of forming acid and
flavor compounds (cheese, fermented milks, cream butter, [Table 1.2, 1.3]) as well as in
thermophilic starters capable of growing at high cooking temperatures (yogurt, a large variety
of cheeses such as Grana, Gruyere, Emmental) (Sharpe, 1979; Auclair and Accolas, 1983).
Their main function for manufacturing these products is fermentation of sugars, protein
hydrolysis, synthesis of texturizing agents and flavor compounds and production of
inhibitory components which avoid cross-contaminations. Basically, when temperature,
water activity, pH conditions are optimal and allow the growth of these bacteria in these
products, lactic acid bacteria develop faster than competiting microflora, the pH decreases
by acid production leading to a microbiologically stable fermented product (Steele and Unlii,
1992)

The same principle is applicable in meat and vegetable products, except that in some
cases the meat industry has to add fermentable sugars, salt, and spices to adjust the product
to a more favorable conditions for lactic acid bacteria to grow (Salminen and von Wright,
1993; Egan, 1983). Vegetable products can be fermented using pure cultures of lactic acid
bacteria (cucumbers, cabbage, olives). Nevertheless, these products undergo natural
fermentation if the product is handled adequately and by establishing a designated salt
concentration, which is typical of that particular product.

Lactic acid bacteria starters continue to be used in the baking industry in manufacture
of breads such as natural sour rye bread, San Francisco sourdough French bread,

Pumpernickel, Italian Panettone and Pandoro cakes, and soda crackers (Gilliland, 1985).

27

 

SNEQSEES 3886323

 

38 38.2333 35.83083 85:58 £553 tamo>
38 88.2933 3333083 «Em—am SEE Sgwim
£83 @8883th
B8 SNEQQEQG 8:83863 tamomm
802m .38 305..» mm>m_mow:> 88.5w #2:va
Q85 .38 309$ 335:3 aux—mm 9:85
Accwoaoaové
82% .38 msote> Banana 9“me :3820
80% 309$ «was? 8M
80% .38 £595 Swazi cash
2&3 .827. snow 3093 383 8:82
38 Eaton—Sn mm 083 .<.m.D E88 Sow
EEQBEQ 88.43833
£86meth ugueeeumwhm.
53258883 883884
nicest 83.9583
nicest 388.68%th
38 £83 wgueufimvhm. .<.m . D 93 2.3 figofism
38 32.838 258383 .3.: can. ”22.5294,
“.58 «0 09C. mEmEdmuooCoaz .9850 28 Z

 

 

.692 .8338 23 88548.5: 338 essence co 8.5 .2 59¢

 

IIIIIIIII I'll-I

28

 

82.8 wsfi

 

38 £883 883m

38 289$ .838 Z £028 388
V58 83m 38 £883 @885 a8:&
38 305.3 Edam BE
82% 8285/ .683. 38¢.

88%

8:8? 088 .&&8 85288883
38 8.83 3.388888% «$8888m 883.
38 .808 .Sow 8059 A385 «868m 3285
28mg .888 .38 2853 8% 28 83mm 884

8588883.:

33$

8.83 89988:.»th

8:33.838 3383883

8:88 £00 38838 832.8883
80m .388 .088 833385me mauuouoamhm £38m $850M

858203 33238

285203 haugfimuw

28.8888 8388883

.383 9.888888%

M2288 .388888w

68% $0.0 :3 5.28
“mow .827. .38 $3 3.858858% £83m 8.8M
x=8 mo 09? m8m8mm808=2 88:00 08 Z

 

A32 .8828 8d 88-85.38 9.8. 8.5.8.8 828... mo 8.5 .2 882.

 

 

 

 

29

Lactic acid bacteria can be both useful or detrimental in brewing and wine making.
In the first case, products contain large quantities of organic acids which lactic acid bacteria
can metabolize thus affecting the final product favorably or unfavorably. The predominant
organic acids in grape wine, cider and other fruits are malic acid and tartaric acid. Malic acid
is transformed to lactic acid and CO2 (malo-lactic fermentation) producing not only the
reduction on the acid content but also a characteristic flavor. In contrast, the transformation
of tartaric acid by these bacteria leads to a spoiled wine (Radler, 1975).

Lactic acid bacteria are also of interest as biopreservation or biocontrol cultures. The
purpose is to extend the storage life and safety of certain foods by using the natural flora and
the products of their metabolism. In the United States lactic acid bacteria are generally
considered as harmless and they are afforded GRAS (Generally Recognized as Safe) status.
They are capable of preserving the food by lowering pH and by producing lactic acid as well
as bacteriocins (Stiles, 1996). This application of lactic acid bacteria has been studied in
dairy and meat products, and recently is being examined for use in minimally processed fruits
and vegetables, creating the so-called minimally processed refrigerated foods (MPR) (Breidt

and Fleming, 1997).

1.1.6 Lactic acid bactcg'a as prcbictics and mimal supplcmcnt§.- The term probiotic

derives from the greek meaning “for life”. The first definition of probiotics was as
“organisms and substances which contribute to intestinal microbial balance”, where
antibiotics and organic acids could be included. Fuller (1991) redefined the term as “ a live

microbial feed supplement which beneficially affects the host animal by improving its

3O

microbial balance”. There are numerous reports of potential health and nutritional benefits
afforded by lactic acid bacteria (Gilliland, 1990; Marteau and Rambaud, 1993; Lee and
Salminen, 1995; Hughes and Hoover, 1991; Tanaka, 1995; Teuber, 1991; Sanders, 1993;
Wood and Holzapfel, 1995). These include studies of lactose digestion, cholesterol
metabolism, diarrhea] disorders (Gonzalez et al., 1995; Salminen et al., l996a, l996b),
prophylaxis of intestinal or urogenital infections (McGroarty, 1993), immunomodulation
(Perdigon et al., 1986a, 1986b, 1987, 1988, 1990, 1991a, 1991b, 1992, 1995) and oral
vaccination (Wells et al., 1996).

To better understand previous results and to rationally develop new healthy foods or
even drugs using lactic acid bacteria, the pharmacology of these microorganisms must be
determined. This includes precise knowledge of the active constituents of lactic acid bacteria
responsible for each potential effect, the target site in the body, the ecological conditions
necessary for the activity of the active constituent, the requirement for viability, the
pharmacokinetics of ingested lactic acid bacteria as percentage of survival at the target site,
the concentrations of bacteria, the duration of the passage (Fuller, 1991; McCann et al.,
1995). Also research must consider possible negative side effects of lactic acid bacteria.

Factors required to maintain high viability include the proper ratio of strains and
inoculum levels, suitable growth factors, and control of pH and redox potential (Shin, 1997).
Once bacteria are ingested, they can be destroyed by the acid in the stomach and bile in the
intestine. Their survival depends on their intrinsic resistance but also on the host and the
product in which they are ingested. Yogurt bacteria (Lactobacillus bulgaricus and

Streptococcus thermophilus) have a very poor intrinsic resistance to acid and bile.

31

Lactobacillus acidophilus and Bifidobacterium are more resistant but there are numerous
strain differences. Most information concerning the pharmacokinetics of lactic acid bacteria
relates to the human gut. After ingestion, L. acidophilus, L. casei and Bifidobacterium can
colonize the small bowel and colon and reach high concentrations (106 - 108/ml or more)
(Marteu and Rambaud, 1993). Lactic acid bacteria administered exogenously can persist in
the colon for at least 3 weeks or more.

Some of the lactic acid bacteria used as probiotics may offer an alternative to the use
of antibiotics to improve growth and performance of livestock. Much of this interest has been
due to the concern over the use of banned antibiotics for this purpose (Gilliland et al., 1990).
Beneficial effects have been observed for some strains of lactic acid bacteria administered
orally in pigs (Tortuero et al., 1995; Abe et al., 1996), calves (Nousiainen and Setala, 1993)

and poultry (Jin et al., 1996a, 1996b).

1.1.7 W9: Relative to immunity, the effects of lactic acid bacteria
are intrinsically related to the diet and nutritional status of the individual. There is a strong
relationship between nutrition and immunity created by the presence of mechanisms
responsible for decreasing the resistance to infection. Specific nutrient deficiencies can
interfere with nonspecific defense mechanisms that include flora, anatomical barriers (skin,
mucosa, and epithelium); secretory substances such as lysozymes, mucus, and gastric acid;
the febrile response; endocrine changes; and binding of serum and tissue iron (Solis-Pereyra
et al., 1997). Protein deficiency produces impaired antibody formation, decreased serum

immunoglobulin, decreased secretory immunoglobulin A, decreased thymic function and

32

splenic lymphocytes, delayed cutaneous hypersensitivity, decreased complement formation,
decreased interferon, and effects on nonspecific mechanisms that include anatomic barriers
and secretory substances such as lysozymes and mucus (Scrimshaw and San Giovanni,
1997). Deficiency in vitamin A increases susceptibility to infection, decreases thymus and
spleen sizes, reduces natural killer cell activity, lowers production of interferon, impairs
delayed cutaneous hypersensitivity and lowers lymphocyte response to stimulation by
mitogens. B-carotene can stimulate mitogenesis in lymphocytes and increase human natural
killer cell and T helper cell numbers. Vitamin B-12 and folic acid interfere with cellular
replication, antibody formation. Deficiency of these vitamins can produce anemia and cell-
mediated immunity depression. Vitamin C deficiency also decreases immune function,
besides decreasing iron absorption, but there are claims of a favorable effect on infection
with massive doses of vitamin C, studies which have not been scientifically confirmed. The
killing power of lymphocytes is reduced if the dose of vitamin E is not appropriate. Clearly
impaired phagocytic killing power has been reported with iron and zinc deficiency, as well
as cytokine function or production, immunoglobulin and B cell function (Schrimshaw and
San Giovanni, 1997). All these effects are more severe in the elderly (Lesourd, 1997;
Paavonen, 1994; Schmucker, 1996). Thus, when studying immune function, it is very
important to design experiments in which the nutritional status of the population under study
is controlled and equilibrated.

Several studies have tried to decipher the events which occur after the first contact
of ingested lactic acid bacteria with the immune system in the gut-associated lymphoid tissue

(GALT). Perdigon et a1. (1990) showed an increase in levels of immunoglobulins against

33

Salmonella in the intestinal fluid of mice after oral administration of L. casei, L. acidophilus,
L. bulgaricus or S. thermophilus and challenge with Salmonella typhimurium. De Simone
et al. (1987a, 1988b) reported that yogurt or heated yogurt, when administered to mice,
increased the percentage of B lymphocytes in the Peyer’s patches. The antibacterial activity
of the Peyer’s patch cells is also increased by viable yogurt (De Simone et al., 1987a). These
effects might be due to the activation of the macrophages and non-specific immunity by cell
wall components of lactic acid bacteria.

In human studies, De Simone et al. (1986, 1987b) showed stimulation of interferon-y
(IFN-y) production in supematants of blood lymphocyte cultures when lactic acid bacteria
were administered as a yogurt filtrate to the cultures. At very large doses (10“ - 10‘2/ day)
of yogurt, serum IFN-y is elevated (Halpem et al., 1991). While lower doses of yogurt did
cause a significant increase of serum IFN-y, blood lymphocytes did have an increased
capacity to produce IFN-y in vitro. Because such a stimulation of IFN-y might exert both
beneficial and detrimental effects, more experiments and trials are needed to reach a
conclusion (dose-response studies, clinical end-points, etc.). Tables 1.4 through 1.6
summarize the recent researches on lactic acid bacteria modulation of the immune system

which is relevant to this dissertation.

1.1.8 5ch cflects and tcxicity cf lactic acid bacteria.- Lactic acid bacteria have been

observed to produce side effects after intraperitoneal injection, causing septicemia,
endocarditis in mice (Okitsu-Negishi et al., 1996), systemic infections, fever, and arthritis

in rats (Wilson et al., 1993; Blancuzzi et al., 1993). However, during oral administration no

34

complications have been shown in mice (Perdigon et al., 1991a). In humans, cases of
clinical infections are extremely rare, but the fact that they appear in several publications
suggest that lactic acid bacteria might act, rarely, as opportunistic pathogens. Nevertheless,
all severe cases were reported in immunocompromised patients (Abgrall et al., 1997), and
only a few cases describing local infections were produced in immunocompetent individuals

(Gasser, 1994; Donohue et al., 1993).

35

 

mag r? «o 2:28—

mog .. 3 #0 :thom
3&3 :3 8 2888 on
33 :3 «o 9228“

R2 .22 a 2:22

22 .22 a... 8225 on

32 .22 2., 22282.22:

32 :3 Ho agom

0&3 :3 Ho «5832

S2 .22 a 22%

~93 :3 Ho sou—E

33 :3 Ho :82“?

2838.62-25-223 2:2 22 38865

.2323?ng
3629.5 0985 ac .m_mo§oowmnn 32822:

2.2.2 28 28 22 822202

8:526 :82 .0388 £83.28 3
<2 2:8 22228... 22 .03 422 82855

coo—o 25:22.82 5 5.227: on :8 m @8385

<3 288 39 a. 523 w 98 «2? 3382:

28233 8.62295»—
owfifio o: 53222 238me 32322:

v28? 8: 28.5282 3 322805

2892322 «8:805

gem name» E 8280.25 32228 295m

23:86
2225850wa 28 22260.2: 223.» Each

2:252 .222. a m .3822 do .328 .2

223 in .22 2 _ 22222
53.223223me +23 2 b #3 2332.63.66 d

u 2N .5222 2 com
+ 2.6 228:5 m 223 23882 822283

20 :2-22 .228 do .828 .2

u 22 23338 2
22222 .m a. 2002 2222828. q

:35 .m 325:2 + 22a 20.223 2222.2 .m
a. 3.2828 .2 22 mm: .2 can 2.223 m

2223 2 22: + 2022 2 b @239 22.3%
- .8 q s 22222 .2 12238222 :2 as 2 m

20.2 2232838 .2 co 2 22 22 22 5 2 N

E can 2 a a .22 2.2222
Eatfioaaohmbm @ wzzxmosoo Q as x 2.

Eye: 33328 q .23 35222 2° 2

222.2 2 S $222232 2
+ 3.2.828 .2 2883 + 8222.22 25

2822823 2225

£00 803 28:52

2:223;

28:53

2:020 2

333 cm

316w mm

93v<

332 cm

5895 mm

E S A 25qu R

 

35.52

.2an

$22. .3.—2:227”.—

9.322

.2883 Bus 262 .3 562385022882 no 8.682 2255 258264 mqmflw

36

 

3m 2
225883 98 .2022

22 2 2-2 282602 2.2 82385

2.2 .32 2.2.2 225:2

52:? a b + .23 u 2. .5202 . 22.5.63 622 2:8
.. 035% .285 W .233Q238 Q 2.222305ng .m. 2332:2208 v85

$233235me M. 23 x _.v.m.m + 232.833 4

 

S2 .22 s 228.5, .m.<.2 3 .258 285 39 5 28238222 02 3.2 2 2-2 .522: 2 .20 2 .2858 5282 232
82 .22 a 2:2: <2 22 82885 322 .2022 a 2 do .228 .2 £8 2.85 228::
2&0 22 2 m .23 2
82 .22 2. E22 5E 82855 228 35.2222 222228222 2.63 q 2:8 822. 22:52
822.83— 2.33— .222. 355.2322 .835

 

 

 

8.283 .1 mums.

37

 

32 :3 3 .333:

32 .23 30 32238:.
$3 .23 :0 023m

32 :3 3 2852

.32 :3 3 83333»
32 :33 33332

$2 .23 00 2032382.“

wwm: .23 :0 03:32

82 :3 3 8333.:
32 :33 5338

N03 .20 :0 30w830m

N03 .23 :0 $633335

52 :3 3 83333

.332 :3 3 8333

30303 30333333 3823033500 3. 3830305

303033: 83:30:02: 383202
w2=2 323383.32 333 20030033: 383205

82032230: 3830305

083303023 3 3203: 03382.8 38303032

30002 w33233338$3200 3830305

323203333 2 <3 333822 3830305

233033 33332 2 8:00 @3208 E 383205

328:3 09330303:— 3832303 3830305

7:: 3:3 6&7; 33338 3830305

303032 83330303:— 38300303 .23 3830305

303033 0333030333 38:03:03 .23 388305

30302.2 .933: on £3383:ng .m

33325333 3 .8133 32. .333 .3

83035:: no .2 . .680 .

Q

83033:: 72.0 8300:5338 . 830 Q
830:3»:3 md .2 . 800 q
2333 :03 3338 01. .3 .338 .3

83033330 33 x N .—

.3 b 480 .Ewcud £55838 d 3.32.6333 4N

888330 :2 x N3 .330 333333.33 .3

8:03:38 33 x N A
.3 wé 43:0 .wasimnfifiu d :0 .860 Q

3.333 3.3 3338 an
.883 .0933» . Ema .3823M30 .3

30:3 x «A .3 w
.2 3:3 380 65385335 .m a. «ENEMEQ Q

33.2 x 2 .3 w .3 .38.: .3 a. 3.8.5833 .3

832

83,:
022

022
00:2

00:2

83: £3.23 33.33
83. 33:23 33333

8E 23:23 33:33

00:2

833 $3.23 3333:

023H 8QO o—mfiom

 

330.50%

8.303—

3wmao3 fiEoEv—oanfi

{I

3.38

"‘

2:803 203 0302 .3 30333333003383: :0 82338 33233 «3000: .n._ 033,—.

38

 

82 :3 5 «>25

32 :3 3 832$

33:: “a 83388
$2: «a: :3 :0 820m

mom: in :0 85$ on

cam: in 8 ofiEcO

:2 r: 8 820m
$2 a: 3 :58:

33 8:50 :5 28:32
33: 52:8:

82 :3 :0 28:3“:
82 c: s 835:

mama—floss: 85:8: 5 7d 8885

8885 mogoagwh :5 £8 «Ems:
8885 35.353on .m..::a-<wH beaoom

95.2.: 93 E 3.855

338% bags: 8808 =8 H

EMS—:8: 2.80% a 058%-:0: @8885

03 52:2: 8:89? 888::

8:9: 3

42:5 32:. .22: an: .93: ESE
8:268: 83:5“ 888::

88:8wa Ho 8838:: 888::
8658 £983 :o:a::::m .m: 8885

vizaafisud mo 5832.6 :5 :«3 =8
€255 888:: RE .«o 8:38: 3:80.58:

53:95:.» 8 853:8: 88.83

23%: o: a 2 .38 .32: ”8.283 4

as? 572%: .352 x N a N .38 6:8 .:

Echo: x 3&6 .mommnqobfi: 8:38:
i Q. A: «3385me .m 82.8%}: .4

8:? £1 GE .8 J3 v 4:25 :8 35%;: .m
2.5%: x S d N “:8 .33%E§ .m
.nSERchu .4 68.2633 d .88 4

4.? ES: 22 x _ 8908 m: can 5:5 m
480 6328:8872 + 5.39388:

Eu no: x 5m .: m 8:: 885889: 88:8:
.«o :o:8=ou d: .3353: .m .Eamfix .m

Bu :2 x N 6 h .8me .383 .m
65385.85. 9mm

”3.633: .M 6.83 d £35.20 :8 .503. 4
{8.83.8 Q . :88 Q .mnutuMNS dam
:«8358 d .23M23 .m H8:3: 88056.: N.
:do go?

658% a. :9; =8 .222 x a 35:8:
x3 w-~ 4:8 . «33:56.8: d . 539:8 .m

332 x S a h :3 gsmofius. m

00.52

8:: 28% 230:

8:: 35m

8:: Qmim 2:2

8:: 05175

8E 05.2: 22...:

8:: Smim 2:2
8:: o\m\.~<m 2.22

8:2
8:2

8E: 8QO 258$
8:: 2QO ofifiom

 

85.52

 

3:53—

:mmnu: Eng—5.8.7:

30.3.5
€288 .3 mama

 

 

39

 

baa“
a .881: 5 838m

82 :3 a 888m

33 in 8 256m

mag in 5 23:83.5

8% .8 s 8328

~93
r? 8 $622 on $32

82
.8 8 28$: 8 882

23 626$ mommies—WE 13:8qu 8382:

mcmzofizmzm
822383 85%me 98 CE 888 68885

:23“:on 8:59

bans—Em x33 :5 028%“: 0288an
Ewwaoiu £53 :8 wSEmoEuu Q @8885

38:8 5 5928? £28

wfiozcoa G3 £3 n£3 «Ema—n 39885

8% E 86898 ~=owEméSw 8322:

Son 5 36328 5353.28 @8365

£32 an N a w is giants a. $8 8

v 2 £85 :08 mo
ecu: x m .38 £593 w 8583 Bow 283

:55 :8 cocoon: .flEQE .m

888 :8, =8 8m $285 a? S
.330 22 x w .EaM23 .m 8 8.523858 4

c b .m ,N
3083 x NA .856» 8 Eng .4 .mzsimchfic Q

8:28. SEMEM womaflwfio .u w
33838 q a. $8 q 82% MWS x 2

:8 M 8 nmtmmgagcs Q woman—:30 6 w
$3838 .q a. $8 q .38 M"A: x 3

8:: 2QO 220m

8%

8:2

00¢: £QO ofifiom

8:: 2:93 mmrsm

8E. 833 3:6

8? 833 8%

 

3:383—

mu—ao.“

:Wfioc 3.5810ng

«835m

6.883 .3 359

 

 

40

 

82 :3 3 «388:2
82 :3 3 38885

32 :36 032

cog :3 8 ES

33 :3 8 8E8

3383 M2 33 $25 $885
on: .8129 .1: 83085

88.885 So .50 $3

czwzoxcumég
<3 88% doe—am detach—ca @8385

888 E E22928 @393 38385

omnoaBBwo ”A: x w
c 2.5 480 .QEEQQEES W 58.2833 .‘N

3002 .3 33 E .35 $8.838 q

E v .8335 083885 a o
$8 28 23.33% q .32 .838 q

#3 v .c w E undo—Eu SNESENQM + 20%

“58:35 38 93K 438 EU w:

2: + e .5 8 8m £88 23 8m x N .38 .8“
.2325; $8 q .2582st 28.8.8:
3°33 9.538585“: + 32% 3:qu a

£8 52% 0552

933

83m

3sz

o\m..—<m

 

35.3.3—

233.

$.32. 3.885%”.—

30.3.5

6.283 .2 39¢

41

 

£63 :3 6 552
033 :fi 8 552

5555858 \3 5:535 0:385 vegansm
mad .N-d @8383

0523 u N 455.3 ”2-02.de Eatmgouaohgm
23:5 c N 4an wofiegssaests w

 

«33 :R 6 F52 53525—0 58% 526565 2585 33205 0555 c N .255 ”2-02 .dnm 25.83803 :8 «8055:. 74m
933 :3 8 552 855558 \B 58255 259.6 Bonandm 0523 c N .555 ”3-03.53 Eztfiugohgm
032 .aa 3 :52 on: .UQZP @8385 05:3 v N 5an ”3-9345355ng w

«33 :3 8 552 55858on \B 55535 338.6 cougncm 05:3 6 N .555 ”3-03:5; 52.83893 0505285 EvoN >33
.095:on Eztzfiigo .0 555685 w: ONV

83 :3 go 5:032 53580.5«3 55583 $53882 m 2 .mmq + . 5:532 .m £33538 Q emanaouowa 3.?

£8 m 5:05 B

5&3? .525 mnemon—

32 :3 8 30wa >90 53» 85%2 2.85:5 3.0.365 .5 cm .30 .0 3 . tamam .0. 523.0354 .83851w 2:52

82 :3 a «3355 @592": 3265 a N5 isms 2: 35.08: 45.35% q mowégobma 382.com

83 :3 6 mod 3 E 33805 3835 =55 =8 E 02 £9» :8 MES .m £8 683 29:5

£8 25: :55

£28 :83 €9QO

82 :E 6 82 282596 $0.58 5:585 Eu: m0 .03 .0m .30 6.259835 .m 8&8:2& 2:52

£8 533% .950 53 =8 .69 £8 £8 sea {95

33 :R «o 288: E8 52% 0o 555205 30858 A030 :8 20:3 . £58535 .m .8§5_% 0552

8:339: 8:53— :wmuoc BEST—03H 80.3.5

 

 

 

5563 Bus 0.58— 3 5235589555 .8 865m 93> 5 580m 6.2 mafia.

42

 

033
£53 .33 .oaoEEoQ

32 .3320 a. 83>
$2 :3 a «3339
82 ca 3 8:882

82 i s Eusizom
82 4a a a2;

82 re s axiom

5A.": 3885

mowmnmobmfi 3 58258 0:238 @8885
c3385 <ZME 5.2% .595 ~

2%: 6.1: 8.2 @8355

RTE/E. d7d $2": 3885

5.2 @3385

5.2 .1: 8,555

:Ewi m; KSERQEKNS .m. + “52.5me .4
36.5 .m

Saam 4 8:53; .8 2E3 2
ESQ—NA: £583 Eon 262 mo 885 3
30%: x N £883 Box 0:820 38.8 n
15w: 3 ,mcouoma 39883th

:5»: m .Bmmb uaotqug .‘N

838353 888
mo§5a§_ 8:2:
mowwnaouomE Scam
02m

E52 x N 035
:5 2: 02m

252 x a 02m
38 Mao—Sacco:— v85

 

85.5%:

3:504"—

:wmmoe Rune—3.3%”.—

.323
3288 6A mAm<H

43

1.2 Gastrointestinal immune system

The gastrointestinal system consists of stomach, small intestine, and large intestine.
Defense against infections can involve nonimmunologic mechanisms such as gastric acidity,
digestive enzymes, peristalsis, mucus, enteric microflora, and the self-regenerative capacity
of the epithelium (Pestka, 1993). Specific immunologic responses are mediated by the gut-
associated lymphoid tissue (GALT) (McBurney, 1994). GALT consists of lymphoid
follicles (Peyer’s patches) located in the small intestine, appendix and large intestine,
isolated follicles, the mesenteric lymph nodes, and a diffuse tissue comprising the
intraepithelial lymphocytes (IELs) and lamina propria (LP) (Shanahan, 1994). Peyer’s
patches consist mainly of lymphocytes of type B, T, macrophage and accessory cells.
Intraepithelial lymphocytes are essentially regulatory and cytotoxic T cells. Lamina propria
possesses mainly plasma cells (lg-secreting B cells) and memory T cells. The Peyer’s
patches are the central focus for the induction of T and B-cell responses following an oral
immunization. These organs lie below a specialized layer of epithelial cells called M cells.
The mucosal immune response is initiated by contact between the antigen and the
lymphocytes and antigen presenting cells in the dome of the Peyer’s patch. Here, once the
antigen has traversed the M cells, antigen presenting cells within the Peyer’s patch can take
up the antigen and present it to adjuvant regulatory T cells. These B cells expressing IgM
or IgD on their surface upon stimulation proliferate and differentiate into lymphoblasts
expressing IgA on their surface. The response induced in Peyer’s patches leads to activation,

switching, proliferation and differentiation of B cells under the control of T cells and other

44

accessory cells (Ogra et al., 1994).

The mucosal immune system B lymphoblasts derived from the GALT can migrate
to respiratory, genital tissues as well as other sites in the intestine. Lymphocytes migrate
fi'om Peyer’s patches via the mesenteric lymph node into the efferent lymphatics, thoracic
duct, and the systemic compartment to the lamina propria located beneath the intestinal
epithelia. These lymphoid cells which originate at the inductive sites can enter distant
effector sites as lamina propria of the intestine repopulating it. At these effector sites B cells
proliferate and undergo terminal differentiation to IgA secreting plasma cells in response
to antigen and fiirther signaling by cytokines produced by T cells and macrophages. Lamina
propria plasma cells synthesize monomeric IgA which eventually is joined by a J chain to
form dimers or longer polymers. Polymeric IgA binds to the secretory component (SC)
which is the epithelial receptor responsible for the transportation of this immunoglobulin,
being then secreted across the epithelial cell into the lumen (Pestka, 1993). This complex
IgA-SC binds to the antigen present in the intestinal mucus and lumen, avoiding antigen
access to the mucosa and producing its degradation. IgA secreted by B cells in the lamina
propria and released to the intestine follows later the bowel content or goes to the blood
stream and raises the serum IgA levels. Although Peyer’s patches are primarily inductive
sites for IgA response, IgG+ cells are known to be also present in these organs. During this
process cytokine production can be also stimulated, controlling first cytokine mRN A
formation and resulting eventually in an elevated translation.

Of particular relevance to this dissertation are the recent reports indicating that B-

cell activation, switching, proliferation and differentiation to IgA synthesis as well as

45

cytotoxic T-cell development are regulated in the mucosal and systemic immune
compartments by several cell types and their cytokines (McGhee et al., 1989). Cytokines
produced by activated macrophages and mucosal T cells also mediate the gastroimmmune
response. There are two different types of T cells: T-helper and T—cytotoxic cells. T-helper
cells can be classified as Thl (producers of IL-2, IL-3, IFN-y) and Th2 (producers of IL-6,
IL-4, IL-5) (Pascual et al., 1996). T lymphocytes located in lamina propria and Peyer’s
patches are capable of cytokine production such as IL-2, IFN-y, IL-4, and IL-5. T cells
present in mesenteric lymph node have a potential for IL-4 and IL-5 production, and to
some extent for IL-2 and IFN-y. Activated macrophages are able to secrete IL-l , TNF-oc, IL-
12 and IL-6, as well as some other mediators that can participate in the immune response
such as nitric oxide. Some of these cytokines mediate proliferation, switching and
differentiation of B cells to become IgA plasma cells and up-regulation of the expression
of secretory component (Shanahan, 1994).

The most likely site for interaction among immune cells, food antigens and lactic
acid bacteria is the GALT, where the intestinal microflora is also a source of antigenic
stimulation. Possible targets are the Peyer’s patches, via specialized epithelial cells called
M cells (Pestka, 1993). They are more permeable to luminal components and as such
encourage interaction with underlying immune cells. Bacteria signals can also act directly
on the proximal small intestine, increasing the proportion of intraepithelial cells (Muscettola
et al., 1994). This lymphoid population can produce cytokines which can act locally or alter
the systemic reactivity of the host.

Thus, numerous factors of host and of microbial origin participate in the

 

 

 

 

46

development of an individual gastrointestinal ecosystem and in the generation of gut
immune responses. The composition and number of microorganisms in the gut are mostly
influenced by bacterial physicochemical activities. There are specific substances excreted
by some bacteria which inhibit the growth of others (bacteriocins, organic acids, short-chain
fatty acids, H202), allowing them a better adherence to the gut epithelial cells, forming a
local microbial community which will determine for instance gas concentration and
composition, the pH and reduction-oxidation potential, and the mucosal secretions
(lysozyme, cytokines, immunoglobulins) in the gut (Sanders, 1993). It is very difficult to
have available data about interactions between different niches established along the gut in
a particular host and obviously there is great difficulty in getting individuals with a similar
or identical GI ecosystem. The microbial ecosystems are individually different inside the

gastrointestinal system even in cases of animals with a similar origin or/and husbandry.

1.3 Rationale for this research

A number of published studies suggest that lactic acid bacteria and their fermented
products have beneficial effects on the health of human beings due not only to the
nutritional properties of such bacteria but also due to their action on the immune system.
Much interest has been focused recently on the immunological effects of these bacteria.
Several lines of evidence using the mouse model suggest that probiotic bacteria could
enhance immunity both locally on the mucosal surfaces and at the systemic level.

To enhance immune function in severely immune-compromised individuals, high

47

doses of IFN or strong immunopotentiators such as S. pyogenes (OK432), Bacille bilie de
Calmette-Guerin (BCG) have often been administered. However, repeated use of them has
been reported to cause severe side effects such as fever, malaise, depilation or depression
in some cases (Kishi et al., 1996). Oral administration of lactic acid bacteria could be an
alternative way to stimulate host immunity and stimulate the systemic immune response.
Immunological effects of dairy cultures reported previously in our laboratory and by others
suggest that some strains of lactobacilli as well as bifidobacteria (Marin et al., 1997a,
1997b) or their cellular components can stimulate macrophage activation in animals. As of
now, the biological significance of this enhancement in macrophage activity is not clear.
Researchers have made the assumption that lactic acid bacteria are beneficial to
humans and most of them have designed experiments to prove this hypothesis. There is a
lack of studies relating to a clear identification of most of the strains used and consequently
the claims of positive effects are not well sustained. Furthermore, not all data reported in
literature are based on well designed and controlled experiments and often lack statistical
analysis. Research needs for specific target areas have been suggested by a panel of experts
on the subject (Sanders, 1993), recommending that further research on immune system
stimulation must use "focused animal studies prior to human studies until a working
hypothesis is obtained". There is a clear need for critical study of the effects of strain
selection on health, and also there is a lack of clear mechanistic information about how
cultures could affect immunological function in leukocytes of gut origin (Peyer‘s patch,
mesenteric lymphoid node, lamina propria, intraepithelial lymphocyte cells). Elucidation

of specific effects after ingestion of lactic cultures on leukocytes found in the immune

48

system is becoming a critical need. Recent advances in the immunology field may now
allow us to explain the reasons for immune enhancement by consumption of dairy cultures
that have been observed previously. Thus, the dairy industry could be provided with a basis
for choosing species and strains to optimize immune enhancement and this could be used
in new or improved fermented dairy products.

In this research, I hypothesized that lactic acid bacteria have an immunomodulatory
effect in the gastrointestinal system and that the immunopotentiating activity resides in the
cell walls of lactobacilli, streptococci and bifidobacteria. To determine the effects of lactic
acid bacteria on immunity, in vivo feeding studies were conducted using the mouse model.
Secondly, I conducted in vitro studies, where the conditions were defined and kept
constant. By exposing clones and primary leukocytes to the whole bacterial cell and to
different fiactions of lactic acid bacteria, I attempted to identify the cellular components that

possess the immunopotentiating activity.

CHAPTER 2

EFFECTS OF LACTOBACILLI, STREPTOCOCCI AND
BIFIDOBACTERIA INGESTION ON MUCOSAL AND SYSTEMIC
CYTOKIN E GENE EXPRESSION AND SECRETION IN A MURINE
MODEL

49

50

2.1 ABSTRACT

Lactic acid bacteria are essential for the fermentation of products such as cheese,
buttermilk and yogurt. An increasing number of functional foods and pharmaceutical
preparations are being promoted with health claims based on the potential “probiotic”
characteristics of some of these bacteria and on their capacity for stimulating the host
immune system. However, the immune effects of oral administration of these microbes has
not yet been well defined. In this study, I tested the hypothesis that cytokine gene expression
and production in mice is affected by orally administered lactobacilli, streptococci or
bifidobacteria. The specific objectives of this study were to evaluate the effects of in vivo
exposure to viable and non-viable strains of lactic acid bacteria (Lactobacillus acidophilus,
L. bulgaricus, L. casei, L. gasseri, L. helveticus, L. reuteri, Streptococcus thermophilus and
Bifidobacterium) in mice on: (1) cytokine mRNA expression in systemic (spleen) and
mucosal (Peyer’s patches) lymphoid tissue, (2) cytokine and nitric oxide production by
mitogen-stimulated Peyer’s patch, spleen and peritoneal cell cultures and (3)
immunoglobulin levels in serum and fecal pellets. Growth rate of mice as well as
immunoglobulin levels were not affected by treatments. Although basal cytokine mRNA
expression in spleen and Peyer’s patches was not affected by repeated lactic acid bacteria
administration, single exposures to certain bacteria altered subsequent mitogen-induced
cytokine and nitric oxide production by peritoneal cells. The results suggest that lactic acid

bacteria may alter leukocyte responsiveness to an activation signal in a strain dependent

manner.

51

2.2 INTRODUCTION

Almost a century ago, Metchnikoff (1907) suggested in his book "The Prolongation
of Life" that consumption of fermented dairy products resulted in improved health and a
longer life. Lactic acid bacteria are mainly associated with fermented dairy products such as
cheese, buttermilk and yogurt. Today, an increasing number of health foods and so-called
functional foods as well as pharmaceutical preparations are promoted with health claims
based on the “probiotic” characteristics of some of these bacteria. The most evident effects
of probiotics involve changes in viable counts of microorganisms in the intestinal flora after
ingestion, which can be caused by competition for adhesion sites and nutrients between the
ingested microorganisms and potential pathogens (Rafter, 1995). Another known mode of
action for probiotics is production of antibacterial substances. Some of the health effects
attributed to lactic cultures are improved absorbability of certain nutrients, alleviation of
lactose intolerance symptoms, improved metabolism of some drugs, serum cholesterol
reduction, improvement of intestinal motility, anticancer effects, creation of an antagonistic
environment for intestinal pathogens by production of inhibitors, inactivation of enterotoxins
from pathogens, alleviation of constipation, relief from vaginitis and stimulation of the
immune system (Gilliland, 1990; Fuller, 1991).

T cell regulation, IgA synthesis and cytotoxic T cell development are tightly regulated
in the mucosal and systemic immune compartment by several cell types and the cytokines
they produce (McGhee et al., 1989). The gastrointestinal immune system is a critical site of

host defense (Hanson et al., 1983) formed by Peyer’s patches, lamina propria and mesenteric

52

lymph node, which represent the mucosal compartment, while spleen and lymph nodes
dispersed along the body represent the systemic immune compartment. The intestinal wall
contains macrophages and over half of the lymphocytes (T and B-cells) present in the body.
Differentiation of B cells to committed IgA producing cells requires T helper type 2 cells
(Th2), which secrete IL-4, IL-5 and IL-6. In contrast, T helper type 1 cells (Th1) can secrete
IL-2 and interferon-y (IFN-y) (Pestka, 1993). These matured B -cells produce mainly IgA
which is located in the Peyer's patches and lamina propria in the small intestine, or travel to
spleen using the systemic circulation. The secretory IgA system acting by binding microbes
and toxins, is the biggest component of the gut immune defense system. A number of
investigations suggest that lactic acid bacteria immunopotentiate this gut system (Perdigon
et al., 1986a, 1987, 1988, 1990, 1992, 1995; Saito et al., 1987) via activation of the
macrophage fraction. Macrophages can regulate T and B cell activity by activation and
secretion of IL-1 and tumor necrosis factor-a (TNF-a) (Abbas et al., 1994). Increased
secretion of these two cytokines by peritoneal macrophages occurs after intraperitoneal
exposure to Bifidobacterium Iongum, or B. animalis (Sekine et al., 1994) and Lactobacillus
bulgaricus, or Streptococcus thermophilus (Solis-Pereyra et al., 1991). Even though the
peritoneal cavity surrounds the gastrointestinal system, the exact relationship between these
two components of the immune system remain unclear.

The effect of oral administration of lactic acid bacteria on cytokine expression and
production is not yet well defined. In general, cytokines are considered a diverse group of
protein hormones produced through the effector phase of an immune response (Abbas et al.,

1994). Their synthesis is initiated by transcriptional activation of DNA, mRNA production,

53

and translation of this mRNA to proteins (i.e. cytokines). Once synthesized, cytokines are
usually secreted as needed, playing a very important function in host immunity.
In this study, I hypothesized that cytokine expression and production in mice is affected by
orally administered lactobacilli, streptococci or bifidobacteria. The objectives of this study
were to evaluate the effects of in vivo oral exposure to viable and non-viable strains of lactic
acid bacteria (Lactobacillus acidophilus, L. bulgaricus, L. casei, L. gasseri, L. helveticus, L.
reuteri, Streptococcus thermophilus and Bifidobacterium) in mice on 1) cytokine mRNA
gene expression in systemic and mucosal lymphoid tissue and 2) cytokine production by
mitogen-stimulated Peyer’s patch, spleen, and peritoneal cell cultures. These latter effects

were further related to immunoglobulin and nitric oxide production.

54

2.3 MATERIAL AND METHODS

2.3.1 Microorganisms

Representative cultures of L. acidophilus, L. bulgaricus, L. casei, L. gasseri, L.
helveticus, L. reuteri, S. thermophilus and Bifidobacterium were obtained from three
different sources: American Type culture Collection (Raleigh, NC), Dr. T.R. Klaenhammer
(North Carolina State University) and Sanofi Bio-Industries (Waukesha, WI) (Table 2.1).
Initially, lactobacilli and streptococci were grown in MRS broth (De Man et al., 1960) while
bifidobacteria were grown in MRS + 5% (w/w) lactose (MRSL) broth. All organisms were
incubated at 37°C for a period of approximately 15 h. A 1% (v/v) inoculum was transferred
to fresh broth, and this was incubated for 15 h at 37°C. Bacteria were recovered by
centrifugation at 1100 x g for 15 min, resuspended in 10% (w/v) non fat dry milk (NF DM)
plus 10% (v/v) glycerol (cryoprotective agent) and frozen at - 80°C for long time storage.

Growth curves were calculated for each bacterial species, by measuring every 30 min
the percentage of light absorption at 650 nm in a newly growing culture, which was kept at
37°C after each measurement until the turbidity no longer increased. These measurements
are directly proportional to the cell concentration in the culture and can be correlated with
total number of bacterial cells. Aliquots at each time point were plated in MRS agar for
lactobacilli, modified ST agar (Lee’s agar) (Lee et al., 1974) for streptococci and neomycin
sulfate, paromomycin sulfate, nalidixic acid, lithium chloride (NPNL) agar (Teraguchi et al.,
1978) for bifidobacteria in order to determine bacterial counts at these same points. Agar

plates were incubated at 37°C for 48 h aerobically for lactobacilli and streptococci and

55

 

5» $8.33 sumac—:55 «2%

a: £58m 580:8 0530 25 gouge
:5 again? $535.05 «2%

oz 553 Saga: 3% 9505 €02

a: .238m 4.50:8 2.26 25 $825.
5 Sousa .sfimésém «2%

02 £33 .5535 saw 586 :52

5 ages; .ssméaém mesa

oz .322 5925 saw «533 5.2

42.5 .2 Em .mflm .mmEm
ER 8.?

3-:

z: 062

42 8: .923 8.?
3.:

RN 062

3-: £3 .3-:

336% Nzéaoz .Gaxuzv 488m

wSEQoExmfi agooooamhm.
ESE: 23.83893
mammfifi $303893

32 {33% 35.33834
39.3 .Qmm 8.3.» $3,838.93
Eco 25838....3

9m: maotoflaa 3393893

macromNE SSBGQSQS

 

o2 .oaioom a88=8 233 25. gene... 39% 8? .5 on:
E £3833 .seaésém 8:8 is $3 .73 $3838 $3383
5 gave“? $53555 eofim mam .NEm sum .73 sausages?
350m gum atofimm

 

 

 

35.8 no a: .3 ”.39;

 

56

anaerobically with Gas Pak® (Becton Dickinson Co., Cockesysville, MD) for bifidobacteria.
Colonies were counted using a Quebec counter (Fisher Scientific, Pittsburgh, PA). The
absorbance at 650 nm versus log number of bacterial cells per ml were plotted and used
when an estimation of bacteria present in a suspension was needed.

Prior to in vivo experiments, frozen bacteria were thawed with moderate agitation
in a 37°C water bath, inoculated (1% v/v) using broths described above and incubated at
37°C until entry into log phase (8 - 20 h, depending on the species). A 1% (v/v) inoculum
was transferred to fresh broth and the culture was incubated at 37°C for an identical time
period. Total cell numbers were estimated at this point by measuring the culture’s absorbance
at 650 nm and relating these data to previous quantitative plate counts.

Some bacteria cultures were heat killed at 100°C for 50 minutes in their own culture
broth. Viable and non-viable cultures were centrifuged at 1100 x g for 15 min to recover
bacteria, washed once with a physiological saline solution and resuspended in 10% (w/v)
NFDM. Doses were aliquoted in single vials containing 0.3 ml 10% NFDM with 1.0 x 109
cells and frozen at - 80°C for no longer than 3 weeks. To determine the effects of freezing
storage in lactic acid bacteria doses prepared, viability was checked by plating a single dose

every 5 days during 15 consecutive days, using agar media and conditions described above.

2.3.2 Experimental Design
2.3.2.1 In vivo studies
Animal model. Eight week old female B6C3F1 mice (6-8 mice per experimental group)

were used for in vivo experiments. All animal handling was conducted in strict accordance

 

57

with regulations established by the National Institutes for Health. Experiments were designed
to minimize numbers of animals required to adequately test the proposed hypothesis and
approved by Michigan State University Laboratory animal Research committee. Mice were
housed 3-4 per cage in a 24 hr light/dark cycle. Water was provided ad libitum. The basic

feed was nutritionally complete (AIN-93G) (Reeves et al., 1993).

Lactic acid bacteria administration. Bacteria were orally administered as
suspensions in 10% (w/v) NFDM by gavaging doses of 1.0 x 109 cells per mouse per day.
Matching controls were gavaged 10% (w/v) NF DM without organisms. At 1, 7, 14 days, and
2 h after last dose was administrated, blood was collected and mice were sacrificed by
cervical dislocation after gentle anesthesia for extraction of Peyer’s patches, mesenteric

lymph node, and spleen.

Serum preparation. Blood was collected from anesthetized mice from the
retroorbital plexus. Serum was obtained afier overnight incubation at 4°C and centrifugation
at 1,000 x g for 15 min. Serum samples were aliquoted, stored at -80°C prior to assay for

immunoglobulins.

Semiquantitative RT-PCR for cytokines. Total RNA was extracted from Peyer’s
patches, spleen and mesenteric lymph node using the RNA STAT-6OTM isolation reagent
(TEL-TEST “B” Inc., F riendswood, TX) according to the manufacturer’s instructions. Total

RNA content was determined spectrophotometrically at 260 nm (Sambrook at al., 1989). A

 

58

reverse transcriptase (RT) reaction for first strand cDNA synthesis was carried out by the
method of Kawasaki (1991). Cytokines were detected by the semiquantitative RT-PCR
method of Svetic et al. (1991) using the PCR primers and probes for DNA hybridization
synthesized at GIBCO-BRL which sequences are described in Table 2.2. PCR was performed
in a final volume of 50 ul using a 9600 Perkin Elmer Cycler (Perkin-Elmer Corporation,
Norwalk, CT) with a 5 min incubation step at 95°C for denaturation, followed by a three step
temperature cycling (1 min denaturation at 95°C, 1 min primer annealing at 52°C, and 3 min
extension at 72°C) terminated by a 72°C incubation for 10 min then cooled to 4°C. The cycles
were repeated for an optimized number for each transcript (15, 20, 24, and 25 cycles for B2
microglobulin [BZ-MG], TNF-a, IFN-y, and IL-6, respectively). BZ-MG was used as a
“housekeeping gene” to verify initial equal quantities of RNA and the integrity of the RNA
preparation.

Hybridization analysis were carried out to determine the relative abundance of PCR-
amplified cDNAs using previously described protocols (Azcona-Olivera et al., 1995; Zhou
et al., 1997). Briefly, PCR products (10 pl) were treated with 1N NaOH and 0.2M EDTA,
boiled for 10 min and applied to a Nytron membrane (Nytran, Schleicher &Schuell Inc.,
Keene, NH) using a Bio-dot SF micro filtration apparatus (Bio-Rad Laboratories, Hercules,
CA). Hybridizations employed 32P-labeled probes prepared using a DNA 3'-End labeling
system which uses terminal deoxyribonucleotide transferase (TdT) (Promega Co., Madison,
WI) (Table 2.2). All oligoprobes had at least a specific activity of l x 109 cpm/ug and all
hybridizations were performed with no less than 1 x 107cpm/ml. Blots were exposed to

Kodak X-OMAT-AR film (Eastman Kodak, Rochester, NY) with an intensifying screen at

59

.0358 22 fies: £85585 389.5 850 a? 2.8ch AB 3 $8: a e @3383 3
.23: .3 e 08% 3 55563. 2m 2m BBQ 25 2:5 .5 3 B Bursa 22.» §§§ Boa as $an

 

 

sham AU<<U<HGO<<U<0<HO<OOO
omvm .O<HOUHOU<UUUO<HOHUU m2 m aUUUOOH<HUEO<UH<UHUH sod—Z...
SN .UOHUHUOOHUHiHa‘Oa‘Og
omm s<8<<<<80hu<<8<8 on FUOU<<OH<<U<O<GHUHUO .56:
E m aO<HOO<UUHU<HOH<UEU
3 ~ a<0<COUUHO<<U<UFHU< om AGGREUEU DOEO<UUH<UUFH v-1:
w fl m .O<H<H<8<UUO<OHOHO<U
N: 23885850858 8 bpfiuoefiotuoooogh oz 3
.m .m .m .m
dogma .m 3on :oEmom .m BEE 3835» can 85w 25.8.6

 

.8389 <ZQo commas—m 88% 8 confining? 83.8% no.“ 805368 39a 98 mega—bag <75”. 2.22.6 we 82258 5.55 dd mam»;

60

-80°C for variable length periods depending upon signal strength. For the purposes of
comparison, exposure periods for each cytokine were identical among untreated and treated
groups. Autoradiographic bands were quantified with an Epson ES-lOOOC scanner and
Sigma-Scan program (Jandel Scientific, San Rafael, CA). Relative amounts of cytokines
mRN A were estimated by dividing the densitometric area of the cytokine autoradiography

band by the densitometric area of the housekeeping gene BZ-MG autoradiography band.

2.3.2.2 Ex vivo studies

Animal model and Lactic acid bacteria administration. Eight weeks old female
B6C3F1 mice (6-8 mice per experimental group) were also used for ex vivo experiments.
Bacteria were orally administered to mice by gavaging a single dose of 1.0 x 109 cells per
mouse in 10% (w/v) NFDM. Eight hours after administration of the bacteria, fecal pellets
and blood were collected, and mice were killed by cervical dislocation after gentle anesthesia

for peritoneal cells, Peyer's patches, and spleen extraction.

Serum and fecal pellet preparation. Blood was collected from anesthetized mice
from the retroorbital plexus. Serum was obtained afier overnight incubation at 4°C and
centrifugation at 1,000 x g for 15 min. Serum samples were aliquoted, stored at -80°C prior
to monitor for immunoglobulins.

Fecal samples were prepared as described by de Vos and Dick (1991). Briefly, feces
were collected, aseptically weighted and placed into centrifuge tubes. Ten ml of 0.01 M

phosphate buffered saline (PBS) per gram of feces (v/w) were added, and the mixture was

61

incubated for 15 min at room temperature. Samples were mixed by vortexing until
suspended, left to settle for 15 min, mixed again, and centrifuged at 22,000 x g for 10 min
in a Sorval RC 5C centrifuge (Du Pont Co., Wilmington, DE). The supernatant was removed

and stored at -80°C for immunoglobulin measurement.

Leukocyte preparation. Peritoneal cells were prepared by injecting 2-3 ml RPMI
1640 medium containing 10% (v/v) heat inactivated fetal bovine serum (F BS) supplemented
with 100 U/ml penicillin, 100 pig/ml streptomycin, 5 x 10'5 M 2-mercaptoethanol, 25 mM
Hepes buffer, 1 mM sodium pyruvate, and 0.1 mM nonessential amino acids (10% FBS-
RPMI 1640) into the peritoneal cavity of euthanized mice with a 10 ml syringe fitted with
a 20 gauge needle. After gentle massage of the abdomen and without withdrawing the needle,
the remaining 7-8 ml of medium were injected. Abdomen was massaged again and the liquid
was slowly drawn back inside the syringe. Afier all the fluid was evacuated with the syringe,
a small hole was cut in the peritoneum and the remaining fluid was collected with a sterile
Pasteur pipette. The lavage fluid was centrifuged 800 x g for 5-10 min and the supernatant
discarded. Ten ml of same medium was added to resuspend the cells. Cells were counted

with a Neubauer hemacytometer to determine number of cells per ml.

Lymphocyte preparation. Peyer’s patches were removed, teased apart, and passed
through 85-mesh stainless-steel screen. Cells were suspended in 5 ml 10% FBS-RPMI 1640
medium, washed by centrifugation at 450 x g for 10 min and resuspended in 2 ml of the

same medium and counted. Spleen was removed aseptically, teased apart with tissue forceps

62

in 10 ml 10% FBS-RPMI 1640, and centrifuged at 450 x g for 10 min. Erythrocytes were
lysed for 5 min at room temperature in 5 ml of a buffer containing 9 parts of 0.16 M
ammonium chloride plus 1 part 0.17 M TRIS buffer (pH 7.2). Ten ml of fresh medium were

added and cells were centrifuged at 450 x g for 10 min, and counted.

Leukocyte cultures. Spleen, Peyer’s patch lymphocytes (5 x 105 cells/ml) and
peritoneal cells (1.5 x 105 cells/ml) were cultured in 24-well tissue culture plates with 10%
FBS-RPMI 1640 medium in the absence (unstimulated) and in the presence of inducing
agents as lipopolysaccharide (LPS, 1 ug/ml) from Salmonella typhimurium, phorbol 12-
myristate-13 acetate (PMA, 10 ng/ml) and ionomycin (I, 0.5 ug/ml) and incubated at 37°C
in 7% CO2 atmosphere. The tissue culture supematants were collected after 2 and 5 days and

monitored for cytokine levels (IL-12, IL-6, IFN-y, TNF-or) and IgA production by ELISA.

Cytokine quantification. Cytokines were quantitated in supematants by enzyme-
linked immunosorbent assay (ELISA). hnmunolon 4 Removawell microtiter stn'ps (Dynatech
Laboratories Inc., Chantilly, VA) were coated overnight at 4°C with 50 ul/well of a purified
rat anti-mouse cytokine capture antibody (PharMingen, San Diego, CA) in 0.1M sodium
bicarbonate buffer, pH 8.2 (1 11ng [IL-6, TNF-a], 2 ug/ml [IFN-y] or 0.5 mg/ml [IL-12]).
Plates were washed three times with 0.01M phosphate buffered (pH 7.2) saline (PBS)
containing 0.2% Tween 20 (PBS-T). Plates were blocked with 300 pl of 3% (w/v) bovine
serum albumin (BSA) in PBS (BSA-PBS) at 37°C for 30 min and washed 3 times with the

PBS-T. Standard murine cytokines (PharMingen) or samples were diluted in 10% (v/v)

63

FBS-RPMI 1640, and 50 ul aliquots were added to appropriate wells. Plates were incubated
at 37°C for 60 min, washed 4 times with PBS-T, and 50 ul of biotinylated rat anti-mouse
cytokine detection monoclonal antibody (1.5 ug/well; PharMingen) diluted in BSA-PBS
were added to each well. Plates were incubated at room temperature for 60 min, washed six
times with PBS-T and 1 additional time with distilled water. Fifty ul of streptavidin-
horseradish peroxidase conjugate (1.5 ug/well; Sigma Chemical Co., St. Louis, MO) diluted
in BSA-PBS were added to each well and plates were incubated at room temperature for 60
min. Plates were then washed 8 times with PBS-T and two more times with distilled water,
and 100 pl of substrate [10 mM citric phosphate buffer (pH 5.5), containing 0.4 mM
tetramethylbenzidine (TMB; Fluka chemical Corp., Ronkonkoma, NY) and 1.2 mM H202]
were added to each well. The reaction was stopped by adding an equal volume (100 pl) of
6N H2804. Absorbance was read at 450 nm on a Vmax kinetic Microplate reader (Molecular
Devices Co.) and cytokine concentrations were quantitated by using Vmax Software. The
sensitivity of this assay was 0.5 ng/ml for IL-6, 1.2 ng/ml for IFN-y, 3 ng/ml for TNF-a and

75 pg/ml for IL-12.

Immunoglobulin quantification. IgA and IgG were quantitated in supematants and sera
by enzyme-linked immunosorbent assay (ELISA). Irnmunolon 4 Removawell microtiter
strips (Dynatech Laboratories Inc., Chantilly, VA) were coated overnight at 4°C with 50
ill/well of a heavy-chain specific goat anti-mouse IgA or IgG (Capel Worthington, Malvem,
PA) (10 ug/ml) in 0.1M carbonate buffer, pH 9.6. Plates were washed three times with

0.01M phosphate buffered (pH 7.2) saline (PBS) containing 0.2% (v/v) Tween 20 (PBS-T).

 

64

To reduce nonspecific protein binding, plates were blocked with 300 pl of 1% (w/v) bovine
serum albumin (BSA) in PBS (BSA-PBS) at 37°C for 30 min and washed 3 times with the
PBS-T. Standard mouse reference serum (ICN Immunobiologicals, Costa Mesa, CA) or
samples were diluted in BSA-PBS (sera) or in 10% (v/v) FBS RPMI-1640 (supematants),
and 50 pl were added to appropriate wells. Plates were incubated at 37°C for 60 min, washed
5 more times, and then 50 pl of goat anti-mouse immunoglobulin horseradish peroxidase
(y- or a- chain specific, Cappel Worthington, Malvem, PA) diluted 1:1000 for IgA and 1:500
for IgG in BSA-PBS was added to each well. Plates were incubated at 37°C for 30 min and
washed 6 times with PBS-T. Bound peroxidase was determined with 2,2'-azino-bis (3-
ethylbenz-thiazoline—6 sulfonic acid) (ABTS) substrate [0.4 mM ABTS, 50 mM citrate
buffer (pH 4.0), and 1.2 mM hydrogen peroxide] as described previously by Pestka et al.,
1980. Absorbance was measured at 405 nm on a Vmax kinetic Microplate reader (Molecular
Devices' SOFTmax, Molecular Devices Co., Menlo Park, CA) and immunoglobulins were
quantified using the Vmax software. The sensitivity of this assay was 0.1 pg of IgA or IgG

per ml.

NO production. The method of Dietert et al., (1995) was used to detect nitrite
production by macrophages. Cells were cultured as described above for 2 and 5 days and
nitric oxide (NO) production was assessed by measuring nitrite accumulation, a stable
metabolic product of NO, in the culture supematants. Nitrite concentrations were determined
by the Griess reaction. Equal amounts of N-(l-naphthyl) ethylenediamine dihydrochloride

(NED; Sigma) [100 mg dissolved in 100 ml of distilled water] and sufanilamide (Sigma) [1

 

 

65

g dissolved in 100 ml of a 5% phosphoric acid solution] solutions were mixed prior to each
assay (Griess reagent or chromogenic reagent). Nitrite standards [2 mM stock solution; from
0 to 200 pM] were diluted in the same media in which the cells were suspended. Equal
amounts of Griess reagent and NaNO2 standards or samples (100 p1) were placed in a 96 well
plate in duplicate and incubated for 5 min at room temperature to allow the chromophore to
develop and stabilize. Absorbance was read at 550 nm using the Vmax kinetic Microplate
reader (Molecular Devices Co.) and nitrite concentrations were quantitated by using Vmax

Sofiware. The sensitivity of the assay was 4 pM.

2.3.3 Statistical methods

Data summary was done with descriptive statistics such as the mean and standard
error of the mean. Statistical comparisons of treatment and control groups were analyzed
using a Student’s t test for comparison between two groups or by Dunnett’s test following
one way analysis of variance (ANOVA) using the SigmaStat Statistical Analysis System
(Jandel Scientific, San Rafael, CA). A p value of less than 0.05 was considered statistically

significant.

 

66

2.4 RESULTS

2.4.1 Viability of lactic acid bacteria.

An initial study was performed to determine the ability of representative lactic acid
I bacteria to recover, remain viable and in high numbers after being frozen at -80°C over 15
days. MRS agar was used to enumerate these cells after being frozen for 0, 2, 5, 10, and 15
days. Table 2.3 summarizes the survival of several lactic acid bacteria suspended in NF DM
at this temperature. None of the tested lactobacilli exhibited a significant decrease in cell
concentration, which verified that primarily viable organisms would be used in in viva
studies. Since all species and strains tested showed a similar behavior, I assumed a similar

trend for the rest of the species used in my studies.

2.4.2 Study I.- In viva effects of lactic acid bacteria on cytokine mRNA expression.

All animals appeared healthy during these experiments. For all in vivo studies, weight
gain in mice was monitored to determine if any change occurred due to feeding lactic acid
bacteria during 1 and 2 weeks (Figure 2.1). No significant effects on weight gain were
observed in mice fed lactic acid bacteria as compared to NFDM controls.

To assess potential immune modulation of bacterial feeding, basal cytokine mRN A
expression in spleen, Peyer’s patches and mesenteric lymph node was evaluated after
administration of lactic acid bacteria for 1, 7 and 14 days at 1.0 x 109 cells per dose per day.
Figure 2.2 represents a typical result of exposure to viable and non viable lactic acid bacteria

(L. bulgaricus 1489 [NCK 231] in this case) for basal expression of IFN-y and TNF-or

 

 

67

@8583 no: u DZA~

._8 con 3:5 menace.“ .828 .«o wo— ofi 3 votes“:
mm :58 3583 2C. .Uonm 8 a 8» com 823:2: gm.» mg 5 BEE 803 .3889.

 

 

 

no ad ad ma 3 33m ooh... 383
$2 a a a a 4mm 8.? 38.3
a 3 3w 2 3 am 85.. 33838 q
a a a a a 2mm 85. simenei
m2 2: 2: 2: mo 73 33838 q
2: ”.2 : : 2 W3 $336284
3 no no no 3 was: ”8.8%qu
no 2: m2 2: we was §§m§q
S 2 3 M; a 3-: ”SNSMNEQ
«92358 we. mEmEmmeoobaz
2 S m N o awn

 

 

£083 28 58:0 33% 8 6°03 mfiooc mo 385 .2 mqmfi

68

.mEoEtomxo EoBbE 025 we o>flm€82g8 28
Q n E 2mm H :38 08 8mm .QmEomoEéoo a3 x C 99% 3 93 N. no.“ 3.883 Eon 38“— vom 028 mo Saw Emma? in charm

69

 

 

I: 7 days

m 14 days

 

 

 

W

 

 

 

l
f

I§\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\V

 

 

+R\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\V

 

 

 

‘9
III!

I
V.

I
N.

I I I I I
q °°. ‘0. V. N.
—- o o o o

(8) 0193 wfilam

CD.
0

S. therm ophilus
Sanofi St133

L. casei L. bulgaricus 1489 L. acidophilus
Sanofi L32

ATCC 39539

CONTROL

NCK 231

non-fat dry milk

70

$808598 3838 95 mo 0355858 08 98 G u 5 SEE um :38 08 8mm .33 N.
2a fl 5 2 mm 56 a: grumse ”$233993 3% Ba 2,: 3 osmaxo :20 5% 8E 5 232 <zma 8296 .3 2:3

71

2%; 8.3.3»: .238

 

O

fl

M

IIIIIIITIIIIITIIITI
N

 

 

 

v

2%; Sign 62.8

g E E6

 

IIIIIIIIIITIrIIIIII

 

 

 

v

MQQZ ENE» Q DENKZMVJN:

2%; 35.3»: 39.8

”32> 33-30: .880

 

 

(\l v—n

M

IIIIIIIIIIIIIIIIITT

 

 

22> 8.3.3»: 328 2§> 33-39: .880

 

 

EL EL o

H

N

8% N. g
82: D

M

IIIIIIIIIIIITIIIIII

 

 

”maps 3N magmxmm: ZNMHEM.

 

 

29m omawousuap slump};

cam omawousuap OAQBIOH

RYE/E.

is:

72

mRNA. In spleen, basal expression of IFN-y and TNF-a mRN A was higher than in Peyer’s
patches and mesenteric lymph node for all groups. No difference was found between
treatment and controls of the species tested (Lactobacillus casei ATCC 39539 and L.
bulgaricus 1489 [NCK 231]) regardless of whether viable or non-viable bacteria were used.

When L. bulgaricus 1489 (NCK 231) and Streptococcus thermophilus St-133 were
administered for 1 and 2 weeks (1 x 109 cells/mouse/day), significant induction or inhibition
was not observed for the basal mRNA for any cytokine tested in either spleen or Peyer’s
patches (mesenteric lymph node was not tested) (Figure 2.3). Expression of IL-6, IFN-y and
TNF-a mRNA was similar for mice treated during 1 and 2 weeks with L. bulgaricus 1489
(NCK 231), S. thermophilus St-l 33 or NFDM used as a control. These results typify those
obtained from 3 separate experiments not only for L. bulgaricus 1489 (NCK 231), S.
thermophilus St-133, but also for L. bulgaricus 1489 (N CK 231), S. thermophilus St-133,

L. casei ATCC 39539 and L. acidophilus La-2 (data not shown).

2.4.3 Study II.- Ex vivo effects of lactic acid bacteria on cytokine, nitric oxide and
immunoglobulin production.

The effects of exposure to different strains of L. acidophilus, L. bulgaricus, L. casei,
L. helveticus, L. gasseri, L. reuteri and S. thermophilus in mice on cytokine production were
assessed in mitogen stimulated and unstimulated leukocyte cultures. Mice were fed a single
dose of a microbe suspension and sacrificed after 8 hours. Peyer’s patch and spleen cells
were isolated, peritoneal cells were extracted, and cultured with or without mitogens (LPS

or PMA + ionomycin [PMA+I]) for 2 and 5 days. LPS induces activation of B cells, antigen

 

 

73

Figure 2.3. Cytokine mRN A levels in mice after oral exposure to Lactobacillus
bulgaricus 1489 (N CK 231) and Streptococcus thermophilus St- 133 for 7 and 14 days.
Data are mean i S.E.M (n = 6) and are representative of three separate experiments.

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2 0 SPLEEN PEYER'S PATCHES
co ' 2.0
E 3 |:] 7 doses :
.E 1'6 —: '//////, 14 doses 1'6 .3
‘8’. 12 I I E
- '2 1.2 -
IL-6 ‘3 g E
g 0-8 '5 0.8 f
s : 1
o - I
‘3‘ 0.0 7 oo ‘
Control Lbl489 St133 ' Control Lbl489 St133
8 4 4
53 j 3
E 3 "_ 3 —:
o -I
IFN 9; 3 :
o q --d
1: : -
E 1 ‘2 1 "3
M 0 O 7 ,
Control Lbl489 St133 Control Lb1489 St133
4 4
TNF-a 2

 

A

0)
lllllllllllLllllLL
u—I DJ
IllllllllllllllLlll

Relative densitometric Area
N

 

 

 

 

 

 

 

1%?

k\\\

 

 

 

0
Control Lb1489 St133 Control Lbl489 St133

75

presenting cells and macrophages (Beutler and Kruys, 1995). PMA is an activator of protein
kinase C while I is a calcium ionophore which increases the levels of intracellular calcium
(Truneh et al., 198 5). Together they are strong enhancers of T cell activation.

Lactic acid bacteria did not affect cytokine production by Peyer’s patch and spleen
cell supematants (TNF-oc, IL-6, IL-12, and IFN-y, or NO). Furthermore, TNF-a was
undetectable in spleen, Peyer’s patch or peritoneal cultures. However, IL-6 was produced in
considerable amounts by peritoneal cells from mice fed NFDM, L. bulgaricus 1489 (NCK
231), S. thermophilus St-l33, L. casei ATCC 39539 and L. acidophilus La-2, when
stimulated with LPS or with PMA + I, with LPS treated cells showing the highest IL-6 levels
(Figure 2.4). Effects on cytokine production by peritoneal cells were strain dependent. IL-6
was also produced in appreciable amounts by peritoneal cells from mice fed L. helveticus Lr-
92, L. gasseri ADH (N CK 101), L. reuteri ATCC 23272, and Bifidobacterium Bf-l when
cells were stimulated with LPS, but synthesis of this cytokine seemed to be inhibited when
these same cells were stimulated with PMA + I (Figure 2.4).

Peritoneal cells stimulated with PMA + l but not LPS showed a high capability for
IFN-y production (Figure 2.5). Those cells coming from mice fed just one dose of L.
bulgaricus 1489 (NCK 231), S. thermophilus St-133, L. casei ATCC 39539 and L.
acidophilus La-2 produced a equivalent or greater IFN-y level when compared to the control
group fed NFDM. In contrast, those cells coming from mice treated with L. helveticus Lr-92,
L. gasseri ADH (NCK 101), L. reuteri ATCC 23272, and Bifidobacterium Bf-l, exhibited

markedly depressed IFN-y production as it was observed for IL-6.

76

.God w S 055 83m wow anew 8980 05 8 “8&8 53>
8055wa 38$:me 8:865 AL 358598 8383 03:0 o>uficomoa2 8w new 3 n 5 85:8 832330 Ema H
988 8a Sam .Extmtgohfim ”mm .1833 4 U5 image 4 "un— .wzotmfig Q ”:1— .§~.£Q.§oo 4 Had .828 4 Mod
.§~EQQE§S w. Ham .wzotaMBQ Q Haw— .<m5m 3 can 8w 3.3me use made m 28 N Baa 380:8 803 ngfiomsm
82:30 .cstwamd £53522 + 5%: E roe/E ”ER: _ .mmd macaw“ wfiogfi mo ooaomoa on“ E 98 00:83
05 E wee—:0 was £552: 5 “Contact mm «£825 Eon 0:03 mo cozmbmmEEcm Bad 2 x 35390 803 £00 Region
$583 Boa 282 we omoc o_wEm a Rm 028 Bob 85:3 GEE—8 me x m. 3 :8 Region 5 £32 can «an 9.55

77

 

[:1 No 1nduccr
- LPS
“ PMA +1

M.
k

 

 

 

E

 

 

 

 

 

m
>~ W 9. m

'3 a

'5 1:

“3 "3

52 P.

= = m

° c

5 .fl

‘30 co
jilllIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII1II
In o m o m o o tn o In C
N N .— u— N .— u—n

(IN/3“) 9-11 (IN/‘3“) 9-11

Lg Lr Bf

Lh

Lb St

Control

78

.8888 8c M Q2 Amod w 3 :2: 83m 8.: 30% 75:8 2: 8 80%8
5:5 8288b? EmoEcw_m 88865 0.: 988898 088:8 93 :0 8388852 0.8 :5 G .I. 5 85:8 88:3:
.8 E.m.m H :88 08 3mm $38888me Hum @838 4 US 58.8% 4 Had £8.83»: 4 Mn: .§~EQ038 Q Had
8.89 Q How— .§~.E8E8£ M. Ham “azutcmwza 4N "A: .<m:m E 36: 8: 8:88 :5 998 m :8 N 8% 880:8 083
3582.895 88:6 .958 W: 89888: + 3%: 3 {SE 28%: fimmd $88 wfioswfimo 8888 2.: 5 :5 888m
05 5 85:8 98 80:88 E 8988: mm «883 E8 383 .«o 888388 8% n w 8598 08>» :8 Houston
drowns 28 28B .«o 88 Bwfim a 8: 8:: 88.: 85:8 288:8 no: x m. S :8 3888: 5 £881.; .m.~ Quanta

79

 

 

 

 

 

 

 

 

 

:- . ‘9...
g : g Pg“)
-: c:
S m E Z
2 8 a. H
7 “ ' s.
U" 2 '84
n
2
fl
4 @530
52¢ a
«2 .
E 'fifi
z
I——V/////////////////////////// « W '3
G
2 1
2
W‘ W
a: g .3
E
l—W l—W H
a g (I)
Z
g a
m m 8
‘ .D
8‘ i? i:
'U “U
6: '9
8 if. '3
5 W 5 W1:
O O G
a: u: o
co co 0
TTIIIIII‘IIIIIIIIIIIII1I llTlllTlll1lTTllllIlT—TTI
‘13 V M N F‘ W V M N F‘ O

(Ila/‘3“) k-NJI (luv/‘3") A-Nal

80

Peritoneal cells produced more IL-12p40 when stimulated with LPS than when
stimulated with PMA + I (Figure 2.6). No consistent effects on IL—12p40 were observed in
cultures from treatment mice. IL-12p35 was undetectable in all cases.

NO production was relatively low after incubation of peritoneal cells for 2 (1. After
5 days, levels of NO increased for those cells stimulated with PMA + I (Figure 2.7),
following a pattern very similar to the one presented for IFN-y production, with L. helveticus
Lr-92, L. gasseri ADH (NCK 101), L. reuteri ATCC 23272, and Bifidobacterium Bf—l
inhibiting production of this mediator.

Regarding immunoglobulin production, sera from mice fed lactic acid bacteria
revealed no difference in total IgA (Figure 2.8) and IgG (Figure 2.9) levels when compared
to control mice fed with NFDM. Similar results were obtained when analyzing for
coproantibodies (Figure 2.10). Supematants from Peyer’s patch and spleen cell cultures had

undetectable levels of either IgA or IgG.

81

803% so: u :z .38 w 3 85
82m 8:: anew 38:8 2: 8 888: 5:5 8088mm: 88:8me 888:8 AL 8888:88888 03C: 03888888:
88 :8 G n 5 88:8 88:8: mo 2mm H :88 88 San 83888::me "mm .888: {N ".5 888% {N nua—
48838 4 ”:1— .§~E:§88 4 Max— 88: Q Mud £32885me w Lm “8.8.8838 Q ":1— .888 888Eomsm 2:
(«o :8 8.: 8888: 8: 83 2:de {mfim 3 98de :8 8:88 :8 88: m :8 N 88 888:8 88>» 8:88.888
88:50 .2881 md 88883 + 8%: S {SE ”8&1 :nmmd 88w: M8268 :o 8:88: 85 8 :8 8:88 2: 8
88:8 :8 8:88 8 8:88: 8 «888 :8: 282 .«o 88.88888 8:: n m 888:0 P83 m:8 8:88: 8888
:88 382 mo 80: 288 m 8: 8:8 80: 88:8 988:8 me: x WC :8 88988 8 £88: 382$: 6d Baum—

82

 

E] No inducer

- LPS
“PMA”

LL1*££&£i

 

 

* Llflifiwfiplm

 

 

250

 

:1 a.

a a

'6 '5

S“ "9

2 r.

= =

O O

.E I

00 Q

IIIIIIIWIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIII
O O O O O O O O O O
O V) O W In O V) O In
N '_' '—‘ N N v—n —
(Iva/3d) ovdzml (flu/3d) ovdzml

Lg Lr Bf

Lh

Lb St Le

Control

 

83

.:0:00:0: :0: .I. QZ .God m 3 0:8 8:8 :0: :88 6:800 05 0: :00:8: 5:.»
80:0:0bm: 082:8? 88088 T: 8:088:08 088:8 03: :0 03880880: 08 :8 G n :v 88:80 008:::: :0 Sam:
8 808 08 San 8320:0303bm ”.5 ..:.:0:8: 4 U5 488.3% 4 nu..— .302030: Q HS .§~.E:§80: .4 ”:4 .880
4 H04 .§~.EQ.QE:0S M. n:m 8:02:83: Q "a: .088: :0: :088: :8 88: m :8 N :08 :0:00:00 0:03 88880::8
0:380 .A_8\w1m.o 80:88:: + 8%: o: £2: 28%: immuc 8:08 883:8 :0 00:08:: 0:: 8 :8 00:88 0:: 8
:08800 :8 80508 8 :0£:08: 8 8:88: :6: 0:02 :0 88.8888 :08 n m :2850 0:03 m:00 80:88: 8:008:
28 0:02 :0 80: 08:8 8 :0: 00:8 80:”: 8:880 288:8 no: 0: mi :00 80:88: 8 208— 0:80 0:82 fin 0.5”:—

84

 

 

 

 

 

 

:m a: 35:00
mm mm - o
,. m N
m. v
m. 0
w 0
08: misc:— w m
8:0: 2 88%;
m N
w v
w 0
:32: Z m. w
m: I H
88:8 02 D 08: $0.50.. a n

 

(wfi) EON

(wfi) EON

85

.mEoEtoaxo 1838 no azuficomoume 0.8 98 G n 5 Edam H 508 2a 8mm ATE
§§8€§m u E .NRMN 8: .532 q u .5 A _ 2 52. $3183 q u 3 .83 §§§§ q n 5 $3
33%38 q u 3 «ma 8: $8 4 u S .mmim ENEQEQS §88§Em n .m .: mm 52. $3 §Em§
azzwugofiuu u ad 3.583 Eon 0:03 03m? 3.55m mo €28 o3 x S 80c _ .m 28 Qmubmogmzoo o3 x S 3 mm
MUZV $3 gmxamge waEuBSuSmo women h 23 g .< wfiuoom “can 028 88m 380:8 88 5 £32 <3 .wd «Sur—

 

86

 

 

1.7
A
aw h
m
V

 

 

 

 

6666666
33333

came «-2 .23

 

V////A 8 hours

 

 

came ”-2 x <m_

87

.God w 3 x28 88? wow anew 75:8
85 9 8898. 2:3 moocfiobmn 88$:me 880me AL 888898 882% .8 83888888 88 28 G u 5 Eng a S88
2“ 53 A _ .8 §§§o§m "E fig 8: $38 4 n 5 A 8 2 8% E2 55% q n 3 8.5 3.23%
q n .5 $3 3.3838 4 u S .933 8,? ES 4 n S .3; $38285 “388938 u .m ,: MQQE
$3 SBEUMBQ “32838an M £16 3883 Bus 38E 0383 888% mo 8:8 o3 x S omow # .m 28 383888—88
oo _ x : mm Em 338585 388288 u .m 23 A H mm 888 $3 828% q u 5 imam 8? .88 q u 3 .N
.8; §~E§E§ 83:83883 u «A .8 888 3 new 5 .< magma 8% 8:: Bob 88280 88 E 228— 03 .afi «Suw—

 

88

 

 

 

{:1 7 doses

A

W 14 dosa

 

 

V/////%
.II

 

 

 

 

 

 

%///A

%

Z

 

 

 

 

 

 

 

 

 

 

0.
2

da—
5
1.
A

______q —
0. 5.
1 0

0.
0

.5382 x as

La LC Lb St

Control

 

 

 

T

 

§
T§

\
\\

A

T§
T§

 

 

m 8 hours

 

 

 

T

 

 

 

 

 

 

T

 

 

 

 

 

 

 

 

 

 

 

 

 

:33 we x o 3

Lb St Le La Lh Lg Lr Bf

Control

 

89

.God m 5 858 82m “8,“ 3on _ob:oo 85 8 88%»: 53> mooceotmc
282.3%? 88865 AL £88898 888m (8 8388888 Pa 28 G u 5 Edam H 588 88 Sam .SmEomogéoo

as x a 23m $38285 $88888. u 5 2a A _ mm 88 a: 28:83 q u 3 «ma 8,? 38.8 q n 3
N64 §~£Qoh8¢ 3583883 U «a .8 888 E 28 n 958$ 8% 838 88w 888:8 88m 5 £98— <w~ .cfin PSME

9O

 

 

 

W//////////A

 

_

 

 

 

 

.T%/////////////////%

I

 

 

 

 

 

%////////%

 

 

 

 

 

Z

 

 

 

 

7/////%

 

 

 

 

 

 

 

20.86420.

€5ch <3

St

Lb

ntrol

91

2.5 DISCUSSION

Although several studies have reported the in vitro effects of administration of
lactic acid bacteria on cytokine production, very little information is available about cytokine
production in vivo. This study was an attempt to determine potential in vivo effects of oral
administration of lactic acid bacteria on cytokine gene expression in mucosal and systemic
sites. The main findings of study I were that: (1) repeated administration of lactic acid
bacteria did not affect the growth rate of mice and, (2) basal mRNA expression of IL-6, IFN-
7 and TNF-oc was not affected by repeated lactic acid bacteria administration. From study 11,
it can be concluded that: (l) a single exposure to certain lactic acid bacteria altered
subsequent mitogen induced cytokine and nitric oxide production by peritoneal cells, (2)
immunoglobulin levels were not affected either at the mucosal or systemic sites.

Freezing between -20°C to -80°C produces variable results depending on bacterial
species. The use of NFDM to resuspend bacterial cells has been reported to be successful in
long-term preservation for small volumes due to a cryoprotective effect (Silliker et al., 1980).
Typically some bacteria can be kept for 6 months to 2 years by this method. Here, I grew
cells under optimum conditions and to a sufficient number of cells to provide a good cell
suspension. Thus, ordinary freezing proved to be a good means for preserving the viability
of my bacterial cultures for studies.

I chose to measure basal cytokine mRNA in spleen, Peyer’s patches and
mesenteric lymph node because they represent the systemic and mucosal immune

compartments. The gut immune response is induce in the Peyer’s patches where activation,

92

switching, proliferation and differentiation of B cells is under the control of T cells, cytokines
and other accessory cells (Ogra et al., 1994). As an initial approach I decided to analyze basal
cytokine mRNA instead of serum cytokine levels for several reasons: 1) the short half life
of some cytokines makes their quantitation a weak indicator of cytokine production in vivo;
2) uptake of cytokines by some receptors makes it difficult to accurately measure these
proteins. I chose a very sensitive technique, RT—PCR which includes amplification of mRNA
present on a tissue sample for a desired cytokine (Svetic et al., 1991).

The in vivo results may differ from previous in vitro studies because lactic acid
bacteria probably do not survive digestion. For those bacteria to target the intestine, they
have to tolerate a number of physical and chemical conditions present in the gastrointestinal
tract, i.e. gastric acid and small intestinal secretions. Additionally, lactic acid bacteria may
need to establish and become active in the gut prior to having immune effects. The influences
of various factors of host and of microbial origin, that participate in the formation of an
individual gastrointestinal microbial ecosystem, are considerable. This creates great
methodological limitations for in vivo research (Sanders, 1993). The intestinal microflora,
one of the most complex ecological niches, are still not fully understood. Many types of
facultative anaerobic bacteria live together, influencing each other. It is very difficult to
predict these interactions and obviously there is great difficulty in finding individuals with
similar or identical GI ecosystems. Thus, potential beneficial effects of lactic acid bacteria
often prove to be statistically insignificant when attempts are made to confirm the
conclusions made by different authors.

The result observed in my ex vivo study (study 11) with peritoneal cells suggests that

93

macrophages are a key target for lactic acid bacteria immunomodulation activity. Oral lactic
acid bacteria administration to mice altered peritoneal cell activity in vitro. The effect of
various strains relative to their capacity to secrete IL-6, IFN-y, IL-12 and NO ranged from
marginal stimulation or no effect to inhibition. The potential to alter macrophage function
is very important for several reasons. Macrophages constitute the second major cell
population of the immune system with phagocytosis as their primary function. They originate
in bone marrow, and after migration and maturation they settle in tissues as mature
macrophages (J aneway and Travers, 1997). They can be activated by a variety of stimuli and
their principal function includes phagocytosis of foreign particles, production of cytokines
(IL-6, TNF-a, IL-l , IL-12) that recruit other inflammatory cells and act as antigen-presenting
cells. Thus, macrophages can participate in humoral immune responses and/or as effector
cells of cell-mediated immunity (as antigen presenting cells).

This study examined the effect on peritoneal cells of different mitogenic signals.
These include lipopolysaccharide (LPS) from Salmonella typhimurium, which is a B-cell
and macrophage activator (Beutler and Kruys, 1995), and a combination of phorbol-12-
myristate-l3-acetate (PMA), which activates protein kinase C (PKC), and ionomycin (I),
which increases the concentration of intracellular calcium [Ca 2+] (Truneh et al., 1985). The
latter two compounds can affect both macrophages and T cells. The peritoneal cavity is
frequently used as a concentrated source of mature murine macrophages. Peritoneal lavage
yields 2 - 8 x 106 cells of which 20 - 40% are macrophages, but there is also contamination
by other cell types (Herzenberg et al., 1997). B cells represent 10 - 4O % (Kim et al., 1996),

and thus, T cells and other type of cells as natural killer (NK cells) may represent at least

94

20%. Thus other cells besides macrophages may contribute to the findings observed in
peritoneal cell cultures.

It should be noted that vehicle plus dietary factors may influence basal levels in
control groups in this study. Casein is a dietary component present in NF DM and in murine
diet AIN-93G which could be able to produce a mucosal response in fed-NFDM groups.
Peptides derived from casein were shown to contain an immune activating factor (Y amauchi
and Suetsuna, 1993). Diets containing between 20-40% of casein increase cellular immunity
in spleen cells and macrophages (Ueda et al., 1990).The casein fraction is composed of
different proteins (a, B, and K). Notably, K casein triggers lymphocyte proliferation (Werfel
et al., 1996). Also, B-casein promotes cellular and humoral responses (Cavallo et al., 1996).
During enzymatic digestion of human and bovine caseins, peptides are formed and released,
which are able to stimulate phagocytic activity of murine and human macrophages and these
apparently exert in viva a protective effect against infections (Migliore-Samour et al., 1989).

Other authors have similarly reported that macrophage activation occurs after lactic
acid bacteria administration. Saito et al. (1987) found that intramuscular injection to mice
with heat-killed L. casei LC 99018 increased macrophage function. Perdigon et al. (1986a)
observed an enhanced macrophage and lymphocytic activity by administering a mixed
culture of Lactobacillus acidophilus ATCC 4356 and L. casei CRL 431 to mice. This same
group later reported a similar conclusion in a feeding study with L. casei and L. acidophilus
to mice infected with a lethal dose of Salmonella Iyphimurium (Perdigon et al., 1990). L.
casei and L. bulgaricus were also studied and found that both given either orally or

intraperitoneally were able to activate peritoneal macrophages in mice from day 2 onward

95

(Perdigon et al., 1986b) and they suggested that these bacteria, when passing through the
intestinal tract, may be responsible for an enhanced host immune response. L. bulgaricus
was generally less effective than L. acidophilus increasing the activity of peritoneal
macrophages measured by phagocytic assay. Similar results were obtained when using
Streptococcus thermophilus and L. acidophilus (Perdigon et al., 1987) or fermented milks
with L. casei, L. acidophilus or/and a mixture of both (Perdigon et al., 1988, 1991b, 1992).

Lactic cultures have been shown to stimulate cytokine production in earlier studies.
Halpem et al. (1991) reported a rise in serum IFN-y production in young human adults when
they consumed two cups of yogurt daily. Kishi et al. (1996) reported that the oral
administration of viable L. brevis ssp. coagulans produced a significant increase in IFN-a
levels in human blood in a dose dependent manner, but heat treated bacteria did not elicit
a similar response. Kitazawa et al. (1992, 1994) demonstrated induction of IFNs in murine
peritoneal macrophages when stimulated in vitro with L. acidophilus, and an increase of the
expression of mRNA encoding IFN-a in spleen-macrophages and Peyer's patch-adherent
cells stimulated with L. gasseri DSM 20243. L. bulgaricus and S. thermophilus added to
human peripheral blood lymphocytes have an adj uvant action, potentiating IFN-y production
(De Simone et al., 1986).

In my experiments, cytokine production by lactic acid bacteria (or food component)
apparently involves not only macrophages which are capable of producing lL-6, IL-12 and
NO, but different types of cells, as indicated by the production of IFN-y which is not
produced by macrophages. Further studies are necessary to elucidate the mechanisms

involved in these processes.

96

IL-6 is a B-cell growth factor that operates in a paracrine manner enhancing
immunoglobulin production from previously activated B-cells. IL-6 secretion usually
requires a previous stimulation with other cytokines. IFN-y is known to interfere with the
induction of IL-6 mRNA accumulation and protein secretion by macrophages in humans
(Cicco et al., 1990).

IFNs are proteins with a variety of physiological effects. They are produced mainly
by T lymphocytes, macrophages and natural killer (N K) cells. IFN-0t and B are closely related
and produced by T cells and macrophages. NK cells are large lymphoid cells with many
intracellular granules, which are able to recognize and kill abnormal cells even though they
do not bear antigen-specific receptors. IFN-y is produced exclusively by T lymphocytes and
NK cells whereas macrophages do not express the IFN-y gene (Kirchner, 1986).

IL-12 is a heterodimer and its principal biologic effect is upregulation of the Thl
response via induction of IFN-y production, T-cell proliferation and NK cell-mediated
cytotoxicity (Hendrzak and Brunda, 1995).

NO is produced by immune cells in response to cytokines such as TNF-a, IFN-y and
IL-2 (Lin et al., 1995) by stimulating the expression of inducible NO synthase and acting in
combination with LPS in murine macrophages. IFN-y activates macrophages and these
activated macrophages are able to produce TNF-a, IL-6, IL-12, oxygen radicals and nitric
oxide (NO) (Janeway and Travers, 1997). The evidence in the literature suggests that NO
participates in inflammatory and autoimmune mediated tissue destruction (Yamashita et al.,

1997)

In this study, elevation of IL—6 secretion was observed in LPS-stimulated peritoneal

97

cell cultures after a single oral exposure to lactic acid bacteria. Since B cells and
macrophages are responsive to this mitogen but not T cells, it is likely that they are the cells
of choice associated with the lactic acid bacteria immune effect (Figure 2.11). Also, murine
peritoneal cells released IL-6 into their culture supematants after being stimulated with PMA
+ I combined. The use of PMA + I together stimulates the phosphatidylinositol signal
transduction pathway by increasing [Ca 2l] and by modulating PKC in murine cells, which
is also the functional role of IFN-y. Since IFN-y is able to increase [Ca 2+] inside murine
macrophages stimulating their cytotoxic activity, I believe that stimulation of murine
peritoneal macrophages with PMA + I could exert the same effect. Several studies support
the fact that the combination of PMA + I is capable of stimulating macrophages by
mimicking IFN-y activity (Klein et al., 1990; Gumina et al., 1991). PMA by itself does not
activate macrophages to be cytotoxic what suggests that PKC activity may be required but
not sufficient for macrophage activation (Gumina et al., 1991).

The results in study 11 are not surprising if, as I described above, PMA + I together
are capable of activating macrophages. After stimulation with PMA + I, macrophages could
be activated to produce IL-6 and IL-12 (Figure 2.11). IL-12 may be able to stimulate IFN-y
production by T cells or NK cells present in the peritoneal cells (Puddu et al., 1997). NO is
produced from L-arginine via catalysis by an enzyme that is activated by [Ca 2+] (Dietert et
al., 1995). Thus PKC and [Ca 2+] by PMA + I could increase activity of nitric oxide synthase.
In my study, nitrite accumulation was enhanced when IFN-y was present for more than 48
h following the stimulation with PMA + 1. Thus IFN-y production could be responsible for

the induction of NO in murine peritoneal macrOphages. In contrast, other authors (Eason and

98

.mtogomn Eon 038— .«o Aomsocfizoo an: x 3 80¢ 28 we.“ 8:: Bow c8855 mega—so :8 Region co SEE:
no a 50382 + as? o: <33 .28? : sasssqb u. see mad 2&0? .8 some... 2958.2 .:.~ 2:5

sz
2.8 xz e mamacaocoms.

 

 

 

 

11 l .
oxm
/.Al BENCH— H+NMUH
mzmo v.2
3:8 x2 93 p .m
EmEEflcoo + mommcaohoasv
e w._._m_o
maggots—2 -_<m20.:~_mn_
Lt); mm.—

02
@Al\mM «my ....V... .
. ..x
1 tr
., H .HWHH nilll.....rH...H.. ..
.\ @Alnl m=mo m

 

100

Martin, 1995) found this effect when the mitogen used for stimulation was LPS, but not
PMA + I alone or in combination, and when using a macrophage cell line and not primary
cell culture.

Neither peritoneal cells, spleen nor Peyer’s patch lymphocytes responded to LPS or
PMA + I by producing TNF-oc, typical cytokine produced by activated macrophages. Studies
of the kinetic properties on the transcription of mRN A and secretion of IL-6, IL-1 and TNF-a
showed that TNF-a mRN A transcription and protein production occur very rapidly (Martin
and Dorf, 1990). TNF-a mRNA peaks 1-2 h after the stimulus was produced and the
cytokine can be collected in supematants after 2-4 h culture. IL-6 and IL-1 mRNA peak after
4-8 h and simultaneously but the protein is present at the highest concentration between 8-12
h for IL-6 and 12 h for IL-1. That means that I could have missed TNF-a detection.

The question that remains to be answered is what is the importance of the peritoneum
in relation to the immune system. Bellingan et al. (1996) recently showed that macrophages
migrate from the peritoneum to the draining lymph nodes. The migration process takes more
than 4 h, but once started these macrophages are cleared within 96 h from the peritoneum.
Thus, macrophages homing the peritoneal cavity may mediate a strong activation as a
consequence of lactic acid bacteria and NFDM administration, perhaps prior to migration
within gut associated lymphoid tissues. It is known that in rodents and other animal species
some antigen presenting cells (dendritic cells) are able to migrate via afferent lymph into
draining lymph nodes or from intestinal spaces, via lymph, to lymph nodes and via blood,
to spleen (Herzenberg et al., 1997). Cells from the peritoneum also seem to be able to

migrate to Peyer’s patches (Heel et al., 1997; Kawabata et al., 1995), which might suggest

101

a possible role of peritoneal cells activating mucosal immune tissues. Kroese et al. (1989)
observed that a high percentage of the IgA-secreting cells located in the lamina propria had
their origin in B cells from the peritoneal cavity. There are the so called B-l cells possessing
a CD5 marker, while the conventional cells from Peyer’s patches, spleen, and lymph nodes
are called B-2 cells. The former have a longer life span and appear to be at a greater
activation state, being able to populate the lamina propria inside the gut.

This research presented herein suggests that lactic acid bacteria can alter peritoneal
cell function and this may subtly impact immunity. On a speculative level, these effects could
be produced by strains possessing specific components in the cell wall or soluble substances
which affect immune cells via receptors present mainly in macrophage cell surface, and
perhaps in some other cells such as NK cells. Further mechanistic understanding is warranted

on how these and other immunologic effects are mediated by lactic acid bacteria.

CHAPTER 3

EFFECTS OF LACTIC ACID BACTERIA AND THEIR EXTRACTS ON
CYTOKIN E PRODUCTION IN VIT R0

102

103

3.1 ABSTRACT

Cells from a large number of bacterial genera as well as bacterial products have been
shown to possess mitogenic and polyclonal activating properties when cultured with cells of
the immune system. Notably, gram positive cell walls contain teichoic acid and
peptidoglycan which have been shown to induce cytokine production by immune cells and
suppress tumor development. In this study, I tested the potentiating effects of representative
lactic acid bacteria and their extracts on leukocyte function. Specifically, the effects of in
vitro exposure to heat-killed cells of Bifidobacterium, Lactobacillus acidophilus,
Lactobacillus bulgaricus, Lactobacillus casei, Lactobacillus gasseri, Lactobacillus
helveticus, Lactobacillus reuteri, and Streptococcus thermophilus, their cell walls, and their
cytoplasmic extracts on proliferation, cytokine and intermediate production were examined
in the RAW 264.7 macrophage cell line. A similar strategy was applied to spleen and Peyer’s
patch lymphocytes as well as in peritoneal cells from mice. The results suggest that lactic
acid bacteria as well as their cytoplasmic and cell wall fractions are able to stimulate cloned
macrophages to produce very significant amounts of TNF-a, IL-6 and nitric oxide. While
similar effects were not noted in spleen and Peyer’s patch cell cultures from mice, a
pronounced enhancement in IL-6 production by peritoneal cells was observed when cultured
with those extracts. The results suggested that as a group, the lactic acid bacteria are capable

of stimulating macrophages and/or other immune cells to produce cytokines and nitric oxide.

104

3.2 INTRODUCTION

Cells of a large number of bacterial genera as well as bacterial products have been
shown to possess mitogenic and polyclonal activating properties. As mitogens they induce
DNA synthesis, blast formation and ultimately division of lymphocytes. One of the most
widely studied bacterial products with mitogenic activity is lipopolysaccharide (LPS), which
is present in the cell wall of gram negative bacteria. The basic molecule of LPS consists of
a lipid component, the lipid A, covalently bound to a heteropolysaccharide of two distinct
regions: the core oligosaccharide and the O-specific chain. It can also contain polysaccharide
chains of variable lengths. LPS can activate macrophages and induce the release of many
mediators and cytokines such as TNF-a, IL-1, IL-6 (Gammage, et al. 1996).
Lipopolysaccharide (LPS) has also been shown to exert a mitogenic effect in B cells in mice.
Since the structure of LPS is complex and variable even within the same species, the degree
of lymphocyte activation can vary. Some bacterial species that represent major human
pathogens are gram positive and possess a basic gram positive cell wall structure
(streptococci, micrococci, staphylococci, etc.). Gram positive cell walls contain teichoic acid
and peptidoglycan. These gram positive components have been shown to induce the TNF-a
and IL-6 production of human monocytes (Heumann et al., 1994).

Baricault et al. (1995) reported that the use of fermented milks manufactured with
lactic acid bacteria may exert an attenuating effect on cancer cell growth and differentiation.
Anti-tumor activity observed with Lactobacillus bulgaricus is associated with the bacterial

cell wall (Bogdanov et al., 1975), whereas other studies indicate a possible role for the whole

 

It." ‘ I?

1'9".

 

105

bacteria and/or cell extracts for Bifidobacterium infantis (Kohwi et al., 1982). Either viable
or intact dead bacteria seem to be able to induce these effects. One explanation may be that
tumor suppression observed is mediated through an immune response.

A number of investigations have focused on the capacity of gram positive lactic acid
bacteria to alter immune fiinction. Perdigon et al (1986a) observed enhanced macrophage and
lymphocyte activity in mice after administering a mixed culture of Lactobacillus acidophilus
and L. casei. This group also reported activation of peritoneal macrophages in mice after oral
administration of L. casei and L. bulgaricus (Perdigon et al., 1986b). Similar result were
found for Streptococcus thermophilus, L. acidophilus orally delivered (Perdigon et al., 1987)
and heat-killed L. casei administered by injection into mice (Saito et al., 1987). Previous
studies conducted in this laboratory found that heat-killed lactic acid bacteria stimulate
production of IL-6 and TNF-a in RAW 264.7 macrophages (Marin et al., 1997a, 1997b,
1997c). Although many of the studies regarding lactic acid bacteria are controversial, they
seem to intersect in a common result, activation of macrophages after lactic acid bacteria
treatment.

In this study, we assessed the potentiating effects of representative lactic acid bacteria
and their extracts on leukocyte function. Specifically we evaluated different commercial U.S.
dairy strains, either as heat inactivated whole cells or as fractions (cytoplasm and cell wall)
for their ability to alter immune functions by measuring effects on secretion of cytokines.
Initially I determined the effects of in vitro exposure to Bifidobacterium, Lactobacillus
acidophilus, Lactobacillus bulgaricus, Lactobacillus casei, Lactobacillus gasseri,

Lactobacillus helveticus, Lactobacillus reuteri, and Streptococcus thermophilus and their

 

 

106

extracts on proliferation, cytokine and intermediate production by a macrophage cell line. I
then examined the same effects on spleen and Peyer’s patch lymphocytes as well as by
peritoneal cells from mice. The results suggested that lactic acid bacteria as well as their
cytoplasmic and cell wall fractions are able to stimulate macrophages to produce large
amounts of TNF-a, IL-6 and nitric oxide (NO). Similarly, an enormous increase in
production of IL-6 by peritoneal cells is observed. However these effects were not observed

in spleen and Peyer’s patch cell cultures.

107

3.3 MATERIAL AND METHODS

3.3.1 Lactic acid bacteria fractionation

Representative cultures of Bifidobacterium, L. acidophilus, L. bulgaricus, L. casei,
L. gasseri, L. helveticus, L. reuteri, S. thermophilus were obtained from three different
sources: American Type Culture Collection (Rockville, MD), Dr. T.R. Klaenhammer (North
Carolina State University) and Sanofi Bio-Industries (Waukesha, WI) (Table 3.1). From our
frozen (-80°C) bacterial stocks described in chapter 2, bacteria were thawed with moderate
agitation in a 37°C water bath until all the ice melted. Lactobacilli and streptococci were
grown in MRS (Difco) broth (De Man et al., 1960) and bifidobacteria in MRS + 5% (w/w)
lactose (MRSL) broth at 37°C for 15 h. A small part of each culture (1% v/v) was transferred
to fresh broth and then incubated at 37°C until exponential phase was reached. The culture
was plated after being serially diluted to assess the number of bacteria per ml. Bacteria were
recovered by centrifugation at 1,100 x g for 15 min.

Figure 3.1 shows an schematic representation of the cell fractionation procedure
followed. For heat-killed bacterial fi'actions, bacterial cultures in early stationary phase were
quickly chilled to 4°C in an ice-ethanol bath. Bacteria were harvested by centrifugation at
16,000 x g for 10 min at 4°C and then washed with sterile distilled water and centrifuged at
16,000 x g three times. Bacterial yield and number were determined by weight and by
plating respectively. After resuspending in Hank’s buffer saline solution (HBSS, Sigma
Chemical Co., St. Louis, M0) to achieve a concentration of 25 mg/ml, the bacteria were

aliquoted in 1 ml vials and heated to a temperature of 100°C for 50 min, then quickly chilled

108

 

 

:5 damage? .moEmsuEéE «cam 97% «3.23.323» “.3883th

Q2 .oLLEoom £28260 83:0 33. 585:2. thmm 00.5. .2833 3583823
:5 £58.33 $3385-05 wocmm No-5 $3833 $583033

02 amend image: 2% 2:88 552 z: 062 :3 team £8883

DE .ozfioom £26260 05:20 093. 585:9. ammom 00H< Show SNNBBQBS
oz emcee image: 3% 8:05 £82 mm 52 33 senses 303383

E £5333 $3335-05 wocmm Nam:— mSEQSEQQ ESERQBS

=5 £5813 3 $3333-05 098m Em Eatfiogohmbm

coca—om Edam mtoUmm

 

beam 5 BS 8330 ._ .m mnmfi.

109

Figure 3.1. Schematic representation of lactic acid bacteria fractionation.

110

Frozen cells

Activation

Cell Suspension

Transfer

Bacterial cells

Centrifuge 16,000 x g 10 min 4°C
Supernatant
Wash and centrifuge 3 times Discard

Chilled ice-ethanol 4°C

 

 

<—-—

—
~
0
H

Pe

Pellet —) HEAT-KILLED FRACTION

l Resuspend in ddeO

Sonicate, 5°C, 30 min

Broken cell suspension

Heat 60°C 15 min
Centrifuge 800 x g 10 min
Pellet (discard)

 

Supernatant

Centrifuge 40,000 x g, 30 min at 4°C

 

 

\
Superratant
CYTOPLASMIC
FRACTION
(____ Protease 20pg/mg pellet, RNAse A 250 ug/ml

DNAse] 250 ug/ml, at 37°C overnight

Pellet (crude wall fraction)

Centrifuge 800 x g
30 min at 4°C

 

Protein, RNA
or DNA
free
cytoplasmic
fractions

 

Pellet (discard) Supernatant

30 min at 4°C

 

Centrifuge 40,000 x g

 

Pellet (pure walls) Supernatant (discard)

Centrifuge 40,000 x g 30 min at 4°C
Heat 90°C, 15 min aliquot and freeze - 800C

CELL WALL FRACTION

111

and stored at -80°C.

For cell wall and cytoplasmic fraction preparations, bacterial cells were harvested and
washed as described above and then subjected to a series of fractionation steps (Work, 1971;
Takahashi et al., 1993; Heumann et al., 1994). After resuspending in sterile distilled water,
bacteria were disrupted by sonication for 30 min at 5°C in a Branson sonifier W-350
(Branson/ Sonic power Co., Danbury, Conn). The suspension was heated at 60°C for 15 min
to inactivate autolytic enzymes. The suspension was centrifuged at 800 x g for 30 min at 5°C
and the pellet (unbroken cells) was removed. Cell walls were sedimented from the
supernatant by centrifuging at 40,000 x g for 30 min. The supernatant was then aliquoted in
1 ml vials and frozen at - 80°C [cytoplasmic fraction]. The pellet was further treated with
protease (20 pg/mg of crude cell wall), ribonuclease and DNAse (250 pg/ml) in a 0.1 M Tris-
HCl buffer (pH 7.4) at 37°C overnight to eliminate the contaminating cytoplasmic material.
The cell wall fraction was centrifuged at 800 x g for 30 min at 4°C and then at 40,000 x g for
30 min, weighed, heated to 90°C for 15 min, aliquoted at a concentration of 25 mg/ml and
then frozen at - 80°C [cell wall fraction]. Aliquots of cytoplasmic fraction or control buffer
were treated with protease (20 pg/mg), ribonuclease or DNAse (250 pg/ml) at 37°C
overnight to eliminate proteins, RNA or DNA respectively. These were filter sterilized and

kept at -80°C until used for assay.

3.3.2 Sugar determination

Sugar content on bacterial fractions was determined using the method described by

Dubois et al. (1956). Briefly, a stock solution containing 100 pg glucose/ml was prepared.

112

From this stock solution, serial dilutions were made and placed in glass tubes ranging from
0 to 90 pg sugar/ml (2 ml of each). Two m1 (2 ml) fractions were prepared for heat-killed,
cytoplasmic (1/ 10 dilution) and cell wall fiactions (1/ 100) and placed in glass tubes. Fifty pl
of a 80% phenol solution were added to each tube. Concentrated H2804 (5 ml) was quickly
dispensed, mixed and allowed to stand for 10 min. Tubes were shaken and placed in a water
bath at 25-3 0°C for 10-20 min. Absorbance at 490 nm was measured and carbohydrate levels

were quantified from a standard curve. The sensitivity of the assay was 10 pg/ml.

3.3.3 Protein determination

Protein determination on bacterial fractions was made in accordance with the method
of Bradford (1976) using commercially prepared dye reagent (Bio-Rad Laboratories,
Hercules, CA) and bovine serum albumin (BSA) standard (Sigma). Briefly, a stock solution
containing 1.4 mg BSA/ml was prepared. From this stock solution, serial dilutions were
made and placed in glass tubes ranging from 0 to 40 pg protein/ml. Dilutions were prepared
for heat-killed (1/ 100), cytoplasmic (1/500) and cell wall fractions (1/5) and placed in glass
tubes and 800 pl of standard or sample were mixed with 0.2 ml of concentrated dye. Tubes
were mixed thoroughly. Absorbance at 595 nm was measured and protein concentrations

determined from a standard curve. The sensitivity of this assay was 5 pg/ml.

3.3.4 DNA determination
DNA was determined by a modification of the method of Ausubel et al. (1994). Heat-

killed bacterial fractions (0.5 ml) were centrifuged at 200 x g for 10 min at 4°C. The

113

supematants [vial S] were transferred to a 1.5 ml vial. The pellets [vial B] were resuspended
in 0.5 ml of a buffer prepared with 0.01 M Tris-Cl with lmM EDTA and 0.2% Triton X-100
pH 7.4 (TIE-0.2% Triton). After vortexing vigorously, the lysed cells were centrifuged at
1,100 x g for 10 min at 4°C. These new supematants [tube T] were transferred to a 1.5 ml
vial. TIE-0.2% Triton (0.5 ml) was added to the vial B still containing the pellets. One half
(0.5) ml of 25 % (w/v) trichloroacetic acid (TCA) was added to each vial S, T and B and they
were vortexed vigorously. Vials were held at 4°C for 18 h to allow the DNA to precipitate.

DNA standards (salmon sperm DNA, 10 mg/ml stock solution), lipopolysaccharide
(LPS) fiom Salmonella typhimurium and cell wall fractions (0.5 ml) were centrifuged at
1,100 x g for 10 min at 4°C. The supematants [tube T] were transferred to a 1.5 ml vial. TTE-
0.2% Triton (0.5 ml) was added to the vial B containing the pellets. One half (0.5) ml of 25
% (w/v) trichloroacetic acid (TCA) was added to vials T and B and they were vortexed
vigorously. Vials were held at 4°C for 18 h to allow to allow the DNA to precipitate.

Cytoplasmic fractions (0.5 ml) were treated with 0.5 ml of 25% (v/v) TCA, vortexed
vigorously, and held at 4°C for 18 h to allow to allow the DNA to precipitate.

After DNA precipitation for all the bacterial fractions and standards, vials were
centrifuged at 1,100 x g for 10 min at 4°C and supematants were aspirated and discarded.
Eighty (80) pl of TCA solution (5% v/v) were added to each vial and heated for 15 min at
90°C to hydrolyze DNA. Freshly prepared diphenylarnine (DPA; 150 mg DPA powder, 10
ml glacial acetic acid, 50 pl acetaldehyde solution [16 mg/ml]) reagent (160 pl) was added

to each vial and also to a blank vial containing 80 pl of TCA 5% (v/v). Vials were vortexed

and incubated for 4 h at 37°C for color to develop. One hundred (100) pl of the resulting

114

colored solution were transferred to a 96-wells plate and absorbance was read at 570 nm
using a Vmax kinetic Microplate reader (Molecular Devices’ SOFT max, Molecular Devices
Co., Menlo Park, CA). DNA concentrations were determined from a standard curve. Total
DNA for each bacterial fraction was calculated by addition of the DNA amounts quantified

independently for vials S, T, and B. The sensitivity of the assay was 15 pg/ml.

3.3.5 RAW 264.7 Macrophage culture

A murine macrophage cell line obtained from the American Type Culture Collection
(Rockville, MD) was used (RAW 264.7 ATCC TIB 71). RAW cells were maintained in
Dubelcco’s modified Eagle’s medium (DMEM) (Sigma Chemical Co., St. Louis, MO)
supplemented with 10% (v/v) fetal bovine serum (F BS) (Gibco Laboratories, Chagrin Falls,
IL), sodium bicarbonate, non-essential amino acids (1 mM) (Gibco), lmM sodium
pyruvate (Gibco), 10 ml NCTC-l35 medium (Gibco), 100 U/ml penicillin and 100 pg/ml
streptomycin (Sigma). Cell number and viability were assessed by trypan blue (Sigma) dye
exclusion using a Neubauer Bright Line counting chamber (American Optical Co., Buffalo,
NY). RAW cells were cultured at a final density of 5 x 105 cells/ml in 48 well flat bottomed
tissue culture plates with and without bacterial fractions. Salmonella typhimurium

lipopolysaccharide (LPS [Sigma] 1 pg/ml) was used as a positive control for stimulation of
RAW cells. Triplicate cultures of this cell line were exposed to 50, 250, 500 pg (wet weight)
of heat-killed bacteria, cytoplasm or cell wall fraction/ml and incubated for 48 h at 37°C in
7% C02. RAW cells were also stimulated with 500 pg (wet weight) of the protein, RNA and

DNA free cytoplasmic fractions, as well as similarly prepared controls. After suitable time

115

intervals supematants were harvested, stored at -80°C for cytokine assay, and cells were used
for MTT assay to determine proliferation/differentiation. Results fiom the protein, RNA and
DNA free fractions were compared to the untreated cytoplasmic fractions from each specific
microorganism.

Contamination with exogenous endotoxins was ruled out by using a test with
polymyxin B. Various concentrations of polymyxin B (Boehringer Mannheim, Germany) (5,
25, 50 pg/ml) were used as the LPS inhibitor to confirm that fractions were not contaminated
with lipopolysaccharide. Bacterial fractions were mixed with the various polymyxin B
concentrations and after a 30 min incubation at room temperature the mixtures were added
to the cultures. After a 48 h incubation, supematants were harvested and stored at -80°C for

cytokine assay, and cells were used for MTT assay to determine proliferatiomdifierentiation.

3.3.6 Leukocyte preparation

Primary cell cultures from B6C3F 1 female mice, 8 wk old (Charles River Labs,
Raleigh, NC) were used. All animal handling was conducted in strict accordance with
regulations established by the National Institutes for Health. Experiments were designed to
minimize numbers of animals required to adequately test the proposed hypothesis and
approved by Michigan State University Laboratory Animal Research committee. Mice were
sacrificed by cervical dislocation under gentle anesthesia. Peritoneal cells were prepared by
injecting first 2-3 ml RPMI 1640 medium containing 10% (v/v) heat inactivated fetal bovine
serum (FBS) supplemented with 100 U/ml penicillin, 100 pg/ml streptomycin, 5 x 10'5M 2-

mercaptoethanol, 25 mM Hepes buffer, 1 mM sodium pyruvate, and 0.1mM nonessential

116

amino acids (10% F BS-RPMI 1640) into the peritoneal cavity of mice with a 10 m1 syringe
fitted with a 20 gauge needle. After gentle massage of the abdomen and without withdrawing
the needle, the remaining 8-7 ml of medium were injected. Abdomen was massaged again
and the liquid was slowly drawn back inside the syringe. After all the fluid was evacuated
with the syringe, a small hole was cut in the peritoneum and the remaining fluid was
collected with a sterile Pasteur pipette. The lavage fluid was centrifuged at 800 x g for 5-10
min and the supernatant discarded. Ten ml of same medium were added to resuspend the
cells. Cells were counted with a Neubauer hemacytometer to determine number of cells per
ml and cultured to a final density of 2 x 105 cells/ml.

Peyer’s patches were aseptically removed, teased apart, and passed through 85-mesh
stainless-steel screen. Cells were suspended in 5 ml 10% F BS-RPMI 1640 medium, washed
by centrifugation at 450 x g for 10 min and resuspended in 2 ml of the same medium and
counted. Cells were cultured to a final density of 5 x 105 cells/ml.

Spleen was removed aseptically, teased apart with tissue forceps in 10 ml 10% FBS-
RPMI 1640, and centrifuged at 450 x g for 10 min. Erythrocytes were lysed for 5 min at
room temperature in 5 ml of a buffer containing 9 parts of 0.16 M ammonium chloride plus
1 part 0.17 M TRIS buffer (pH 7.2). Ten ml of fresh 10% FBS-RPMI 1640 were added and
cells were centrifuged at 450 x g for 10 min, counted and cultured to a final density of 5 x
105 cells/ml.

Leukocytes were cultured in a 48-well flat bottomed tissue culture plate in triplicate
with 500 pg (wet weight) of heat-killed, cytoplasmic and cell wall fraction of

Bifidobacterium, L. acidophilus, L. bulgaricus, L. casei, L. gasseri, L. helveticus, L. reuteri,

117

S. thermophilus or Salmonella typhimurium LPS 1 pg/ml. Cells from spleen, Peyer’s patches
or peritoneal cavity were incubated at 37°C in 7% CO2 and supematants were harvested at

time intervals and stored at -80°C for cytokine assay.

3.3.7 MTT Assay

MTT assay was performed as described by Mosmann (1983). The reagents were
prepared fresh weekly and consisted of 5 mg of MTT (3-[4, 5-dimethy1thiazol-2-yl]-2, 5-
diphenyltetrazolium bromide)(Sigma)/ml in 0.01 M phosphate buffer saline (PBS) at pH 7.4.
This solution was sterilized by passage through a 0.2 pm filter and store in the dark at 4°C.
MTT reagent (150 pl /well) was added to each well after the supernatant was removed, and
plates were incubated for 75 min in a C02 incubator at 37°C. Plates were centrifuged at 450
x g for 10 min, MTT reagent was removed from wells using a syringe with a thin needle
trying not to disturb the pellet and the crystals formed. Solubilization solution (100%
dirnethylsulfoxide [DMSO] from Sigma) (450 pl /well) was added to each well and plates
were gently agitated to uniformly resuspend the solubilized dye. The solubilized dye (200 pl)
was transferred fiom each well into a 96-well plate and read on a Vmax kinetic microplate
reader (Molecular Devices, Menlo Park, CA) at 570 nm. Unstimulated cells values were

considered as control cells.

3.3.8 Cytokine quantification
Cytokines were quantitated in supematants by enzyme-linked immunosorbent assay

(ELISA). Irnmunolon 4 Removawell microtiter strips (Dynatech Laboratories Inc., Chantilly,

118

VA) were coated overnight at 4°C with 50 pl/well of a purified rat anti-mouse cytokine
capture antibody (PharMingen, San Diego, CA) in 0.1M sodium bicarbonate buffer, pH 8.2
(1 pg/ml [IL-6, TNF-a], 2 pg/ml [IFN-y] or 0.5 mg/ml [IL-12]). Plates were washed three
times with 0.01M phosphate buffered (pH 7.2) saline (PBS) containing 0.2% Tween 20
(PBS-T). Plates were blocked with 300 p1 of 3% (w/v) bovine serum albumin (BSA) in PBS
(BSA-PBS) at 37°C for 30 min and washed 3 times with the PBS-T. Standard murine
cytokines (PharMingen) or samples were diluted in 10% (v/v) F BS-RPMI 1640, and 50 pl
aliquots were added to appropriate wells. Plates were incubated at 37°C for 60 min, washed
4 times with PBS-T, and 50 pl of biotinylated rat anti-mouse cytokine detection monoclonal
antibody (1.5 pg/well; PharMingen) diluted in BSA-PBS were added to each well. Plates
were incubated at room temperature for 60 min, washed six times with PBS-T and 1
additional time with distilled water. Fifty pl of streptavidin-horseradish peroxidase
conjugate (1.5 pg/well; Sigma Chemical Co., St. Louis, MO) diluted in BSA-PBS were
added to each well and plates were incubated at room temperature for 60 min. Plates were
then washed 8 times with PBS-T and two more times with distilled water, and 100 pl of
substrate [10 mM citric phosphate buffer (pH 5.5), containing 0.4 mM tetramethylbenzidine
(TMB; Fluka chemical Corp., Ronkonkoma, NY) and 1.2 mM H202] were added to each
well. The reaction was stopped by adding an equal volume (100 pl) of of 6N H2804.
Absorbance was read at 450 nm on a Vmax kinetic Microplate reader (Molecular Devices
Co.) and cytokine concentrations were quantitated by using Vmax Software. The sensitivity

of this assay was 0.25 ng/ml for IL-6, 0.4 ng/ml for TNF-or and 1.25 ng/ml for IFN-y.

119

3.3.9 Nitric oxide determination

The method of Dietert et al. (1995) was used to detect nitrite production by
macrophages. Cells were cultured as described above for 2 and 5 days and nitric oxide (N 0)
production was assessed by measuring nitrite accumulation, a stable metabolic product of
NO, in the culture supematants. Nitrite concentrations were determined by the Griess
reaction. Equal amounts of N-(l-naphthyl) ethylenediamine dihydrochloride (NED; Sigma)
[100 mg dissolved in 100 ml of distilled water] and sufanilamide (Sigma) [1 g dissolved in
100 ml of a 5% phosphoric acid solution] solutions were mixed prior to each assay (Griess
reagent or chromogenic reagent). Nitrite standards [2 mM stock solution; from 0 to 200 M]
were diluted in the same media in which the cells were suspended. Equal amounts of Griess
reagent and NaNO2 standards or samples (100 pl) were placed in a 96 well plate in duplicate
and incubated for 5 min at room temperature to allow the chromophore to develop and
stabilize. Absorbance was read at 550 nm using the Vmax kinetic Microplate reader
(Molecular Devices Co.) and nitrite concentrations were quantitated by using Vmax

Software. The sensitivity of the assay was 12 pM.

3.3.10 Statistical methods

One way analysis of variance AN OVA with Dunnett’s test for multiple comparison
versus a control group was applied to data, using Sigma-Stat Analysis System (Jandel
Scientific, San Rafael, CA). A Ps0.05 was considered statistically significant. Results were
expressed as relative change in the cytokine production compared to control by dividing

experimental data by control values (fold control).

120

3.4 RESULTS

3.4.1 Bacteria fractionation process

Lactic acid bacteria yields from MRS broth cultures were between 1 and 1.5 g/ 100
ml. Wall fractions from bacterial cells were prepared from mechanically broken cells by
removal of extraneous material by washing, differential centrifugation and enzymatic
digestion. The supernatant collected after centrifugation of the sonicated whole cell fraction
represented the cytoplasmic content of the bacteria, and the pellet resulted in the crude cell
wall. In our studies the crude cell wall obtained from the whole cell preparation represented
15-30% (wet weight) of our initial bacteria, and this is in agreement with the yield reported
in general by the literature (Work 1971). Purified pure cell wall fraction ranged between 36-
80 % (wet weight) with respect of the crude cell wall fraction, depending on the bacterial
species. By weighing each fraction, total colony forming units for each bacteria strain could

be correlated with their wet weight. Those values varied from strain to strain (Table 3.2).

3.4.2 Cell fraction constituents

Heat-killed, cytoplasmic and cell wall fractions were analyzed for protein, total
carbohydrate and DNA content (Table 3 .3). The highest protein content was found in general
in the cytoplasmic fraction. Carbohydrate concentration made up a significant part of the
cytoplasmic and cell wall fractions. Heat-killed bacterial fractions possessed also a fairly
high carbohydrate content but lower than the corresponding cell wall for that same bacteria.

DNA was also detected in both heat-killed bacteria and cytoplasmic fractions.

121

85:52. see u oz

 

 

 

m2 x 3 N2 x 2 N2 x 3. 3% 338.5% M.
m2 6. 2 m2 x 3 .2 x 2 ER 8.? teased
n2 x i ..2 x Z .2 x S 8-3 288%: 4
m2 0. 2 .2 0:; m2 x1 :2 202 team 4
.2 x 3 m2 x .2 n2 6. .2 28m 8: see 4
n2 3. e.» 2 x 3 .2 x 3 28 652 a: 3:833
m2 x 3 Q2 m2 x Ne 23 32838 4
n2 x 5 m2 x I .2 x A: Em sateegenesm
as. :8 8585 85-232
m1\m=oo

 

.EwmoB 83 28368.0 REE Com 35:33am :8 Etopoam .N. m mam/E.

TABLE 3.3. Chemical composition of bacterial fractions“.

 

 

 

 

 

 

 

 

 

Protein pig/mgb

Whole cells Cytoplasmic Cell wall
Bifidobacterium Bf-l 39 :1: 9 470 :1: 12 169 :h 2
L. acidophilus La 2 12 :1: 4 35 i 0 1 i O
L. bulgaricus 1489 NCK 231 25 :t 4 171 :1: 12 4 :t O
L. casei ATCC 39539 37 d: 3 212 :1: 4 1i 0
L. gasseri NCK 101 14 i 2 108 i 25 O i 0
L. helveticus Lr 92 15 i 3 294 :1: 10 1 i 0
L. reuteri ATCC 23272 5 i 2 200 i ll 0 :1: 0
S. thermophilus St-133 13 i 3 31 i 4 O :1: O

Carbohxdrate pg/mgb

Whole cells g'toplasmic Cell wall
Bifidobacterium Bf-l 93 :1: 5 180 :1: 22 246 :1: 3
L. acidophilus La 2 28 i l 28 i l 37 :1: 6
L. bulgaricus 1489 NCK 231 40 :1: 5 81 :1: 10 95 :t O
L. casei ATCC 39539 32 :t 0 32 :1: 0 106 i 12
L. gasseri NCK 101 46 i 0 116 :t O 69 :1:1
L. helveticus Lr 92 32 i O 128 d: O 56 :1: 3
L. reuteriATCC 23272 41 :t 2 124 :t 7 39 :1: l
S. thermophilus St-l33 28 :1: 2 10 i 1 47 :1: 2

DNA pg/mgb

Whole cells Cytoplasmic Cell wall
Btfidobacterium Bf-l 11 i 1 0 :1: 0 0 :1: 0
L. acidophilus La 2 24 d: 0 5 :t 1 1 i 0
L. bulgaricus 1489 NCK 231 5 i 0 60 d: 22 4 :t 0
L. casei ATCC 39539 22 i 1 23 :1: 4 7 :1: l
L. gasseri NCK 101 7 i 1 34 :1: 3 1 i 0
L. helveticus Lr 92 21 i 1 57 :h 10 1 :t 0
L. reuteriATCC 23272 5 i 1 21 i 7 1:1: 0
S. thermophilus St-133 3 i 1 5 :t 1 l :b 0

 

“ Values are means i S.E.M (n = 4).
b Relative to wet weight

123

Recoverable DNA in these two fractions varied with the species, but the cytoplasmic fraction
had, in general, a higher concentration of DNA than the whole cells. Surprisingly, cell wall
fractions also showed presence of DNA. Since the reagent used for this DNA assay,
diphenylarnine (DPA), reacts mainly with the sugar portion of the DNA, these values
indicating DNA content in cell walls could be reflecting interaction between DPA and any
of the sugars present in these wall fractions.

Treatment of whole cell and cytoplasmic fractions with protease was very effective
in reducing protein concentrations. Protein concentration (Table 3.4 and 3.5), but not
carbohydrate content (Table 3.6 and 3.7), declined drastically after protease treatment as
compared to the untreated fractions. This provided whole cell and cytoplasmic fractions that
were potentially free only of the protein content that the initial fraction possessed. This type
of treatment was repeated using RNAse or DNAse to obtain a derived new fraction free of
nucleic acids. The protein content was slightly decreased for RNAse treated fractions,
suggesting protein-nucleic acid association. Treatment with RNAse did not affect
carbohydrate or DNA content as compared to the untreated fraction (data not shown).
Treatment of whole unviable bacterial cells with DNAse removed not only DNA from the
original fractions but also some protein. Again, this indicated some DNA association with
the membrane or proteins which were dragged out together with DNA afier DNAse
treatment. Treatment of the cytoplasmic fraction with DNAse did not affect so much protein

or carbohydrate content but removed practically all DNA present in the original fractions.

.3 u 5 2356 we: a 3352 2a
25329 m1 on 5 30,222 3 ca 3 6:8 3563 me 333282 w: 8 523 38: 83 883 322.23: 8285:: _

an a 2% 2268 i Ema _ 85.28.2388:

124

 

S a ”2 2 a m: N a 8 am a 2m 27% 3385828828th
2 a. «2 S a :2 c a o mm a m: ES 8:. .83.: 25.58283
2 a 9: mm in ca N a : S a 8m 3 5 28833 9.328833
a a ...2 S. a M: 2 a a. E a NR :2 202 Seam £28883
2 a 2. 2 a 2: o a o 2. .4 ca 32% 8.5. .588 $238283
2 b a SN 2 4 5 m “an 2: a 48 Rm 2% 3...: 25.8st 328883
3 a 2m 3 a SN 2 a me 82 a 8m N 3 33838 328883
2 in SN 2; a 8m 2 ..." 5 N: a 92 Em snagmcuuaongm
om<ZD ems/E $885 $8 22:6

 

9:53 cougqoocou £88m

..oézo .0222 .3835 55 8329:. s 883 «noes 28 262 cc as? a 8.8g 822.322 2: co .558 5282 ....m 539

125

.c. u 5 232808 o9: a 32885 e5

ESQZQ mi omm 3 3.272 m: omm 8 65¢ 3.223 we wEBmHBoa w: on 55 3235 mm? conga 02:22qu 38225 _

 

 

 

5 H or: bum H bmm 3 H rm 2: H N2. 37% mEEQoExmfi uaoooooiohm
02 H ova v: H Sam 3 H «in 3N H 3% memn DURV .2823 25.838er
wfi H 3.4m N2 H «Sm o3 H 32 «cm H womb Na .5 ”3.2.833 .QSBQQEooQ
o3 H nomm mom H mmmm mm H 2N wmo H no: 2: M02 .23.qu ufizooeofioa
2. H 28 E H 38 SN H 32 2: H 2 mm 38m 8.? .88 3288283
cma H 3% mm H venm S H wt: cam H mmmv RN M92 mm: muotoMBQ §=B§Sooa
N3 H 32 2 H #3 o H o c H mww N 3 mSEQoEoo azaooaoBoQ

mbm H owns 2 m H ooh: o H 0 COM H ooh: Tum Eztfiooaohmbm
0375 0322 032on 88305
98%.: 8952828 £88m

 

..oézo .035 .3335 .23 8386.3 s 8.83 «568 28 88. co 259: 3 8E 25385 2: .6 23:8 5282 .2 32¢

126

.G n 5 2355 we: a 3385 EH.

€35 m1 o2 s @3378 m: cm s 6:09 3568 Ho Habaosa w: 8 55 38: 83 Esme 3 €qu 33.32 385:: a

 

 

M: H S: seas t HE»; _ 0328??qu

2: H g H H 9:. mm H 2» S H NE 3% gmficsss HSSSEEE
2: H :5 «a H S» M: H as 8 H E: E2 8,; 22:2 25.53203
8 H to S H as 2 H Sm c H How a 5 28225 2583393
2: H E. we H o? 3 H H? o H S: 2: M02 Easm §B§§3
an H 08 S H o8 an H ME. a H o3 28m 85. 3,48 EEESHS
R H 2 b x H «8 SH 5 on H :2 5 M92 a»: $9.233: 332293
2 H c3 2 H mm E H 3. S H can N 3 $23038 Engages
o: H £2 3 H 3.2 cm H 8M: c: H 88 Em §§8€§m

0229 3472 09385 £8 225,

 

2:53 8:838:00 BEqufimU

H.945 .852 .888; a? 8535. a 853 «563 28 082 Ho 853 3.2.5: 05 .8 .558 $6335 .3 39¢

127

AH n 5 23:58 ooh». a “.8385 28

Ehézo m: cm a 35372 mi cm s 422 3582 he mesmssa m1 cm .23 38: H3 fine 3 8.8.3 gammaoau 38E?

 

 

 

M: H S ~ H K 2 H 3 3 H men 27% aazamoztwfi “.5883th
Em H m2: 2 H mm o3 H coo 2: H ma m was DUEH tuna? 25.83963
mm H 32 S H 32 «mm H 38 o H oumm ma 5 Haucmém: 25.83083
n: H mg 3 H o2. mm H 83 o H moon z: M02 23%»... HEESSQS

S H at. E H w: mm H 2m a H omw ommmm UUH< Bug 5563963
o2 H E: o: H wma M: H 3: NR H mmom 3m M92 062 anotuwin 25.83893

w H NE E H m3 mm H :N mm H won N 3 “339638 HSNBRSQQQ
mow H came mm; H 2 3 32 H 88 mmm H 83 Em Eatfiogohmbm

02an 83,2 3865 88325

2:35 nougdooaoo ougaofido

 

_.H<zn .8<zm .039on 55 Eggs: s 83%;. «568 28 82: Ho 8&3 again 05 Ho 82:8 25385 .2 BBQ

128

3.4.3 Effects of lactic acid bacteria fractions on RAW 264.7 cells.

To assess the effects of lactic acid bacterial fractions and heat-killed cells on
macrophage function, RAW 264.7 macrophage cells were stimulated in vitro with 50, 250
and 500 pg (wet weight)/ml of each bacterial fraction and with 1 pg/ml LPS, used as a
positive control. After culturing these cells for 48 h, cells were used for MTT analysis and
supematants were monitored for cytokines.

Macrophage structure varies with their degree of activation. Figure 3.2 shows the
rate of activation reached by stimulation of RAW 264.7 with lactic acid bacteria fractions or
LPS. These cells are generally large, irregularly shaped and they often have an eccentrically
placed round or kidney-shaped nucleus with one or two prominent nucleoli and finely
dispersed nuclear chromatin (Figure 3.2a). Upon activation, the cytoskeleton that surrounds
the nucleus extends throughout the cytoplasm to the cell periphery. Actin microfilaments
immediately beneath the cell membrane are responsible for the prominent ruffling,
locomotion, and pseudopod formation, as well as influencing endocytotic events (Figure
3.2b, 3.20, 3.2d). The cytoplasm can be seen containing fine and multiple large granules
(Figure 3.2c, 3.2d). Vacuoles are frequently seen near the cell periphery, reflecting the active
pinocytosis of macrophages.

The proliferation of macrophage cells, measured by MTT assay, was enhanced when
RAW cells were cultured with 50 pg and 250 pg of any of the fractions, generally showing
a significant increase with respect to the unstimulated cells (control), but this effect
decreased when cells were stimulated with 500 pg of either heat-killed cells or cytoplasmic

fraction (Table 3.8, 3.9, 3.10). The results suggested that lactic acid bacteria stimulated RAW

129

.2; 8m: H 58 8:885 a 8 E .883 =83 =8 8&1 8m 88 88885 a E
O .888 08888 8&1 8m 58 88885 a .8 a .8883 888 3888 m8 858 88885 a 8 2 .82 8
£8 amour—@888 Evom a; co ommam UUH< 830 538328qu 88:23 “cocotmu 53, 53335 mo Hoobm .N.& 9.55

130

 

131

mod V A?

$58598 8883 95 mo o>u85moa8

8a 98 a. .I. 5 was?» 3950 3 83 35535 waning 3 331530 33 owSEo gage“ .EKZN
.950: 3 How «€803 mo 3093588 3553 £5 38:3 203 253—8 “A: x 9 E8563 >3):

 

 

*mbd 205:8 +Vw1 _ cutanoommbomomfi

mod we; mo; mmTQm uEEQQEme mgucuSQwhm.
...mmd Mod *mm; mummm DPNV $32 2:28?ch
ad 2: RNA mo 5 2853»: 35.83293
mo; LVN; «mm; 2: M02 tmmaaw 23.63893
m3 2: $2 ammmm up? $8 25.63893
:2 sad .2 ._ an v62 3,: 3.5%3 3293293
had *3; ...mmA N 3 .3330th «<53qu33
mad RNA *2; 7mm Eztfiucechmbm
m: 8m .03 SN m1 on «568

 

N 3&8 2e

 

gamma E >9 35308 on: #30163 3: mo 8:803qu no «£323 Boa oflofl 3:293: mo Bobm .w.m mqmflh

132

36 v gr.
.mEoEuomxo 2qu 95 .8

03558“? can 28 3 u 5 mod—136550 3 Saw Econ—toga wEE>€ 3 381630 33 omSEo 2623 EN

.25.. M3 5 SEE 5:562 we aougaeas 82:3 5E 85:5 22» €38 n2 x a £8>$~ B§_

 

 

 

m3 *9; $2 3% 33%:22: mausoifim
Rod .82 *3; 92m UEE§§ 25.83%3
3o 2: *3 ._ 8 5 25232 328833
E ._ *2 ._ m: .2 M02 .Easm §B§§3
.86 *2: *2 .H 23m 85. E8 3223993
3o *2 ._ .32 Rm 52 am: scrumse $38833
a: .33 :2 N 3 33%38 328393
.33 2: a: Em §t§§€$m
w: 8m w: o2 m1 8 «Baum
mob—Bo 28

 

.53 SE 3 Base on: :8 Baa 22m Ho 89202qu 8 «568 28 262 Mo 888 UEEE 2: mo Spam .3 39:.

133

mod v a?
3:25—35 8838 95 .«o

.oESnomoEB Ea 93 c. n 5 $38, 35:8 3 Sun 3585me magi“. .3 33:28 mm? owns? 03:22 EN

.28: a. E SEE 5:833 .0 Sougfioas Song 5? 85:5 203 2528 “S x a £853” 35:

 

 

 

*MNA $9— *w~.~ mmzm aEEmcELuS maoueofimmhm
*3; *oma *3; memm Ubkwtfiamx mangoUc‘N
Lad *mo._ ...N. A No .5 38.333»: uzzwuceouuu‘w
a: .52 * H E z: 52 Eaum 25.33033
mad *2; *5; ammam 095‘ .828 szucQSocu
...wog *2; “RNA RN M02 mm: SGEQMBQ SSBQQSUQQ
.7ng *9: .5: m 3 méEmoEoc 2583893
*NNA two; “woo; Tam EatuBcethm
3 8m m1 o3 m: on @308
N 85:8 28

 

53mm :2 3 3:538 on: :8 Even 342 .«o doumuomaoa do «€203 Boa 363 mo dog :9.» =8 05 .5 «Sum .3 .m mqm<h

134

264.7 cells exhibited enhanced proliferation that seemed to be concentration-dependent
particularly for stimulation with the cell wall fraction of these bacteria.

Exposure of RAW 264.7 cells to lactic acid bacteria fractions also increased TNF-a
and IL-6 levels in a concentration-dependent manner. For heat-killed cells, production of
TNF-a was significantly higher than in control cells with all bacteria demonstrating a great
activity. In general TNF-oc levels were higher than the levels produced by stimulation with
a positive control, (LPS, 1 rig/ml) (Table 3.11). Similar trends were found when analyzing
for IL-6, except that the levels here were lower as compared to LPS (Table 3.12). For heat-
killed bacteria fractions, pre-treatments with protease, DNAse or RNAse did not
considerably affect the production of cytokines by RAW 264.7 cells (data not shown). Since
non-viable bacteria were still intact, protease or nucleases may have had difficulties
diffusing inside the cell.

Cytoplasmic fractions from lactobacilli and streptococci induced marked TNF-a
production (Table 3.13), with significant increases being observed for all concentrations. L.
helveticus, L. reuteri and Bifidobacterium showed an opposite effect, the latter producing
considerable levels of TNF-oc at 50 pg and 250 pg but decreasing dramatically at 500 pg, and
lactobacilli showing a lower level at the highest concentration. Most of the cytoplasmic
fractions induced lower levels of this cytokine than LPS. In general, similar results were
found for IL-6 (Table 3.14).

For cell wall fractions, TNF-a levels significantly increased with concentration for
all species. In general, these increases were lower than the ones found for heat-killed bacteria

(Table 3.15). Stimulation with cell wall seemed to be more effective than with cytoplasmic

 

135

mod v K...
.fiaoataxo 0:538 28 mo o>u8=82q8

.08 98 3 u 5 85:3 15:8 3 88 358598 mania 3 6833—8 33 0320 0338 .oina
25; we 5 €823 mo aougaooaoo 225 a? 8:28 Boa €28 “S x a m=8§$~ >53

 

 

 

Lam 205:8 +Vw1 ~ oc§88boqoafi
*ovm *02 ... ~ m mmfiém wéimcztwfi @3883th
.33 ... SN ...mN— «mama 992‘ {8:3 33.83825
.52: *0: *avm No 5 5623.3: ufitugofiuq
$2 3% ...me I: 202 :3wa ufizuuefiocq
*wmh .38 .33 ammom 095‘ $.89 §=B§Su¢x~
..on :5 ...wwm Rm 52 $2 ”argues $583393
...Ew 3Q. ed: N 3 9.521038 .SSBBSQQQ
*mam ...omm 992 7.5 Eztfiogcnmbm
m1 8m m: on” wan «56am

«35:8 28 6.2

 

“:8 3% 22m 3 88:85 329 5 £063 28 262 8:38.. mo scum ._ _ .m 53?

mod V a?
.fiaofitoqxo 8838 25 mo 2585852

.08 98 c. u 5 83.2, 85:8 3 83 3855me wEuEv 3 883qu was omfifio 2532 0.15
was: wv 8.“ «563 «o Eougaooaoo £533 55 3:28 803 283:8 mg x 3 £62.63 >339

 

136

 

tom 23:8 t w: _ outfiooabaafl

*Nm *om 1‘ mflém. uEERoEwa mucusooimhm
3.2 A: *8 ~28 on: E32 3293893
*2 an *2 mm 5 gang»: 35.83033
*N: E: *3 2: v62 Easm 328%33
*mm *8 3 Sam 8.: .589 ”3593993
*3 an EH 5 52 $2 25.2333 §=a§§3
*5 .3 QN N 3 “323633 wézuuaoficu
$2 ...m: ...: Em se§8€§m
m: con w: on warm 3588

 

"Bugs 28 9d

 

£8 .52 Ed 3 836.3 on: 8 «.563 2% £8. 8:332 mo Bum .24. 39:.

137

mod V l...

8808595 08898 98 .8 02880382

08 88 3 u 5 82.3 8888 3 83 88088098 @825. E. 388080 v.83 owgo 038—2 5.?
8:0: we 8.“ 80828 88.583 .8 8388888 85.53 55 38:3 895 98330 “2 x 3 £8583 >39:

 

 

L NN 9288 +v m1 _ 3588boqoqfl
*wN *3 m 37% «3388me manoeucimém
*2: .52 $6 NnNMN 092x .583 85.83883
*ofl “..ch 1.3 Na 5 88.89:»: 8583833
o: 3N“ *Nv 2: M02 .28.wa SSBQQSUQ‘N
$8 :8 $2 83m 8: $80 328833
$8 *5 :2 Rm v82 33 ”3ng §E§§3
“33 ...mN~ *2: N 3 8:838:88 .QSBQQSUQQ
m *8 *2: Em 58293838
m: can mi on w; 8 888m
N88:8 28 6.2

 

.88 .38 BE 3 88:88 size 8 8208 28 88. 8 888 08385 2: “a 68m a; 598

138

mod V n?
fiaoafloqxo 88598 95 no

0358858 08 Ba 3. u 5 mos?» 65:8 3 83 Enofitaxo wfivg E. 38328 35 owgo 9532 9):”
go: we Ba nouns 8:583 .8 8093588 325» 55 3:53 803 2:5on “3 x 3 38563 233m“

 

*vom 905:8 +vw1 ~ ocgoommbaoms

o 2 o mflbm wfiERcExuS 9.38803th
3% H 12 3 memm 905V 38:2 3563893
*3 two ...wm mm 5 ”33333 $583823
,5: mm b 2: M02 {863% mazfiueofiuu
,L R .35 2 ammam UUH< Eng 25.83823
52 :2 S 5 52 $2 28:qu 328833
*m 5 *ma *8 N 3 @3230393 uézuueeuua
*3 *3 o Em §§§o§m
mi 8m m: cm” mi on «Baum

 

"3:80 28 can

 

2.8 .28 Ed 3 883on «an 8 8202 28 263 mo 883 €585 2: mo seam is mam—<9

139

36 v a?
.mEoEflono Baa—“mom 95 no

gunfiomoae 08 ES 3 u 5 8:?» 35:8 3 83 358530 waning .3 333—8 83 0?on 2528 6&2?
35: we 8m gaunt 5:583 .«o 3023588 £553 5.5 3:58 89: SEE—8 «A: x 3 £8“..ch Baa}

 

 

 

Lam £888 +vw1 H 85:82—88“;

*NE *3 *mm mmfiém mEEmoEme aguoucimém
tam 3: 3 ER 8: E38 328833
...mov 33 *mom 8 5 naucmé§ “52.83?th
.5: 1B 3% 2: M02 {8.3% uaEugSucu
...w; 3mm mm ommam UUH< .38.» $583393
3% .59». a2 5. 52 $3. 3.23ng assfisau
*w: *3 _ v N «A ufiERchc szuueSuu‘N
*3 3m *: Tum Eigogohmbm
m: con m: cm m: cm «583

«85:8 28 RYE/E.

 

.260 ~38 >32 3 88:85 329 8 $588 28 onus .3 883 =29 =8 2: mo scum .2 .m 39?

140

material and/or LPS relative to TNF-a production, especially for L. bulgaricus, L. casei, L.
helveticus. Production of IL-6 in this case was also significantly increased for all bacteria
species, with L. bulgaricus and L. reuteri showing the greatest effect (Table 3.16). In general,
values were slightly lower than the ones found for cytoplasmic and heat-killed fractions.

The above described immunopotentiating activity in this fractions could be due to
nucleic acid. Table 3.17 represents a typical result for cytoplasmic fractions free of protein
showing in general the highest activity when compared to those free of RNA or DNA for IL-
6 production. Cytokine production by cells stimulated with cytoplasmic fraction was in
general not significantly decreased by the protease treatment respect to the untreated fraction.

Nitrite levels were used to measure production of nitric oxide (NO), another
important macrophage mediator. Table 3.18 and 3.20 summarize the effects of stimulation
with heat-killed and cell wall fractions respectively on NO production by a macrophage cell
line. In general, NO production increased in a concentration-dependent manner reaching
values similar to the ones exhibited by LPS stimulation. L. reuteri and S. thermophilus
exhibited an opposite effect, where levels of nitric oxide decreased with the concentration
of heat-killed fraction used to stimulate cells. Stimulation of cells with the cytoplasmic
fraction of lactic acid bacteria disclosed an inverse effect than the observed for heat-killed
bacteria and cell wall (Table 3.19). NO levels decreased in a concentration-dependent
manner for all bacteria except for S. thermophilus.

To rule out the possibility that TNF-a, IL-6 and NO production was induced by LPS
contaminants within the bacteria extracts, RAW 264.7 cells were cultured with up to 50

ug/ml polymyxin B and with the bacterial fractions or LPS. Polymyxin B is an antibiotic

141

 

 

36 v s?
358530 Swanom 92 mo

0338829“ 0.8 98 G n 5 83:3 35:8 3 S3 aaoauomxo wfiug B 338:8 mg» owfifiu 0258 end”
mac: wv 8m codes 83563 .«o aoumbnooaoo £553 .23 3:53 953 €558 a: x 3 385.com 34$:

 

 

 

:8 2238 t w: fl ougsabaas
*m o o 375 mEERoEme mauocuoiuhm
*0: mm o NbNmN 00& V .2832 maSochuq
*3. *mm N. Na 5 agave»: wfimuuesga
... fl m *8 o z: M02 {3qu .SSBSSUQQ
EN .8 ~ ammom UUH< .838 .SSBESUQQ
..m: * _ S *8 Rm M92 33 §§m§ $203803
*m *N _ a 3 “5293039: SEBQQSUQQ
v m o Em §§8€§m
I? can fix o2 mm on «568
L838 28 can

 

£8 33” Ea 3 8:263 can 8 «583 28 263 he 888 as» =8 2: be firm .2 .m 359

142

888898 388m 23 mo 08888888 08 88% 55938 88888

a: 8%: o 8 820 828 £23 883 c8888. 2: 8 8&8 55 8858 888 8 888 9: 968 82.5 ”8oz

.Q n 8 2: x 8083 085888 88888v 825» 8.88 .3 88 8808898 868% 3 8888 83 0988 8688 ch?
88: wv 88 883 88883 we 80888888 82.28 53> 888 203 A8580 «2 x 3 8.8588 >32:

 

 

o3 own o3 mmTfi 8838885 88883th
oi oi S N mnmmm 905 V {8:3 8303883

c o o2 Na 5 888888 mzfiuueouucu
c: on o2 2: M02 tummuw 88288888
o2 cw o2 33m UUH< .588 8588884
9: o: 8 am 82 82 28.283 328883

c o o: N 3 8838.888 85.83884
2 2 28 Em 53.229388

0375 3472 8885

 

”A8888 88888. .xv 0938 9622 ok:

.88 E8 38 3 8838 on: 8
0378 a 8<zm .3888 SE 2058: .08 8823 28 085 Mo 888 again 2: 8o 38 .2 .m 898

 

143

no.0 V n?
gauge 0538 95 .8

3:85352 8a 98 Av n 5 82? 35:8 3 83 3.58595 36% B 38328 33 owgo 2638 .MOZN
was: me He €803 .8 883588 gong 5E 33:8 203 9:528 “2 x a £85.va 3.4.5

 

*2 £888 +v m1 ~ outwnoommbomomfi

*0 *v .2 mmfiém wfiERoELmS manusuoambm
i. *3 *3 «bag DURV tutu? métuaefiucu
...: *m fl *3 mm 5 nauzmfig SSBSSQQQ
*2 *3 3m 2: M02 235% maEuceSuuQ
...M: *2 *2 ammom UOH< 8.39 5583225
*2 ...m *2 NMN M92 mm: nautaEg $5838.93
...NN *3 L. N 3 QSERQEQQ SSEQQSQS
*2 *2 *2 Em Eatfiufigngm
w: con m; cm m1 8 88mm

 

moans Be .62
£8 3% 22m 3 88:85 .62 8 @063 28 282 33-32 mo 60mm .2 .m 59?

144

mod v a?
.SSEtonxo 2838 95 .8

8355888 2m 98 Av u 5 8:13 35:8 3 Saw aces—coax”. $538 .3 cam—:28 33 0320 0528 .MOZN
was: wv no.“ .883 5:583 mo 83.5588 Bot? 55 35:3 22» GEE—8 a: x my $86ch 3g"

 

*2 905:8 +v w: M 85:88boaoqfi

...w *m _ mmzm §~38585 888883th

c m *2 NbNmN DEV .582 35.83883
*3 *3 *3 Na .5 882838: azimuaaeuuu
*2 *2 *2 2: M02 28wa $583083
...: *2 *2 Swan 00,? .528 5288883
:2 ...: *8 Rm 882 $2 25.2%: ӣ33803
*5 *3 ...: N 3 wEERoEuc SSEueSouQ

N *0 *m Tum Eatfiounohmbm
m: con m: cm w: on 838m

 

"sauce 2e -moz

.2? $48 252 B 8838.“ .62 8 «563 28 onus 8o 888 €385 2: .8 68m .3 .m Mimi

145

mod V Ks,
8:88:38 888:8 95 .8

8380:8852 2: v5 3 u :0 829 85:8 P 8:: 28882898 383% .3 88328 83 omgo 8288: .62H
8:0: we :8 888.2 8:883 .8 2285888 32.5» .23 8:38 883 SEE—8 «A: x 3 mzoofivoN >39:

 

*2 905:8 +51 2 octagoonbomoqfi

*2 *m o mngm 8:558:25: 8888885”
*2 *2: *2 NFNMN 0928‘ .2833 25.888885
*2 *2 ...M: Na 5 388385 w58§o8§~
*5 *2 ...m 2: M02 imamum 25885088q
t: *3 .3V amnam 00.5fl .838 $583883
*2 *3 ...om 5 82 mm: 9.8.832 38383
*2 *3 c N m..— mSERoBuu 85.888884
...: ...m o Em 5588888

m: con an own w; on «:28:

 

N388 28 .62

2:8 :38 22: B 8:883 .62 8 «532 23 as: we 8:8: a? =8 2: mo .88: 52 3:5.

146

capable to induce outer membrane permeability in gram negative bacteria (Nikaido and
Vaara, 1985). The basic molecule of LPS consists of a lipid component (lipid A) covalently
bound to a heteropolysacharide (Stewart-Tull et al., 1985). Lipid A is responsible for the
mitogenic activity attributed to LPS. Polymyxin B is able to inactivate lipid A by attaching
to it affecting overall LPS activity. The results indicated a constant stimulation effect of
bacterial extracts independently of the polymyxin B concentration (Figure 3.3). On the
contrary, LPS activity declined when concentration of polymyxin B was increased, affecting
proliferation of cells and decreasing production of TNF-a.

The effects of lactic acid bacterial stimulation were determined in cultures prepared
from RAW 264.7 macrophage cells at different times of exposure (Figure 3.4). Cells were
incubated with L. casei ATCC 39539 fractions for 1, 6, 12, 24, and 48 h. The capacity for
elevated cytokine production was observed in RAW 264.7 cells as early as 1 h for TNF-a
and 6 h for IL-6. Levels of these two cytokines were constantly increasing for all treatments
regardless the fraction used to stimulate the cells. Detectable levels of NO appeared after
incubating for more than 24 h and these levels could be related with the considerable boost

experienced by TNF-a after incubation for 12-24 h.

3.4.4 Effects of lactic acid bacteria fractions on peritoneal cells.

Peritoneal, spleen, and Peyer’s patch cells were incubated with 500 pg of heat-killed,
cytoplasmic or cell wall fractions from lactic acid bacteria as well as with 1 ug LPS for a
period of 48 h and analyzed as described before. Enhanced proliferation of peritoneal cells

was more evident than activation based on microscopy (Figure 3.5, 3.6). Spleen and Peyer’s

147

Figure 3.3. Effect of polymyxin B on Lactobacillus casei ATCC 39539 fractions and LPS
stimulation properties. A) MTT assay B)TNF-a production.

% control response

TNF-a (ng/ml)

120

00
O

O
O

.h
C

N
O

O‘\
O

K)!
O

A
O

DJ
0

N
O

h—A
O

O

148

 

lllllllllllllllllllllllllllll

 

—I - LPS lug/ml

—A— Heat killed (500 pg)

——y-- Cytoplasmic fraction (500 H8)
.... Cell wall (500 pg)

 

Jllllllllllllllllllllllllllll

 

 

 

 

Polymyxin B (pg/ml)

 

149

Figure 3.4. Kinetics of IL-6, TNF-a and NO after stimulation of RAW 264.7 macrophage
cells with Lactobacillus casei ATCC 39539 fractions for 1, 6, 12, 24, and 48 h.
ND=non detected.

IL-6 (ng/ml)

TNF-a (ng/ml)

N03'(uM)

M O\

A

LJ_llllllJllllllllllJlllllllll

r—A N DJ

t—I t—I N N
VI 0 VI 0 LII

[\J
kl]

20

15

10

VJ]

O

150

 

Control=ND

+ Heat-killed fraction (500 pg)

—l - Cytoplasmic

—‘— Cell wall fraction (500 pg)

—v—- LPS (1 pg)

/

  
  
 
 

fraction (500 pg)

 
 

 

JillllllI1111114111]lllllllllllllIllllllllllllll

 

 

 

 

18 24 3O 36 42 48
Time (hours)

 

151

.24 381 a 58 8:885 a 8 E .888 =8 =8 88 8m 58 8888
a 8. G .888 08888 :81 8m 58 8888 a 8 E .2883 888 8888 8 858 8888 n 8 2
.82 8 88m 8: 8.8 888838 888 88% 58 888288 8888 8 8888 8 88m .m.m 28$

152

 

153

Figure 3.6. Effect of incubation of peritoneal macrophages with different fractions of
lactic acid bacteria (x 100). A) 24 h incubation without any microbial product (control).
B) 24 h incubation with 500 ug/ml Lactobacillus acidophilus cytoplasmic fraction. C) 24
h incubation with 500 ug/ml Lactobacillus acidophilus cell wall fraction. D) 24 h
incubation with 500 ug/ml Lactobacillus bulgaricus cytoplasmic fraction. E) 24 h
incubation with 500 ug/ml Lactobacillus reuteri cell wall fraction. F) 24 h incubation
with 500 pg/ml Streptococcus thermophilus cytoplasmic fraction.

 

155

patch cells did not show any difference respect to the unstimulated cells. Similar results were
obtained for all bacteria species.

Exposure of peritoneal cells to lactic acid bacterial fractions increased IL-6 levels
respect to the control cells (Table 3.21). Production of IL-6 was significantly higher for all
bacterial species, showing very large levels regardless of the cell fraction used. Stimulation
with cell wall and cytoplasmic material seemed to be more effective than stimulation with
heat-killed bacteria and, in some instances, LPS. Bifidobacterium, L. bulgaricus, L. casei,
L. gasseri, L. helveticus and L. reuteri appeared to have a high stimulation rate on these cells.
Since peritoneal macrophages were not specifically isolated from other type of cells, we
cannot discard the possibility that some other cells (i.e. T or B cells) besides macrophages
were contributing to these high IL-6 levels.

Production of TNF-(X by peritoneal cells was not detected in any case. Similar results
were found when analyzing for IFN-y and NO. Spleen and Peyer’s patch lymphocytes did not
show any indication of either IL-6, TNF-a, IFN-y or NO production.

The effects of lactic acid bacteria stimulation were determined in spleen, Peyer’s
patch and peritoneal cell by incubating cultures with L. casei ATCC 39539 fractions for 1,
6, 12, 24, and 48 h. Supematants were collected and analyzed for TNF-a, IFN-y and NO,
showing no production of any of these cytokines or intermediates. Interestingly and only in
peritoneal cells, IL—6 was very rapidly produced, appearing after 1 h in considerable levels
and escalating until incubation was maintained for 12 h. In general IL-6 peaked either at 12
h (for cytoplasmic and heat-killed fractions) or 24 h (for cell wall and LPS). Afierwards,

levels of IL-6 started to decline slowly (Figure 3.7). Spleen and Peyer’s patch cell cultures

156

did not produce detectable IL-6.

157

mcdvm 338385 95 85:8 05 8 8032 5:85:me b83389.
3:08.58me «8898 95 he 0385:3392

0.8 983. u 5 260 3585.58 mos—S» 15:8 .3 Saw acoEflomxo 368% .3 cows—8:8 v.83 owcmno 2558 can”
885: me How cog 3683 .«o w: com 55 698:3 983 £828 «2 x B 38>.ch 3&5

 

 

.55 2888 +5 2881 3 88808588:
.58 58 .53 23m 8.338585 8.888885%
.52: *mwfl ...mmc memN Uptx .2833 5.458388%
.53 ..hog *mcw No .5 «Schmém: SSBQQSUBN
.651 L: S .53 2: M02 88.88% mancaSomq
.58 .55: .58 8.3m 8: 8.8 8288883
.58 .58 .58 Rm v52 82 88.888 2588883
.59“ “.53 3% N «A 833338.83 53285033
.58 5% .53 Em §§88§m

:8» =8 88.85 £8 223
N888 28 can

 

8888888 88888 3 88:88 9.: 8 8.888 28 083 8 a: 83 888.. =88 =8 c8 088828 8:28.85 8 808m .38 855.

158

Figure 3.7. Kinetics of IL-6 after stimulation of peritoneal macrophages with
Lactobacillus casei ATCC 39539 fractions for 1, 6, 12, 24, and 48 h.

ND=non detected.

IL-6 (mg/ml)

159

 

3O

 

 

Control=ND
+ Heat-killed fraction (500 pg)
—l - Cytoplasmic fraction (500 pg)
—‘- Cell wall fraction (500 pg)
—v—- LPS (1 pg)

1 l l l I l

 

12 1 8 24 30 3 6 42 48
Time (hours)

 

160

3.5 DISCUSSION

This research showed that in cloned macrophages lactic acid bacteria and their
fractions altered cells morphology as evidenced by pseudopod formation and increased
number or density of fine granules near the cell periphery and inside their cytoplasm.
Macrophages belong to the mycloid lineage and they play a key role in inflammation and in
host defense (Abbas et al., 1994). Functions including phagocytosis, antigen presentation,
mediator production, anti-microbial and tumoricidal activity contribute to its role in host
defense. They are controlled by numerous dynamic stimuli from the tissue
microenvironment. The effects of these stimuli on the macrophages are usually associated
with specific alterations in gene expression after a series of intracellular signals initiated by
ligand-receptor interaction (Abbas et al., 1994). The cellular spreading observed in RAW
264.7 cells could be an early indication of an increased cellular activity. This possibility is
supported by the observation that lactic acid bacteria and their cytoplasmic and cell wall
fractions were capable of stimulating RAW 264.7 macrophages to produce very significant
amounts of TNF-a, IL-6 and NO. In general, bacterial fractions stimulated macrophages and
increased their proliferation.

The multifunctional cytokines lL-6 and TNF-a are major effector molecules in
bacterially mediated local tissue destruction and necrosis. TNF-a plays a role in such
augmented toxicity inducing gene expression of several cytokines. TNF—a has been reported
to be induced in mouse peritoneal cells 3 h after the intraperitoneal injection of

Bifidobacterium cell wall preparation (Sekine et al., 1995). TNF-a responses involve

161

increased rates of transcription of particular target genes, often through activation of NF-KB
or AP-l transcription factors. At low concentrations, TNF-a acts as a paracrine and autocrine
regulator making cell surface more adhesive. Also, TNF-a stimulates other cell types to
produce cytokines such as IL-1, IL-6 and TNF-a itself (Janeway et al. 1997). At greater
quantities TNF-a enters the blood stream as an endocrine hormone and act as an endogenous
pyrogen and it is able to induce the expression of nitric oxide synthase (NOS), which results
in conversion of arginine to citrulline and NO. Interleukin 6 (IL-6) is synthesized by
macrophages and other cells in response to TNF-a, among others. IL-6 serves as a growth
factor for activated B cells during B cell differentiation (Abbas et al., 1994). It can act as an
autocrine factor and also as a cofactor with other cytokines. Previous research conducted in
this laboratory using heat-killed lactic acid bacteria to stimulate this same cell line, produced
IL-6 and TNF-a (Marin et al., 1997a, 1997b, 1997c). Our findings are similar to Miettinen
at al. (1996) who reported an induction of the proinflammatory cytokines TNF-a and IL-6
by viable lactic acid bacteria and fixed in 2.5% glutaraldehyde (non viable) lactic acid
bacteria when used to stimulate human peripheral blood mononuclear leukocytes. In other
studies where RAW 264.7 macrophages were cultured with microbial spores, induction of
TNF-a, IL-6 and NO secretion was observed (Hirvonen et al. 1997). Purified cell walls from
Streptococcus mitis were also seen to induce TNF-a in vitro in whole blood of LPS sensitive
and LPS resistant mice (Le Roy et al., 1996).

Several investigations have examined the effects of lactic acid bacterial fractions
(especially Bifidobacterium strains), on immunopotentiating activity. For example, using

sonicated cells of a strain of Bifidobacterium adolescentis, Lee et al., (1993) showed that

162

those cells stimulate lymphocytes from Peyer’s patch and lymph nodes. Another species of
Bifidobacterium, B. breve, accelerate proliferation of Peyer’s patch cells, particularly B cells.
The proliferation of B cells is enhanced when the supernatant of plastic adherent cells
cultured with B. breve (either whole cell or a cell wall preparation) is added. This indicates
that B. breve activates plastic-adherent cells and that these cells secrete a soluble factor that
enhanced proliferation of B cells (Yasui et al., 1991). Takahashi et al. (1993) investigated
the interaction of cell fractions of lactic acid bacteria with the immune system, using
Bifidobacterium Iongum and Lactobacillus acidophilus. In mice fed B. longum for more than
8 weeks, a strong antibody response was detected to the cytoplasmic fraction, but not the cell
wall fraction. In mice fed L. acidophilus for more than 6 weeks an antibody response was
detected against the cytoplasmic and cell wall fractions. Sekine et al. (1994) demonstrated
that cell wall preparations from Bifidobacterium infantis induced polymorphonuclear cells
and macrophages when injected in peritoneal cavity. These cells inhibited the growth of
tumor cells in vitro.

During fractionation procedures of bacterial cells that contain large amounts of lipids,
disintegration is impaired by clumping, and they are best subjected to preliminary solvent
extraction at room temperature (Work, 1971). However, these types of fractionation
procedures have the disadvantage that there is considerable destruction of sugars and
presence of organic solvent residues, which is not desirable due to the possibility of
stimulatory effects or cell damage. Less drastic methods of wall isolation from cells were
used in this research, such that polymers (peptidoglycan), mucopeptides, teichoic acids,

polysaccharides and possibly some other components were present together and possibly

163

interacting with each other.

Heat-killed, cytoplasmic and cell wall concentrations needed in this study to elicit
maximal cytokine production were substantially higher than the concentrations of LPS (1
pg/ml) that induce comparable levels of TNF-oc and IL-6. Some components of gram positive
bacteria (streptococci, micrococci, staphylococci, etc.) produce significant amounts of TNF-oc
and IL-6 at concentrations above 100 ng to 1 pg of cell walls per ml with a maximal
production requiring 10 to 100 pg of cell wall material per ml (Heumann et al., 1994). In
those cases the wall of those pathogens may contribute to the septic shock induced by gram
positive bacteria. Whole cells were effective as mitogens and activators, but their use may
make interpretation of results more difficult than purified products because the presence of
diverse components in intact cell, which may act via different mechanisms (Ringden et al.
1977).

When I explored the kinetics of TNF-a, IL-6 and NO production in macrophages, I
saw that lactic acid bacteria induced RAW 264.7 cells to produce significant amounts of
TNF-a and IL-6 after just 1 h or 6 h of incubation, respectively, and those levels were further
increased, while significant increases in NO occurred only after a substantial amount of TNF-
a and/or IL-6 was secreted. These results follow a similar pattern to the effect of cell
stimulation with LPS, which can serve as model for cell activation by microbial products.

LPS has numerous important cellular effects. When macrophages are exposed to LPS
in small concentrations they secrete not only TNF-a but also IL-6, IL-1, IL-8 and IL-12.
These cytokines activate other immune effector cells (Janeway et al., 1997). The LPS-

induced signal transduction pathway has not been completely defined. However, one of the

164

signaling pathways initiated by LPS in macrophages involves the breakdown of
polyphosphoinositides and the subsequent generation of intracellular calcium. LPS is
recognized by high-affinity receptors (Gammage et al., 1996) in soluble and cell membrane
form (CD14 and the B2-integrin CD11c/CD18). The highest affinity receptor, CD14, is
located in macrophages and it is a cell surface protein. LPS binds to macrophage CD14 and
this complex stimulates the induction of cytokine gene transcription and eventually TNF-a
release (Figure 3.8). Mouse TNF promoter/enhancer region contains four functional NF-KB
sites (Beutler et al., 1995). In spite of a completely different chemical structure, Gram
positive bacterial extracts and LPS can induce similar biological effects and this might
activate the same mediator system. Relatedly, peptidoglycan present in gram positive bacteria
has been reported to induce inflammation (Stewart-Tull et al., 1985). Peptidoglycan also
stimulates macrophage activity inducing the release of numerous macrophage products
including endogenous cytokines, intermediates and prostaglandins (Abbas et al. 1994). Their
induced secretion might well have important implications with respect to the pathogenesis
of the inflammatory process that occurs after exposure to peptidoglycan (Dziarski, 1982).
Like LPS, peptidoglycan is a T cell independent B cell mitogen and polyclonal activator.
The cell walls of Gram positive bacteria (not lactic acid bacteria) at a dose of 10 pg
(1 x 106 cells) produced, in vitro, maximum activation of the peritoneal macrophages of
guinea-pigs due to the presence of both peptidoglycan and peptidoglycolipid (Takada et e1.,
197 9). It has been also reported that induction of cytokines is produced after stimulation with
cell wall, acting on B cells through T cell mediation. Isolated peptidoglycan-p0lysaccharide

polymer fractions from cell walls have been shown to produce also these type of effects

165

mowmnmome E 99553 mam—mew? mmq .m.m «...—arm

166

moruz +. . ......... VJ
......... v + m02_
a. ..
Esme:

55223.
/ .. Emails": \/
v.

0v?— + Al 0<o

 

osmEEoE :3 es.

 

 

V. \ «ma

 

o-..—

D

Q... -n_Z +

  

167

(Novakovic et al., 1994). In staphylococcal species, the peptidoglycan seems to be a B cell
mitogen and polyclonal activator which possesses equally high activity in both mice and
humans and is equally effective in the induction of DNA synthesis and secretion of
polyclonal antibodies (Dziarski et al., 1979). Recently, De Ambrosini et al. (1996) found that
specific strains of L. casei, L. acidophilus and Lactococcus lactis contained glycerol teichoic
acids and A4a type murein on their peptidoglycan. When they tried to induced phagocytosis
in peritoneal macrophages, only the L. casei peptidoglycan showed some increased
phagocytic activity. This biological activity associated with lactic acid bacteria may not
always be due to the peptidoglycan structure.

Most of the studies of the immunoactive components of bacteria have been limited
mainly to the cell wall components, such as peptidoglycan, polysaccharides or soluble factors
produced by bacteria (Hosono et al. 1997). Experimental evidence suggests that B-glucans
enhance the host immune system by activating macrophages, NK cells and killer T cells
(Ohno et a1. 1996), their action being somehow related to their molecular weight, degree of
branching and conformation. Incubation of peritoneal macrophages with B-glucans in
presence of TNF-on or IL-6 augmented NO synthesis. It has been demonstrated that some type
of glycolipids from gram positive bacteria activated human monocyte IL-6 expression by an
identical pathway to LPS involving NF-KB and NF-IL-6 (Zhang et al., 1994). Those
glycolipids were unable to stimulate murine resident peritoneal macrophages to release TNF-
a. Proteins (Gomez-Flores et al., 1997) and teichoic acids stimulate macrophage cells to
release cytokines and NO. Loss of an acyl group or any other modification could mean the

abolition in some cases of the cytokine stimulating activity (Henderson etal., 1996).

168

Besides protein and carbohydrates present in the cell wall, cytoplasmic or heat-killed
bacteria fractions, DNA might also have immunomodulatory effects. This possibility is
supported by our findings that fractions treated with RNAse or DNAse exhibited less
activity. There is an increasing evidence that mammalian immune system can distinguish
between bacterial and vertebrate DNA, with bacterial DNA activating immune cells.
Bacterial DNA has been reported to induce murine macrophage and B cell proliferation as
well as immunoglobulin production after intraperitoneal injection. This effect has not been
observed with mammalian DNA (Krieg et al., 1995). The difference between these sources
of DNA is the presence of CpG nucleotides in bacterial DNA at a very high frequency. In
addition to this CpG suppression, the cytosine residue in CpG dinucleotides is highly
methylated in vertebrates, but not in bacteria. CpG motifs appeared to stimulate NK cells and
protect B cells against anti-IgM-induce growth inhibition. CpG motifs were nonstimulatory
if they were immobilized, so that indicates that cell uptake is required for activation.

Stacey et al. (1996) observed that DNA was taken up by RAW 264.7 macrophage
cells. Also, characteristic bacterial DNA sequences as unmethylated CpG motifs activated
a signaling cascade ending in the activation of NF-KB and inflammatory gene induction.
TNF-0t is strongly induced afier 1 h treatment with DNA (20 pg/ml). Treatment of cells with
DNA also increases NF-KB activation. This transcription factor, when activated in
macrophages is involved in the induction of TNF-0. transcription. IL-6 gene was reported
also as being induced in response to bacterial DNA (Yi et al., 1996; Pisetsky et al., 1993),
as well as IL-12 and IFN-y (Klinman et al., 1996). Foreign and synthetic RNAs (poli I:poli

C) stimulate immune responses using possibly similar mechanisms to DNA (Pisetsky 1996).

169

Nitric oxide production by macrophages was measured based in a metabolic pathway
in the cell that converts L-arginine to reactive nitrogen intermediates, via an oxidative
enzymatic pathway involving a nitric oxide synthase (NOS) (Dietert et al. 1995). Nitric oxide
produced in vivo reacts with molecular oxygen to generate other reactive oxides of nitrogen,
such as dinitrogentrioxide (N 203) and dinitrogentetroxide (N2 04). These can react with water
to form nitrite, which is further oxidized to nitrate in viva. It is known that NO is an
important mediator in the nonspecific host-defense against microbes and tumors. NO is one
of the cytotoxic agents by which activated macrophages can kill bacteria, tumor cells and a
variety of other pathogens, but also normal tissue cells during auto-immune reactions (Laskin
et al., 1994). NO is a free radical messenger that serves also as a neurotransmitter and
vasodilator. Chemically, NO is a small lipophilic molecule, so few microbes can block its
entry. While large amounts of NO may inhibit production of cytokines, smaller amounts may
induce it. An excessive production of NO plays a role in promoting the classical signs of
inflammation as well as tissue injury (Laskin et al., 1994). Recently there is a great deal of
interest focused on the possible role of NO as an immune defense molecule. Its role in host
defense still needs to be defined.

There are constitutive and inducible forms of NO (MacMicking et al., 1997). There
are three genes encoding NOS. Primary macrophages appear to express only NOS2 and
lymphocytes only rarely express any NOS. NOS2 is independent of elevated intracellular
Ca2+ and it is often called iNOS (inducible) (MacMicking et al., 1997). The pathway where
iNOS is expressed is usually associated with inflammation and infection. iNOS in mouse

macrophage can produce NO for as long as 5 days when care is taken to replenish both the

170

inductive stimuli and the L-arginine substrate. iN OS can bind to calmodulin (CaM) so tightly
that CaM copurifies as a subunit in the absence of added Ca2+. In the induction of iNOS
gene, two regulatory regions placed in the promoter site of this gene are known to be
involved. One of them contains an element for transcriptional induction of iNOS which
involves the NF-KB site. NF-KB site is the enhancer region acting as a functional regulatory
element for the maximal inducibility of iNOS promoter in LPS alone or IFN-y plus LPS
treated RAW 264.7 macrophages. The other region mediates the synergistic induction of
iNOS transcription by IFN-y for which requires IRF-l binding to interferon regulatory factor
element (IRF-E) (Kim et al., 1997). It was suggested also the presence of an enhancer region
containing another LPS responsive element with similarities in sequence to the NF-KB
region.

Numerous cytokines and microbial products, sometimes acting synergistically
stimulate iNOS expression. In mouse peritoneal macrophages, IFN-y seems to be responsible
for this stimulation (Kim et al., 1997). In RAW 264.7, LPS also stimulates iNOS expression
through induction of NF-KB. Ingestion of microorganisms generally elicit TNF-on production
and this cytokine exerts an autocrine stimulus increasing the antimicrobial action of the
macrophage, in particular by inducing the production of NO. Some other studies found that
rodent phagocytes exposed to inflammatory mediators or pathogens produce large amounts
of NO (Chesrown et al., 1994), but not human phagocytes. Eigler et al. (1995) showed that
LPS-stimulated murine RAW 264.7 cells produce both NO and TNF-or and that NOS
inhibition increased TNF-or production suggesting that endogenous NO could suppress TNF-

or production.

 

171

Taken together, these data strongly suggest that lactic acid bacteria can stimulate
macrOphages to produce cytokines. Elevated cytokine production, especially that of E-6 and
TNF-or, was well consistent with later increased NO synthesis. Further purification of these
lactic acid bacteria fractions used in our experiments may result in the isolation and
characterization of stimulatory compounds. It was reported that bacterial cell wall
components can persist in tissues for long periods of time (Dziarski, 1982). Ifactivation by
these bacterial fractions play any role in vivo, one key factor that could influence this effect
might be the requirement for prolonged exposure of leukocytes to these stimulants. Thus, cell
wall may be the best choice to provide a continuous stimulus required for activation. Under
normal conditions, these cell wall components are immediately in contact with cells from the
intestine or the peritoneum. These cells are activated either directly by bacterial products or
by cytokines secreted by peritoneal macrophages or possibly other immune cells. These
release inflammatory mediators, such as TNF-on, IL-6, which can act on the neighboring cells
and quickly amplify the cellular response. Following diffusion in the tissue, these cytokines
may activate local macrophages, fibroblasts and endothelial cells which also produce

mediators and thus contribute to the classical inflammatory reaction.

172

3.6 FUTURE STUDIES

We have found, using peritoneal cells as well as a macrophage cell line, that in vitro
exposure to lactic acid bacteria stimulates the production of IL-6, TNF-a and NO. The
mechanisms involved in the immunostimulative effect of lactic acid bacteria fractions on
macrophage cytokine production have to be determined yet but they could be similar to the
mechanisms followed by LPS. Measurement of intracellular concentration of NF -KB in
macrophages after lactic acid bacteria exposure and comparison with LPS-mediated signal
transduction in macrophages could give us an approximation of how similar these
mechanisms of activation are and the possible differences between stimulation with gram
positive and gram negative bacterial components. Bacterial extracts need further purification
and chemical identification to determine exactly which component exerts the best
immunostirnulatory effect on immune cells. Finally, to provide additional support for the
involvement of nitric oxide in the conditioned immunomodulation, we could assess the effect
of addition of a competitive inhibitor of oxidative L-arginine metabolism to cultures of cells
from animals exposed to lactic acid bacteria.

After purification and characterization of bacterial extracts, several studies could be

conducted to assess the effects of these purified fractions on in vivo and ex vivo functions.

CHAPTER 4

ENHANCED MUCOSAL AND SYSTEMIC IgA RESPONSES TO
CHOLERA TOXIN IN MICE FED YOGURT CONTAINING
LAC T OBA CILL US ACIDOPHIL US AND BIFIDOBA C TERI UM SP.

173

174

4.1 ABSTRACT

Lactic acid bacteria have been reported to have benefits for the prevention and
treatment of some forms of diarrhea and related conditions. I hypothesized that a possible
mechanism for these effects is direct stimulation of the gastrointestinal immune response.
To test this hypothesis, I determined whether administration of yogurt enhances mucosal and
systemic immunity in mice leading to increased levels of antibodies against an orally
presented immunogen, cholera toxin. Yogurts were manufactured with starter cultures
containing different species/strains of lactic acid bacteria. Mice were fed these yogurts for
three weeks during which they were also orally immunized twice with 10 pg cholera toxin.
Blood was collected at day 0 and 21 and fecal pellets were collected weekly. Upon sacrifice,
spleen and Peyer’s patches were aseptically removed, and the lymphocytes from these organs
were isolated and cultured. An enzyme-linked immunosorbent assay designed for cholera
toxin was used for titration and quantification of coproantibodies, serum antibodies and cell
supernatant antibodies against cholera toxin of the IgA and IgG isotype. Mice immunized
orally with cholera toxin responded by producing specific intestinal and serum IgA-anti
cholera toxin. Antibody responses of the IgA isotype were significantly increased in mice fed
yogurts made with starters containing Lactobacillus bulgaricus and Streptococcus
thermophilus (yogurt bacteria) supplemented with L. acidophilus and Bifidobacterium spp.
(B. bifidum, B. infantis). Yogurt manufactured with starters containing only yogurt bacteria
produced decreased IgA-anti cholera toxin when compared to the control group fed non-fat

dry milk. Although a strong response was also observed for IgG-anti cholera toxin in serum,

 

 

175

the response did not differ among groups. Thus, administration of yogurt supplemented with
L. acidophilus and Bifidobacterium spp. enhanced mucosal and systemic IgA responses to

the immunogen cholera toxin.

176

4.2 INTRODUCTION

Lactic acid bacteria have been documented to have benefits in the prevention and
treatment of some forms of diarrhea and related conditions as well as in enhancing immune
firnction (Sanders, 1993). Preservation of the integrity of the normal intestinal flora,
colonization resistance, adherence, production of antibacterial substances appear to be
important for these effects. Although the precise mechanisms of action are still unclear,
yogurt consumption has increased significantly in recent years, presumably because of these
perceived health benefits. Conventional yogurt is a fermented milk produced by classical
addition of Lactobacillus bulgaricus and Streptococcus thermophilus to milk (Trapp et al.,
1993). Based on increasing evidence that other lactic acid bacteria such as L. acidophilus and
the bifidobacteria have therapeutic properties, these species have also been added to
conventional yogurt, or alternatively used as main starter culture for production of new types
of yogurt or fermented milks (Mitsuoka, 1990; Robinson, 1991). These latter bacteria are
advantageous because while conventional yogurt bacteria have a very poor intrinsic
resistance to acid and bile (Marteau and Rambaud, 1993), L. acidophilus and
Bifidobacterium can tolerate a pH 3 and 2-8% of bile acid concentrations (Gilliland, 1989).
Both bacteria can be added in an optimal ratio before inoculation that result in 8-9 x 108
acidophilus cells/ml and 5-8 x 108 bifidobacteria cell/ml (Alm et al., 1993).

There is evidence that ingestion of lactic acid bacteria exert an immunomodulatory
effect in the gastrointestinal system of either humans or animals (Solis-Pereyra and

Lemonnier, 1993; Perdigon et al., 1986a, 1986b, 1987). The gastrointestinal system

177

possesses specialized elements that react upon exposure to antigens coming from diet
resulting in immune reactions (Pestka, 1993). These elements constitute the mucosal immune
system and essentially they are lymphoid tissue containing a full array of immune cells
necessary for induction of an immune response, that is, B, T, macrophage and accessory
cells. The tissues that represent the gut mucosal immunity are Peyer’s patches, mesenteric
lymph nodes, lamina propria and intraepithelial lymphocytes (J aneway and Travers, 1994).
The systemic immune compartment is formed by all tissues involved in defense of the
internal environment from invading microorganisms, and consists of spleen, thymus, bone
marrow and lymph nodes throughout the body. The mucosal and systemic immune
compartments, can overlap in some of their specific activities. This connection is mediated
via circulation of blood and lymph, where B and T cells can migrate from one compartment
to the other. The mechanisms by which lactic acid bacteria might activate the mucosal and
systemic compartments are still unknown.

One potent antigen that stimulates both the mucosal and systemic compartments is
cholera toxin (CT) (Ogra et al., 1994). CT induces an intestinal secretory IgA response to
itself and to additional co-administered antigens (Epple at al., 1997; Robbins et al., 1989).
A strong local IgA response has been reported in the small intestine as demonstrated by the
detection of antibodies against CT in intestinal secretions (Snider, 1995). A strong serum IgG
antibody response is also induced by CT immunization, and the cells secreting this isotype
appeared to be located in Peyer’s patches, spleen and lamina propria. CT is known to act as
an intestinal adjuvant for many common proteins, viruses or bacterial polysaccharides,

creating a memory response with strong IgA antibody response to antigens (Ogra et al.,

178

1994), and maybe signaling by secretion of certain cytokines.

In this study, the effect of yogurt ingestion on gastrointestinal immune system of the
mouse was assessed using CT immunization as a model. Feeding individuals with different
types of yogurt made with L. bulgaricus, S. thermophilus with or without L. acidophilus and
Bifidobacterium spp. and using CT through oral immunizations as a protein antigen, 1 was
able to demonstrate the adjuvant activity of yogurt containing L. acidophilus and
Bifidobacterium spp. The results indicated that ingestion of yogurts supplemented with L.
acidophilus and Bifidobacterium spp. enhanced the mucosal and systemic IgA response to

CT.

179

4.3 MATERIAL AND METHODS

4.3.1 Yogurt preparation

Four commercial yogurt starter cultures were obtained by Sanofi Bio-Industries
(Waukesha, WI), Chr. Hansen’s Laboratories Inc. (Milwaukee, WI) and Rhone Poulenc Inc.
(Madison, WI). Table 4.1 summarizes the type of culture present in each starter and their
source. The four yogurts were made using pasteurized 12 % (w/v) non-fat dry milk (NF DM)
(Michigan Milk Producers Association, Ovid, Ml) according to the instruction of the culture
suppliers. NFDM was inoculated with the starter and mixed. The inoculated mixtures were
aliquoted into 50 ml sterile conical polyethylene centrifuge tubes and incubated for 6-8 h at
37°C developing a typical yogurt consistency. Yogurts were rapidly cooled and stored at 4°C
and ingested within 21 d. Numbers of total aerobes, streptococci and bifidobacteria were

determined throughout storage.

4.3.2 Enumeration of yogurt bacteria

Total aerobic cells, streptococci and bifidobacteria counts were performed at 0, 1, 2,
3 and 4 wks of storage to determine the viability of the cultures. De Mann-Rogosa-Sharpe
medium for lactobacilli (MRS) (DeMan et al., 1960) + 5 % (w/v) lactose (MRSL) agar
plates, modified ST agar (Lee’s agar) (Lee etal., 1974) and neomycin sulfate, paromomycin
sulfate, nalidixic acid, lithium chloride (NPNL) agar (Teraguchi et al., 1978; Rasic and Sad,
1987) were used for total aerobic, Streptococcus and Bifidobacterium counts, respectively.

After the appropriate dilutions were made, samples were plated using the media mentioned

180

 

E .5882
one—somoaonm

E 63323
as 85E .5

E £5333
85559:-05 mom—em

:5 6:53:53
8.5525365 macaw

£8? .5
£3858 .5
5§~Sw~§ q
5335.6:th M.

5855 .5
35858 .5
SEEMNE dam 5.53825th Q

533302225 .amm 5EE>QG5 a.

dam Eatmfigohmbm

35588 .5
SEEMBQ dam 5.383353 a

5SERQE§$ dmm 53.26335 ..m.

5§~§m3¢ .Qmm 5385353 u
53.230583 .mmm 53.8335 ..m.

85m... 55 830 58» do

55% 3.5% SE

8330 tame» 885 mswmfim

83—8 tame» 885 80-855

 

uozmmsm

5:803 Rom 0:84

9:53 tamo>

 

.8523 Steam tame.» 380888 E «583 Boa 382 .«o 8353800 ._.v mafia.

181

above and incubated for 48 h at 37°C, aerobically for total aerobic and Streptococcus, and
anaerobically using anaerobe jars and an anaerobic Gas Pak® (Becton Dickinson Co.,
Cockeysville, MD) system for Bifidobacterium. The colonies were counted using a Quebec

colony counter (Fisher Scientific, Pittsburgh, PA.).

4.3.3 Animal model and diet

Female B6C3F1 mice (C57BL/6 female x C3H/HeN male), 8 week old, were
obtained from Charles River Labs. (Raleigh, NC). Ten mice per experimental group were
employed. All animal handling was conducted in strict accordance with regulations
established by the National Institutes for Health. Experiments were designed to minimize
numbers of animals required to adequately test the proposed hypothesis and approved by
Michigan State University Laboratory Animal Research committee. Mice were housed (5 per
cage) in a windowless room at 25-27°C with a 12:12 hr light/dark cycle and a negative-
pressure ventilated area, in protected-environment cages (Nalgene, Rochester, NY) that
include a transparent polycarbonate body with filter cover and stainless-steel wire lid.
Distilled water was provided ad libidum and changed every 3 days.

Mice were acclimatized to housing and fed nutritionally complete modified
semipurified diet as described by the American Institute of Nutrition (AIN -93 G) (Reeves et
al., 1993) for at least one wk prior to starting experiments. The study lasted 21 d. Yogurt or
NFDM control were mixed (1:1) with AIN-93C} (ICN Nutritional Biochemical, Cleveland,
OH). Yogurt and control diets were prepared fresh and provided daily during the experiment

in clean powder feeders with stainless steel grids and lids to reduce spilling. Figure 4.1

182

.5on 820:0 8 8:858 czsngwogfig co cosmowfi tame» mo Scuba wfimmommm 8m cwfiow Ecofitomxm 4.9 «...—urn

183

95:3 :3
can :otmfifiuoh

 

 

 

cotmumcsaaz
“worm

 

 

cotoozoo 830m 22536255 vacuum cotoozoo 853m $93 9 028
28:00:00 38m cozoozoo 38m 538:3 33m :ozoozoo mooom “882:me
3E0? Emma? 2903 Emma? 3825mm
5 ............................. 3 ........................... n ..................... o .......................... n-

:23 :3 23% 8?} 23:8 6% xi + 08.2?

m><Q

0 3-235. 5:38:03

184

summarizes feeding and treatment schedules during the experimental period. Weight changes
were recorded and fecal samples were collected weekly. Fecal pellets were placed in sterile

plastic containers, immediately processed and stored at - 80°C until analyzed.

4.3.4 CT immunization

Mice were deprived of food for 2 h before oral immunization. Just prior to
immunization, they were gavaged with 0.5 ml of a solution consisting of 8 parts Hanks’
balanced salt solution (Sigma Chemical Co., St. Louis, MO) and 2 parts 7.5 % sodium
bicarbonate to neutralize stomach acidity. Thirty minutes later, 10 ug of CT (Sigma
Chemical Co., St. Louis, Mo) was administered in 0.25 ml of filter-sterilized phosphate-
buffered saline (PBS). Groups of mice were immunized on days 0 and 14. Five experimental
groups were used in the study (control, Ultra-Gro Direct, Sbifidus Direct, PY-3 Redi-Set, and
DPL Yogurt Quick Start ABY-2C). Food and water were restored immediately after

immunization.

4.3.5 Fecal pellet and serum preparation

Fecal samples were prepared as described previously (de Vos and Dick, 1991).
Briefly, feces were picked, aseptically weighed and placed into centrifuge tubes. Ten ml
PBS per gram of feces (v/w) were added, and the mixture was incubated for 15 min at room
temperature. Samples were mixed by vortexing until suspended, left to settle for 15 min,
mixed again, and then centrifuged at 22,000 x g for 10 min. The supernatant was removed

and stored at - 80°C for immunoglobulin measurement.

185

Blood was obtained from anesthetized mice fiom the retroorbital plexus. Serum was
obtained afier overnight incubation at 4°C and centrifugation at 1,000 x g for 15 min. Serum
samples were aliquoted, stored at - 80°C prior to monitoring for IgA and IgG anti-cholera

toxin antibody responses.

4.3.6 Lymphocyte culture

Mice were sacrificed one week after the second and last immunization with cholera
toxin by cervical dislocation under gentle anesthesia. Spleen and Peyer’s patch cultures were
chosen to represent systemic and mucosal immune responses, respectively.

Peyer’s patches were removed aseptically, placed in a small Petri dish containing 5
ml RPMI 1460 medium (Sigma chemical Co., St. Louis, MO) supplemented with 10% (v/v)
fetal bovine serum (FBS) (Gibco Laboratories, Chagrin Falls, IL), 2-mercaptoethanol (50
uM), non-essential amino acids (1 mM) (Gibco BRL, Life Technologies, Grand Island, NY),
sodium pyruvate (100 mM) (Gibco), 100 U/ml penicillin and 100 ug/ml streptomycin
(Sigma), and teased using two sterile cover slides. The cell suspension was passed through
a 70 um nylon membrane that was fixed on top of a 15 ml sterile centrifiige tube. Five ml of
fresh RPMI 1640 were added to wash the plate and membrane. Cells were centrifuged at 450
x g for 8-10 min, resuspended in 2 ml of fresh medium and counted using 0.4 % trypan blue
stain (Sigma) and a hemacytometer. Cell suspensions were kept on ice at all times.

Spleens were removed aseptically, placed in a Petri dish containing 10 ml RPMI 1460
medium as above, teased thoroughly and cell suspensions passed through a sterile 85 mesh

screen to remove tissue debris. Lymphocytes were then placed into a 50 ml sterile centrifuge

186

tube, left to settle for 5 min on ice and then transferred to another sterile tube. The
suspension was centrifuged at 450 x g for 8 - 10 min and supernatant was discarded.
Erythrocytes were lysed for 3 min at room temperature in 5 ml of a buffer containing 9 parts
of 0.16 M ammonium chloride plus 1 part 0.17 M TRIS buffer (pH 7.2). Fresh RPMI 1640
(10 ml) was added, cells were mixed, centrifuged at 450 x g for 10 min, resuspended in 20
ml of fresh medium and counted. Cell were kept on ice at all times.

Peyer’s patch and spleen cells (1 x 105/ml) were cultured in supplemented RPMI
1640 medium in flat-bottomed 48 well (1 m1) tissue culture plates (Fisher Scientific Co.,
Corning, NY) at 37°C in a 7% CO2 humidified incubator. Duplicate cultures were stimulated
with or without 20 ug/ml lipopolysaccharide (LPS) from Salmonella typhimurium (Sigma
Chemical Co., St. Louis, MO). Supernatant was collected at 7 days and stored at - 80°C until

analysis.

4.3.7 Enzyme-linked immunosorbent assay (ELISA)

Antibody titers in serum, fecal extracts and cell culture supematants were determined
by an ELISA as described by Jackson et al. (Jackson et al., 1993) with some modifications.
Immunolon IV Removawell microtiter strip wells (Dynatech Laboratories Inc., Chantilly,
VA) were coated with 100 pl of CT (5 ug/ml) in 0.1 M carbonate buffer pH 9.6. Plates were
incubated overnight at 4°C in a humid atmosphere and washed three times with phosphate-
buffered saline (PBS) plus Tween 80 (PBS-T). Plates were blocked with 200 pl of 1% bovine
serum albumin (BSA) in PBS for an hour at 37°C. The plates were washed again three times

with PBS-T. Fifiy microliter aliquots of serial dilutions of serum (1:50 to 1:6400), fecal

187

extracts (1:2 to 1:320) or supematants in 1% BSA-PBS were added in duplicate, and the
plates were incubated for 75 minutes at 37°C. Preirnmune serum and fecal extracts at similar
dilutions were used on the same plates as controls. Plates were washed three times with PBS-
T. The secondary antibody consisted of 50 pl/well of horseradish peroxidase conjugated to
goat IgG fraction to anti—mouse IgA-0t chain (26 pg/ml) or IgG-y chain (40 pg/ml) (Cappel,
ICN Pharmaceuticals, Inc., Aurora, OH) in 1% BSA-PBS. Plates were then incubated at 37°C
for 75 minutes and then washed six times with PBS-T. Bound peroxidase was determined
by adding 100 pl of 2,2'-azino—bis(3-ethylbenz—thiazoline-6-sulfonic acid) (ABTS) substrate
[0.4mM ABTS,50 mM citrate buffer (pH 4.0), and 1.2 mM hydrogen peroxide] as described
previously by Pestka et al. (1980). Absorbance was measured at 405 nm on a Vmax kinetic
microplate reader (Molecular Devices, Menlo Park, CA). End-point titers were expressed as
the last dilution which gave an OD at 405 nm of 2 0.2 OD units above the OD of preirnmune
serum and fecal extract.

CT-specific IgA and IgG were quantitated gravimetrically in serum, fecal extracts
and cell culture supematants as described by Pestka et al. (1990). Concurrently with the
ELISA described above, 50 p1 of serially diluted standard mouse reference serum were added
to anti-1g wells that flanked CT wells for use in isotype reference curves. F ifiy pl of serum
or fecal sample (1:50 and 1:3 dilution respectively) were added to the wells coated with CT.
Plates were then incubated and washed as described for titration above. Immunoglobulins
bound to CT were assigned concentrations (ng/ml) based on comparison of absorbance to

values obtained in isotype reference curves that were run in parallel.

188

4.3.8 Statistical Analysis
Data were analyzed by Student’s t test using SigmaStat Statistical Analysis System
(Jandel Scientific, San Rafael, CA). A p value of less than 0.05 was considered statistically

significant.

189

4.4 RESULTS

4.4.1 Viability of the refrigerated yogurt during storage

When viability of the yogurt was assessed over a four week period (Figure 4.2), total
aerobic and Streptococcus counts ranged from 1 x 109 to 5 x 10 8 CFU/g of yogurt. Numbers
of viable bifidobacteria declined at a more rapid rate. For DPL yogurt Quick start ABY-2C
and PY-3 Redi-Set yogurt, Bifidobacterium counts declined from 1.2 x 108 to 2.3 x 106
CFU/g and from 6.9 x 107 to 3.7 x 106 CFU/g, respectively, over a 3 week period. For
Sbifidus Direct yogurt, counts declined from 2.5 x 108 to 1.1 x 107 CFU/g. Nevertheless,
viability of lactic acid bacteria and bifidobacteria in yogurt remained above 1 x 106 CF U/g,

which is within normal values reported in commercial milk or yogurt (Shin, 1997).

4.4.2 Body weight and feed intake

All animals remained healthy during the feeding trial. Diarrhea was not observed in
any treatment or control group. No differences in initial mean body weight among the groups
(20.5 :t: 0.4 g) were detected. After 3 weeks, body weights of the treatment groups were
comparable to those of control group (Figure 4.3). No statistically significant differences in
body weights were observed among the treatment groups and/or control group at any time

during the study.

4.4.3 Specific Immunoglobulin production

Intestinal and serum anti-CT responses were measured in mice immunized orally with

190

Figure 4.2. Bacterial counts. Total aerobic, Streptococci, and Bifidobacteria counts in
yogurts manufactured for feeding trial. UG= Ultra-Gro Direct yogurt culture, SBIF=
Sbifidus Direct yogurt culture, PY3= PY-3 Redi-Set yogurt, ABY= DPL yogurt Quick
Start ABY-2C.

 

191

 

 

 

 

 

 

 

 

 

 

 

 

to”? TOTAL couu‘r _O_—°-gg...
—a—PY3
—-x-—ABY
g 10 -
LL
0
g 10‘-
‘0’ 1 | I I l I
0 1 2 3 4
Storage time (wks) ,
—O-—UG
10m STREPTO COUNT —cr—SBIF
—a—PY3
+ABY
.9: 10'--J
E
o
E ,,._
107 l I 1 I I I
o 1 2 a 4
Storage time (wks)
",1 BIFIDO COUNT —a—seu=l
—A—PY3
—-)(—-ABY
g: 10" «-
:>
ll.
0
8’
.l 10""1
10' I 1 I I I 1
o 1 2 a 4

Storage time (wks)

192

.E.m.m “m 538 05 555252 San 458598
2: mo 96 05 En: 5283 @8882: $3 Ego? E owgno 28 RE wficofl 05 mo qufiwon 05 an 3563 203 032
ASHE x3 m S.“ «€883wa USN «£823 Eon 28a— 53, 038 tame» .8 via 8&5 wow 8:: Sm 5% Emma? .m.v charm

193

 

 

 

SW

 

 

 

m\\\\\\\\\\\\\\\\\\\\\\

 

 

 

+—\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

 

 

 

rn\\\\\\\\\\\\\\\\\\S

 

 

 

PW

 

 

 

 

(8) ["93 ILIBIQM

Py-3 Redi DPL Quick

Sbifidus

Ultra-Gro

Control

194

10 pg CT twice (day 0 and 14) and simultaneously fed with the experimental yogurts. To
measure intestinal response, fecal pellets were collected from the various experimental
groups and extracts were tested by ELISA for specific IgA anti-CT (Figure 4.4). Endpoint
titers for CT from all groups of mice before immunization and yogurt feeding showed very
low antibody levels against CT (Figure 4.4A). Anti-CT IgA levels slightly increased one
week after control or treatment mice received the first CT dose, but there were no significant
differences among the groups (Figure 4.48). Levels appeared to decrease during the second
week for the four treatment groups while the control group presented a slightly higher level
(Figure 4.4C). One week after a second dose of CT was given, endpoint anti-CT IgA titers
differed markedly among all groups (Figure 4.4D). Based on optical density measurements,
mice fed yogurt made with Sbifidus Direct yogurt culture and PY-3 Redi-Set yogurt culture
(containing conventional yogurt bacteria supplemented with L. acidophilus and
Bifidobacterium spp.) exhibited significantly higher titers than the control group (p s 0.05),
whereas the group fed yogurt made with Ultra-Gro Direct yogurt culture (i.e conventional
yogurt bacteria) exhibited a decreased specific IgA levels respect to the control diet.
Gravimetric estimates of CT—specific IgA calculated were significantly higher in groups fed
Sbifidus Direct yogurt culture, PY-3 Redi-Set yogurt culture, and DPL yogurt Quick Start
ABY-2C when compared with the control (Table 4.2). Interestingly, group Ultra-Gro Direct
yogurt culture exhibited significantly lower specific IgA than the corresponding control
group and also with respect to the rest of yogurt treatment groups.

Mice immunized orally with CT also responded with specific serum antibody

responses of IgA and IgG isotype. As with coproantibodies, levels of anti-CT IgA in serum

195

.3935 macaw 35:8 Sow Eobfifi bassommawfi ... .2.m.m H :38 05 “55.83“ 9mm
.5on E205 ram-<w~ 8&0on 8m <m5m .3 BSA—mam 225 Sombxm flame» wage.“ 5MB m 23 5x8 820:0 53> gages;
9885 Sam x3 # HQ 305 wage.“ 98 5on 8230 5MB cougfig Ham 8% 5x3 N no shame» wfiuoom Ea 5x3 8085
SE comm—SEE “Em Saw x3 fl um .wfiwoom no one 5% u< 68365 8:5 2: E 995% m .oHHE 3on some E 3308
some 80b 360:8 80>» 53:3 18$ 5on .Amomconuswoaoov .8353 Hon.“ 5 5x8 Eo_o:o-w=m-<wH 2&0on iv charm

196

~-o #

_ FP—__—

_

—

72

.hhb___

cosine 5893:8598 wo—

h

—

oo—

__LF——

1: l:

P_.___— _ -______ _ _

P pp.»—

o3

 

 

 

 

UN->m< Em x080 tame» 15D 3. . :
Ewaeméom ES IT

up?» see 3335 II

tame» 82:. 80.8.5 - ml

3980 lol

 

 

 

od

Kirsueq [condo

o.—

m4

197

were significantly higher in yogurt treatment groups than in control group except again for
group Ultra-Gro Direct yogurt culture which lacked L. acidophilus and Bifidobacterium spp.
(Figure 4.5). Gravimetric estimates of these specific antibodies were higher in serum than
in fecal samples as expected (Table 4.2). Mice also elicited a strong response for specific
anti-CT IgG (Figure 4.6) but titers were lower for IgG than for IgA and there were no
significant differences between group of mice fed yogurt and control diet except for Ultra-
Gro Direct yogurt culture (i.e., conventional yogurt bacteria) which showed a significantly
lower specific IgG levels than control group and other yogurt treatments (Table 4.2).
Together these results showed that this protocol of oral immunization was effective in
eliciting CT-antibody responses and suggested that yogurt made with L. acidophilus and

Bifidobacterium spp. enhanced mucosal and systemic IgA responses.

4.4.4 Specific immunoglobulin production in cell cultures

The effects of yogurt feeding on immunoglobulin production were determined in
cultures prepared from Peyer’s patches and spleen which are representative mucosal and
systemic lymphoid tissues, respectively. Peyer’s patch and spleen cells (1 x 105/ml) were
stimulated with or without 20 pg/ml lipopolysaccharide (LPS) from Salmonella typhimurium
and supematants collected and analyzed after 7 days for specific IgA and IgG. Specific IgA
levels were similar in Peyer’s patch cell cultures for all treatments with respect to the control.
A trend towards increased IgA anti-CT levels was observed in spleen cell cultures stimulated
with LPS from groups treated with yogurt as compared to control group but these effects

were not significant (data not shown). Similar results were obtained when analyzing for CT-

198

specific IgG.

199

.3933 95% 600000 88.0. Empobmu bassomfiwmm ... .2 .o u ad 8805 0:38:85 .E.m.m ”m :88 05 088080 Sam .5on
820:0-35-<3 2.50% Sm <m5m 3 0083050. 083 838% 83805 8205 5 £080. 0.8—0:0 68-4553 058% .m6 9:.me

200

v.2

c2350 83% w2

”OH TO—

_ __———__H p _—_b___

 

 

UN->m< :8m 0330 tame» Ana .0
E?» sméom SE i9:

tam?» 88% msumfim -Idl

tame» 82% 95:35 la]

65:00 |®|

 

<3

o.o

I o.—

 

 

mg

mysuaa [condo

201

$933 @3on 39.8 Bob “55wa 305$?me ... .2 .o n Dd 88% 0:58.505 .2.m.m H :88 05 882mg Sun .508“
8205-35-me #:0on 8m (dam an Bum—mam 95>» moaaam .8383 823 E 5x8 820:0 :5».-me 2&0on .96 0.53%

202

Yog

cows—:8 83% mo“

¢o~ vofi To_

_ _—__h—__ — ____—-—_ L r__———~

 

 

UN->m< 88m 880 tame» .EQ ........0
Ema 3%va SE i9:
tame» 82% mschm - [all

”I’MQI/ tam?» 88% 80-935 I ml

/ 3.980 lml

 

0
IIIIIII
0‘.
c
.0
o
.u

//..
/

4

cu—

oav

 

 

m;

[Cusuaq [condo

E H z 08 z? + a8 8% $3 888: 388 3 882 2888 $8388 .85 v a. ..

God v 3 3888: tame» 8:8 3 88mm: Edouawa bfioumufim o

.2 8888 Room 8m .3: 828 88 0E8 888

888,88 388 c8888 a 8 88888 3 88888 882888 oflooaméowzau 88 888 8:25 3.
AN 8% 8 88.83 083 moon.“ v8 88 .8 888m

.wumcoom raw?» .8 3 23 g 99% 8 88.. .888 .8 w: 2 :85 8858 $38 083 38 A: .8 885 g

 

203

 

 

md H wd ...VN H 0.3 3.0 H _.m Um->m<
No H ~1— *m.m_ H ”K Lao H cs %3_ murm
No H We L8 H m.mm *de 0v 885m
0.56 H Yo o; H #3 10.0 H m: 808.5
_.o H Z No H w.mm _.o H Va 8580
QSEE naaom mamas ..EEom 3E3 amouom 808805 tamer
888 8088-38 0E 88 «888-38 aw~

 

 

“888888 8o» :8 .8 £83 88.. 8am 888Eo88m 058% 88 888 .8 m88mo 08830 .aé mama?

204

4.5 DISCUSSION

This study demonstrated that administration of yogurt differentially affected mucosal
and systemic IgA responses to CT in viva. Specifically, yogurt starter cultures containing L.
acidophilus and any type of Bifidobacterium had an effect on mice that was reflected by
specific IgA anti-CT production as compared to conventional yogurt starter cultures
containing only L. bulgaricus and S. thermophilus. Thus, the bacteria present in the starter
culture seemed to be critically important for immunomodulation. This suggests that several
factors specifically related to a bacteria species might play a role in the extent to which
yogurt alters immune function.

To exert a maximal influence on gastrointestinal function and to be able to act as
probiotics, lactic acid bacteria may need to be present in high number in fermented milks,
and survive the digestive process (Klaenhammer, 1982). In this experiment, the viability of
bacteria in all yogurts remained high and within the normal values for commercial fermented
milks (Shin, 1997). This is important because the ability of L. bulgaricus and S. thermophilus
to persist in the gut after administration in mice is extremely doubtful and because there is
a general consensus that continuous ingestion is needed to maintain colonization in animal
models (Fuller, 1991). The capacity to tolerate low pH and high bile concentrations is
advantageous for the survival of L. acidophilus and Bifidobacterium bifz‘dum inside the gut
of animals (Alm, 1991). Since a mouse model was used, it is possible that starter cultures
containing L. acidophilus and Bifidobacterium spp. were able to persist in the mouse gut

while the traditional yogurt bacteria, L. bulgaricus and S. thermophilus, were washed out

205

faster.

B6C3F1 mice were chosen for this study because of increased hardiness and
longevity characteristics of heterosis and because the wider genetic diversity is more
characteristic of human p0pulations (Festing, 1979). This hardiness is especially critical for
conducting feeding studies and the use of genetically identical mice minimizes variability
encountered in immunological experiments. No diarrhea, food refusal or discomfort were
observed at any time during the feeding trial. We have previously shown that the average
feed intake for yogurt produced with these same starters ranged from 5.4 to 5.8 g/mouse/day
(Ha, unpublished data).

NF DM or yogurt were provided with a semipurified powder diet (AlN-93 G) at a 1 :1
ratio and the gain in body weight was not significantly different between control and
treatment groups. Puri et al. (1996) have similarly found no difference in the grth rate of
mice fed classical yogurt (L. bulgaricus and S. thermophilus) or milk for 4 weeks. This
yogurt diet was also prepared by mixing with powdered pellets at a 1 :1 ratio. In contrast with
these latter findings, better growth has been observed in rats fed freeze-dried yogurt
(Hitchins et al, 1983) and classical yogurt (McDonough et al., 1985) when compared to milk
fed rats. This discrepancy among studies might be explained because a different animal
model was employed and because my study used yogurt mixed with a powdered food while
some others (Hitchins etal., 1983; McDonough et al., 1985) fed only unsupplemented yogurt
or milk.

The observation that fecal and serum IgA anti-CT levels were higher in mice fed

yogurts with bifidobacteria and L. acidophilus when compared to the control-fed group is

206

likely to be related to enhanced stimulation of the gut mucosal immune system. In support
of this contention, feeding mice with L. casei increases IgA to enteropathogens (Perdigon et
al., 1990) and induces a protective effect against E. coli, L. monocytogenes (Nader de
Macias et al., 1993; Nomoto et al., 1985) and Mycobacterium bovis (Saito, 1988). Other
studies have reported that feeding mice fermented milks containing L. casei (Paubert-Braquet
et al., 1995), L. acidophilus and/or yogurt cultures (L. bulgaricus and S. thermophilus)
(Perdigon eta1., 1991b) exerts a protective effect against intestinal pathogens and increases
production of immunoglobulins and activated lymphoid follicles (Bourlioux, 1986). Also,
serum IgA levels after challenging with Salmonella were significantly higher in mice fed
classical yogurt when compared to milk-fed controls (Puri et al., 1996). In a related study,
mice fed yogurt fermented by L. bulgaricus and S. thermophilus, the identical heated
product and a L. casei fermented milk and vaccinated with partially purified CT (three times
intraperitoneally [40-100-200 ug/mouse] and once orally [100 ug/mouseD at weekly
intervals (Portier et al., 1993). When sera were analyzed by vibriocidal test for specific
antibodies against two highly correlated serotypes (Ogawa strain and Inaba strain) of V.
cholerae, significant differences to Ogawa but not to Inaba serotype were found. Thus, in
some cases it seems that classic yogurt has adjuvant properties and the capacity to stimulate
the systemic immune system. This contrasts with the findings in this study where yogurt
made with L. bulgaricus and S. thermophilus had no effect for IgA anti-CT at the systemic
level. This contradiction could be attributed to the different extract of toxin used, the
different dose or immunization protocol followed and/or the different assay used to measured

specific antibodies. It has been reported that mice orally immunized three times at weekly

207

intervals with 10 ug CT produced maximum IgA anti-CT responses in fecal samples and IgG
anti-CT in serum on day 21 (Xu-Amano et al., 1993). Knowing of the capacity of CT to act
as a potent mucosal adjuvant for other antigens or vaccines I reduced the immunization
schedule and administered just two doses. This was because the goal was not to achieve a
peak antibody response to CT alone but to determine whether yogurt treatment facilitates in
any way the induction of antibody responses to the co-administered antigen given orally.
The Peyer’s patches are the central focus for the induction of T and B-cell responses
following an oral immunization (Pestka 1993). These organs lie below a specialized layer
of epithelial cells called M cells (Figure 4.7). Once the antigen has traversed the M cells,
antigen presenting cells (APC) within the Peyer’s patch can take up the antigen and present
it to nearby T cells. There are two different clones of T cells: T-helper cells can be classified
as Thl (producing IL-2, IL-3 and IFN-y) and Th2 (producing IL-6, IL-4, ILS) (Pascual et al.,
1996). Some of these cytokines are able to activate B cells and mediate proliferation,
switching and differentiation of these cells to become committed to secrete IgA (Shanahan,
1994). B cells within the follicles expressing IgM or IgD on their surface upon stimulation
can proliferate and differentiate into lymphoblasts expressing IgA on their surface. B cell
lymphocytes pass into the efferent lymphatics to the mesenteric lymph nodes and from there,
they can enter the systemic circulation via thoracic duct and enter distant effector sites as
lamina propria of the intestine. At these effector sites B cells proliferate and mature into IgA
plasma cells in response to certain signaling by cytokines produced by T cells and
macrophages. Plasma cells produce polymeric IgA that is then secreted across the epithelial

cell into the lumen (Pestka, 1993). IgA secreted by B cells in the lamina propria can be

208

833825: 3.8 wars/0:0.“ uncommon 258:: 353850536 6.9 PEER

209

 

333: All 53.5 coca All «can . «to: FEE».
.832: 2020:... 33503.2
.550
:00 NEON. . 330 m
l@@ i...“ ll
<u % 1% <2 6 a.
a _ a
< _ V4< _ a. 9.58583
925.85 V4 .
.i
#2....- uca viz... 3 2.2.2.53 . . . .
3.3203 3 5.3.3953 . .
/V < 39.3an «=3 05.650

0 Q. no

 

 

 

 

< 5.25 4 o ‘ m3...

:oE:
_ 5x3 9.0.2.935 <.m_ » 220:0

3:332.

Ed mtm MOhUmuum an: mtm ZO_._.UDDZ_

210

released to the intestine following later the bowel content or goes to the general circulation
raising the serum IgA levels.

CT is composed of subunit A (posttranslationally cleavaged into toxigenic A1 and
A2 peptides) and subunit B. CTB is a homopentarner that serves as a carrier for the CTA, by
binding to monosialoganglioside GMl present in the intestinal cells. A conformational
change in this GMl allows A1 peptide to penetrate the cell. The mechanism is unknown but
it could involve some type of endocytosis (Ogra et al., 1994) or direct translocation of A]
component through the lipid bilayer (Fishman, 1990). Intemalized CT drives B cells towards
IgA-committed precursors. CT can change the isotype pattern from IgM to IgG and IgA in
Peyer’s patch B cells primed for an unrelated hapten afier antigen-dependent clonal
expansion in vitro (Elson and Dertzbaugh, 1994). CT might affect Peyer’s patch B cells
either directly by contact or indirectly by activation of macrophages and T cells, which might
produce cytokines that induce differentiation of B cells and immunoglobulin switching.
Stimulated B cells in Peyer’s patches migrate to lamina propria to populate IgA in that organ,
or travel through blood to spleen where they become IgA or G secreting plasma cells.

The mechanisms by which Bifidobacterium and L. acidophilus yogurt stimulate the
gut immune system are unclear. One possible mechanism by which certain yogurts might be
able to alter the immune response on GALT could involve the generation of a local signal
at the intestinal mucosal surface either by the bacteria strain present in yogurt, bacterial
moieties with immunomodulatory activity, substances produced during fermentation, or
translocation of these substances or microorganisms through the mucosal barrier to another

host compartment.

211

Some bacterial components with immunomodulatory activities seem to be
lipoteichoic acids, endotoxic lipopolysaccharide and peptidoglycans which are species and
strain specific. Related to these bacterial fractions, other studies conducted by this laboratory
showed that whole non-viable lactic acid bacteria cells (Marin et al., 1997a, 1997b, 1997c)
and their cell wall and cytoplasmic fractions stimulated macrophages in vitro to release TNF-
or, IL-6 and nitric oxide (see chapters 2 and 3). It has been suggested that L. acidophilus or
some of its microbial components are able to cross the epithelial membrane of human
intestinal cells in culture, but the mechanism is still unknown (Coconnier et al., 1997).

Gangliosides are glycosphingolipids which contain sialic acid and neuraminic acid
(important component of bacterial cell walls) (Pelczar et al., 1993). They are found in apical
membranes of all intestinal epithelial cells. Most pathogens, including Helicobacter pylori,
are able to attach themselves to epithelial cell surfaces through specific structures namely
GM3 gangliosides, and damage the mucus coat by protease activity (Slomiany and Slomiany,
1992). L. casei was also found to bind in the intestinal tract to some specific
glycosphingolipids possessing short sugar chains and galactosyl moiety (Yamamoto et al.,
1996). Yohe and Ryan (1986) observed that both ganglioside composition and sialic acid
composition of macrophages are profoundly altered with presence of gram negative
endotoxin. Present also in milk and other dairy products, they seem able to inhibit
enterotoxin activity from Vibrio cholerae and E. coli, in vitro and in vivo in human studies
(Laegreid et al., 1986). Conventional yogurt might present a higher amount of these
gangliosides which would exert an inhibitory effect on the ability of CT to be internalized.

Regarding the possible internalization of microorganisms or fermentation products

212

through the mucosal barrier to another host compartment, it is known that M cells present
in the intestine are able to translocate bacteria, irnmunogens from the intestinal lumen to
underlying lymphoid tissue (Pestka, 1993). Some studies showed that the proportion of M
cells and lymphocytes was significantly higher in mice fed with live L. bulgaricus and S.
thermophilus as compared to controls (Muscettola et al., 1994). L. acidophilus and
Bifidobacterium spp. might either affect the proportion and permeability of M cells which
translocate CT from intestinal lumen to internal gut lymphoid tissues more efficiently than
some other species of lactic acid bacteria, or penetrate the epithelial barrier of the intestines
themselves providing an antigenic stimulus to the local immune mechanisms of the gut,
activating immune cells to function as APC.

The potential signal produced by any of the above events might initiate a cascade
involving (1) activation of gut-immune cells (macrophages, T and B cells), (2) cytokine
production and (3) eventually translation of this effect into proliferation of B cells and their
differentiation to plasma cells committed to secrete IgA against CT, as it was observed.

Macrophage activation is one of the general mechanisms of adj uvant action as well
as generation of inflamation, stimulation of B cell isotype, switching, proliferation,
differentiation and stimulation of increased cytokine production. Macrophages are normally
in a resting state and a variety of stimuli can activate them, increasing their phagocytic
activity, secreting cytokines (IL-6, IL-1, TNF-a), increasing microbicidal activity, secretion
of inflammatory mediators and ability to activate T cells. In addition, activated macrophages
also function as APC.

Lactic acid bacteria may activate macrophages directly. Administration of fermented

 

213

milks with lactic acid bacteria (L. bulgaricus and S. thermophilus, L. casei, L. acidophilus,
Bifidobacterium spp.) apparently enhances immune response in animal studies by activating
macrophages and lymphocytes (Perdigon et al., 1988). Notably, it has been observed that
classical yogurt induced less phagocytic activity in macrophages than L. acidophilus and
Bifidobacterium fermented milks (Moineau et al., 1989; Moineau and Goulet, 1991). Oral
administration of L. acidophilus strain La 1 and Bifidobacterium bifidum strain Bb 12 at
doses of 7x1010 CPU/day (Schiffrin et al., 1997) to humans increases phagocytic activity in
blood. Activated macrophages play an important role in the resistance of the host to
infections and tumors (Perdigon et al., 1986a, 1986b, 1987). If a mixture of L. casei and L.
acidophilus is administered orally an increased lymphocytic activity and in vitro peritoneal
macrophage phagocytic activity is found (Perdigon et al., 1986a). Macrophages and
lymphocytes are also activated with administration of L. acidophilus and/or S. thermophilus
(Perdigon et al., 1987). When administered to animals in a yogurt form, these bacteria seem
to increase the number of spleen germinal centers, T and B lymphocytes and decrease
preexisting enterobacterial infections (De Simone et al., 1987a, 1988a, 1988b, 1992). In
humans fed yogurt, research showed an increase of B and NK-cells in lymph nodes and IFN-
7 production (De Simone et al., 1986, 1988b, 1988c, 1989, 1993).

Administration of lactic acid bacteria or their fermented milks have been reported
to have an effect on production of cytokines. The administration of 1 x 108 CFU/day for 15
days of L. bulgaricus and S. thermophilus used in the manufacture of fermented milk
products, either in mice (Solis-Pereyra et al., 1991) or humans (Solis-Pereyra and Lemonnier,

1993), resulted in increased levels of IL-1 [3, TNF-a, IFN-y. This effect was even higher

214

when using L. casei, L. acidophilus and Bifidobacterium spp. fermented milks (Solis-Pereyra
and Lemonnier, 1993). Recently, in this laboratory the effects of dietary yogurt manufactured
with different commercial starters were studied regarding the cytokine gene expression in
mice. Interestingly, the yogurts suppressed some cytokine mRN A either at mucosal or
systemic level (Ha, unpublished data) but bacterial strain specificity was not detected.
Given the reported therapeutic action of L. acidophilus (Perdigon et al., 1990, 1991b;
Nader de Macias et al., 1993; Nomoto et al., 1985; Saito, 1988; Paubert-Braquet et al., 1995;
Bourlioux, 1986) and Bifidobacterium spp. (Mitsuoka, 1990, Robinson, 1991; Moineau et
al., 1989; Moineau and Goulet, 1991), a question that remains to be answered is: “why not
manufacture a yogurt or fermented milk using only L. acidophilus and Bifidobacterium spp.
as starter cultures?”. First of all, for a commercially fermented milk to be called yogurt, it has
to contain active L. bulgaricus and S. thermophilus (Trapp et al., 1993). Second, by
excluding the conventional yogurt bacteria fi'om the starter culture, we could be also
eliminating some interactions or formation of by-products which could have an influence in
the outcome of the reported effects. Finally, fermented milks have been produced with L.
acidophilus attempting to produce a yogurt-like product. However, these microorganisms do
not produce acetaldehyde which gives the characteristic flavor of regular yogurt (Sellars,
1991). As a result, fermented milk with only L. acidophilus is tart and plain. Furthermore,
fermented milk produced only with Br’fidobacterium also presents texture and flavor
problems. Thus, the food industry has started adding bifidobacteria and L. acidophilus
cultures to milk and its by-products (yogurt in particular) (Rasic and Kurrnann, 1983; Wood

and Holzapfel, 1995) and the normal practice is to include L. acidophilus and

 

215

Bifidobacterium in conventional yogurt starter cultures. The most common species used are
B. bifidum, B. longum and B. breve often combined with cultures of L. acidophilus and S.
thermophilus to facilitate acidification. Introducing lactobacilli and bifidobacteria into the
food chain to exert the specific therapeutic properties experimentally reported may be
difficult. To achieve high microorganism counts, the ingested portion should be 500 ml of
commercial yogurt or fermented milk per day (Alm et al., 1993), and this fermented milk
must be taken regularly for an unlimited time period as the administrated lactobacilli are
washed out after approximately 14 days from the intestinal tract if not continually
administered.

In summary, a murine model has been established in which the adj uvant activity of
yogurt containing L. acidophilus and Bifidobacterium spp. was demonstrated by generating
a strong gut mucosal IgA anti-CT response. Yogurt manufactured with starters containing
only yogurt bacteria L. bulgaricus and S. thermophilus produced decreased IgA-anti CT
when compared either to the NFDM fed control group or to other groups fed different types
of yogurt made with L. bulgaricus and S. thermophilus supplemented with L. acidophilus and

Bifidobacterium spp.

216

4.6 FUTURE WORK

Since the effect if L. acidophilus and Bifidobacterium spp. has been reported to be
strain dependent and host specific, deve10pment of adequate screening tests are needed to
select strains of these microorganisms for use as dietary adjuncts in dairy products or other
foods, which produce the desired effects at a maximmn level. The mechanistic bases for the
findings in this study require further research. The change in permeability of intestinal cells,
increased immunogen translocation or uptake by APC and involvement of other immune
cells and cytokines would be interesting to decipher, to give a complete picture of the
mechanism of action. The fact that a potential increase in secretion of CT-specific IgA and
IgG was observed on Peyer’s patch and spleen cell cultures coming from mice treated with
yogurts containing L. acidophilus and Bifidobacterium spp., but without statistically
significant differences, might probably be attributed to limitations on the culture system. In
this study, the absence of a direct CT challenge to the cell culture resulted in a lack of
proliferation and/or differentiation to plasma cells capable of immunoglobulin secretion.
These cells probably would need to be co-cultured with the same antigen (CT), and if
immunoglobulin secretion is detected, that could be indicating a local memory B cell

reaction. Other immunogens should be also tested.

CHAPTER 5

SUMMARY

217

218

In this study, I evaluated the potential immunomodulatory activity of lactobacilli,
streptococci, bifidobacteria and their extracts in the gastrointestinal system and in the
immune cells by using the mouse model. In vivo feeding studies as well as ex vivo and in
vitro studies were conducted in an attempt to show such an activity and identify the cellular
components that possess the irnmunostimulatory properties.

This investigation suggests that lactic acid bacteria may alter peritoneal cell function
in vivo, and this may subtly impact immunity. I demonstrated that lactic acid bacteria
stimulate mainly macrophages to produce cytokines (IL-6 and TNF-on) and NO, as well as
some other immune cells (such as NK cells) to produce IFN-y. The results pointed to a
possible relation between peritoneum and gastrointestinal system, suggesting a role of
peritoneal cells in the activation of mucosal immune tissues by migration of cells via lymph
and/or blood to Peyer’s patches or spleen. Further mechanistic understanding is needed on
how these and other immunologic effects are mediated by lactic acid bacteria.

Activation of macrophages to produce very significant amounts of these cytokines
and NO occurred under stimulation of either whole cell, cytoplasmic or cell wall fractions,
showing a similar pattern to stimulation produced with endotoxin, LPS. Further purification
of lactic acid bacteria fractions is required in an attempt to isolate and characterize the
stimulatory components. The mechanisms involved in the immunomodulatory effect might
be close to the mechanisms followed by LPS. Measurement of intracellular concentration of
calcium and/or NF-KB in macrophages after lactic acid bacteria exposure and comparison
with LPS-mediated signal transduction could provide an approximation of how closely

related these mechanisms are, and clarify the possible differences between stimulation with

219

Gram positive and Gram negative bacterial components.

Finally, I was able to establish a murine model in which the adjuvant activity of a
fermented milk containing specific lactic acid bacteria was demonstrated by generating a
strong gut mucosal IgA response against an antigen (CT) used through oral immunizations,
what suggests that species/strains of lactic acid bacteria are critically important for
immunomodulation. Since the effect of L. acidophilus and Bifidobacterium spp. may be
strain dependent and host specific, development of adequate screening tests are needed to
select strains for use as dietary adjuncts in dairy products or other foods. The involvement
of other immune cells and cytokines would be interesting to decipher afier challenging with

this same immunogen and also with some others.

LIST OF REFERENCES

220
LIST OF REFERENCES

Aattori, N. and Lemonnier, D. (1997). Production of interferon induced by Streptococcus
thermophilus: role of CD4+ and CD8+ lymphocytes. Nutr. Biochem, 8 : 25-31.

Abbas, A. K., Lichtrnan, A. H., and Pober, J. S. (1994). Cellular and Molecular
Immunology. 2nd ed. (W.B.Saunders ed.). Saunders Co. Philadelphia, PA.

Abe, F., Momose, H., Igarashi, M., Yaeshima, T., Ishibashi, N., Ahimarnura, S., and
Kawashima, T. (1996). The effect of administration of bifidobacteria on the intestinal
flora and grth of newborn piglets. J. Gen. Appl. Microbial, 42 : 25 7-262.

Abgrall, S., Joly, V., Derkinderen, P., Decre’, D., Carbon, C., and Yeni, P. (1997).
Lactobacillus casei infection in an AIDS patient. Eur. J. Clin. Microbial. Infect. Dis, 16 :
180-182.

Ahn, L. (1991). The therapeutic effects of various cultures: an overview. Therapeutic
effects of fermented milk.(R.K.Robinson ed.). Elsevier. New York.

Alm, L., Ryd-Kjellén, E., Setterberg, G., and Blomquist, L. (1993). Effect of a new
fermented milk product "cultura" on constipation in geriatric patients. In: The lactic acid
bacteria.(E.L.Foo, H.G.Griffin, R.Mollby, and C.G.Hedén ed.). Horizon Scientific Press.

Auclair, J. and Accolas, J. P. (1983). Use of thermophilic lactic starters in the dairy
industry. Antonie van Leeuwenhoek, 49 : 313-326.

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A.,
and Struhl, K. S. (1994). Related isolation procedures and funtional assays. Chapter
supplement 16, section IV, unit 3.17. In Current protocols in immunology.(Ausubel, F.
M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K.
S. ed.). John Wiley and Sons, Inc. New York.

Azcona-Olivera, J. I., Ouyang, Y., Murtha, J., Chu, F. S., and Pestka, J. J. (1995).
Induction of cytokine mRNAs in mice after oral exposure to the trichothecene vomitoxin
(deoxynivalenol): relationship to toxin distribution and protein synthesis inhibition.
T oxical. Appl. Pharmacol, 133 : 109-120.

221

Baricault, L., Denariaz, G., Houri, J. J., Bouley, C., Sapin, C., and Trugnan, G. (1995).
Use of HT—29, a cultured human colon cancer cell line, to study the effect of fermented
milks on colon cancer cell grth and differentiation. Carcinogenesis, 16(2) : 245-252.

Bellingan, G. J ., Caldwell, H., Howie, S. E. M., Dransfield, I., and Haslett, C. (1996). In
viva fate of the inflammatory macrophage during the resolution of inflammation. J.
Immunol, 157 : 2577-2585.

Benoit, A. . (1981) Contribution a l'etude du yaourt et de son conditionnement. Ecole
Nationale Veterinaire d'Alfort. Dissertation.

Beutler, B. and Krays, V. (1995). Lipopolysaccharide signal transduction, regulation of
tumor necrosis factor biosynthesis and signaling by tumor necrosis factor itself. J.
Cardiovasc. Pharmacol, 25(suppl. 2) : 81-88.

Black, F., Einarsson, K., Lidbeck, A., Orrhage, K., and Nord, C. E. (1991). Effect of
lactic acid producing bacteria on the human intestinal microflora during ampicilin
treatment. Scand. J. Infect. Dis., 23 : 247-254.

Blancuzzi, V., Roberts, E. D., Wilson, D., Fryer, L. R., O'Byrne, E. M., and DiPasquale,
G. (1993). Pathogenesis of Lactobacillus casei-induced polyarthritis in Lewis rats: 1. time
related changes in histopathological scores and hematology. Agents Actions, 39 : C183-
C185.

Bogdanov, I. G., Dalev, P. G., Gurevich, A. 1., Kolosov, M. N., Mal'kova, V. P.,
Plemyannykova, L. A., and Sorokina, I. B. (1975). Antitumor glycopeptides from
Lactobacillus bulgaricus cell wall. F EBS Lett., 57 : 259-261.

Bourlioux, P. (1986). Interét de l'ingestion volontaire dy microorganisms: cas de la
microflore du yaourt. Cah .Nutr. Diet, 21 : 204—208.

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

Breidt, F. and Fleming, H. P. (1997). Using lactic acid bacteria to improve the safety of
minimally processed frutis and vegetables. Food T echn., 51(9) : 44-48.

222

Cavallo, M. G., F ava, D., Monetini, L., Barone, F ., and Pozzilli, P. (1996). Cell-mediated
immune response to beta-casein in recent onset insulin-dependent diabetes: implications
for disease pathogenesis. Lancet, 348(9032) : 926-928.

Chesrown, S. E., Monnier, J. Visner G., and Nick, H. S. (1994). Regulation of inducible
nitric oxide synthase mRNA levels by LPS, IFN-y, TGF-B and IL-10 in murine

macrophage cell lines and rat peritoneal macrophages. Biochem. Biophys. Res. Commun,
200 : 126-132.

Cicco, N. A., Lindermann, A., Content, J ., Vandenbussche, P. Lubbert M., Gauss, J.,
Mertelsmann, R., and Hernnann, F. (1990). Inducible production of interleukin 6 by
human polymorphonuclear neutrophils: role of granulocyte macrophage colony-
stimulating factor and tumor necrosis factor-alpha. Blood, 75(10) : 2049-2052.

Coconnier, M. H., Liévin, V., Bemet-Carnard, M. F., Hudault, S., and Servin, A. L.
(1997). Antibacterial effect of the adhering human Lactobacillus acidophilus strain LB.
Antimicrob. Agents Chemother., 41(5) : 1046-1052.

Davidkova, G., Popova, P., Guencheva, G., Bogdanov, A., Pacelli, E., Auter, A., and
Mincheva, V. (1992). Endogenous production of tumor necrosis factor in normal mice
orally treated with Deodan- a preparation from Lactobacillus bulgaricus LBSl. Int. J.
Immunapharmacol., 14 : 1355-1362.

Davidson, P. M. and Hoover, D. G. (1993). Antimicrobial components from lactic acid
bacteria. Chapter 5. In Lactic Acid Bacteria.(S.Salminen and A.von Wright ed.). Marcel
Dekker, Inc. New York.

De Ambrosini, V. M., Gonzalez, 8., Perdigon, G., de Ruiz Holgado, A. P., and Oliver, G.
(1996). Chemical composition of the cell wall of lactic acid bacteria and related species.
Chem. Pharm. Bull, 44(12) : 2263-2267.

De Man, J. C., Rogosa, M., and Sharpe, M. E. (1960). A medium for the cultivation of
Lactobacilli. J. Appl. Bacterial, 23(1) : 130-135.

De Petrino, S. F., De Jorrat, M. E. B. B., Meson, O., and Perdigon, G. (1995). Protective
ability of certain lactic acid bacteria against an infection with Candida albicans in a
mouse immunosuppression model by corticoid. Food Agric. Immunol, 7 : 365-373.

 

223

De Simone, C., Salvadori, B. B., Negri, R., Ferrazzi, M., Baldinelli, L., and Vesely, R.
(1986). The adjuvant effect of yogurt on production of gamma-interferon by con-A
stimulated human peripheral blood lymphocytes. Nutr. Rep. Int, 33 (3): 419-433.

De Simone, C., Vesely, R., Negri, R., Bianchi Salvadori, B., Tzantzoglou, S., Cilli, A.,
and Lucci, L. (1987a). Enhancement of immune response of murine Peyer's patches by a
diet supplemented with yogurt. Immunopharmacal. Immunotaxical. , 9 : 87-100.

De Simone, C., Ferrazzi, M., Di Seri, M., Momgio, D., Baldinelli, L., and Di Fabio, S.
(1987b). The immunoregulation of the intestinal flora bifidobacteria and lactobacilli
modulate the production of gamma-interferon induced by pathogenic bacteria. Int. J.
Immunother., 3 : 151-158.

De Simone, C., Tzantzoglou, S., Baldinelli, L., Di Fabio, S., Bianchi-Salvadori, B.,
Jirillo, E., and Vesely, R. (1988a). Enhancement of host resistance against Salmonella
thyphimurium infection by a diet supplemented with yoghurt. Immunopharmacal.

Immunatoxical., 10(3) : 399-415.

De Simone, C., Baldinelli, L., Di Fabio, S., Tzantzoglou, S., Jirillo, E., Bianchi-
Salvadori, B., and Vesely, R. (1988b). Lactobacilli feeding increases NK cells and
gamma-IFN levels in humans. Dietetics in the 908. Role of the dietitian/Nutritionist, (177)
:180-

De Simone, C., Grassi, P. P., Bianchi-Salvadori, B., Miragliotta, G., Vesely, R., and
Jirillo, E. (1988c). Adherence of specific yoghurt microorganisms to human peripheral
blood lymphocytes. Microbios, 55 : 49-57.

De Simone, C., Bianchi-Salvadori, B., Tzantzoglou, S., Jirillo, E., Camaschella, P.,
Cislaghi, S., Ciardi, A., and Vesely, R. (1992). Bacterial translocation and immunological
responses in mice monoassociated or biassociated with Lactobacillus bulgaricus and
Escherichia cali. Dyn. Nutr. Res. Basel, Karger, l : 57-65.

De Simone, C., Vesely, R., Bianchi-Salvadori, B., and Jirillo, E. (1993). The role of
probiotics in modulation of the immune system in man and in animals. Int. J.
Immunother., XI(1) : 23-28.

de Vos, T and Dick, T. A. (1991). A rapid method to determine the isotype and specificity
of coproantibodies in mice infected with T richinella or fed cholera toxin. J. Immunol.
Methods, 141 :285-288.

224

Dieter, R. R., Hotchkiss, J. H., Austic, R. E., and Sung, Y. J. (1995). Production of
reactive nitrogen intermediates by macrophages. Chapter 8. In Methods in
Immunotoxicology, 2.(Dieter, R. R., Hotchkiss, J. H., Austic, R. E., and Sung, Y. J. ed.).
Wiley-liss. New York.

Donohue, D. C., Deighton, M., Ahokas, J. T., and Salminen, S. (1993). Toxicity of lactic
acid bacteria. Chapter 11. In Lactic Acid Bacteria.(S.Salminen and A.von Wright ed.).
Marcel Dekker, Inc. New York.

Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F. (1956).
Colorimetric method for determination of sugars and related substances. Anal. Chem,
28(3) : 350-356.

Dziarski, R. (1979). Mitogenic activity of staphylococcal peptidoglycan. Infect.1mmun.,
23 : 706-710.

Dziarski, R. (1982). Studies on the mechanism of peptidoglycan- and lipopolysaccharide-
induced polyclonal activation. Infect. Immu., 35(2) : 507-514.

Eason, D. and Martin, W. (1995). Involvement of tyrosine kinase and protein kinase C in

the induction of nitric oxide synthase by lipopolysaccharide and interferon-gamma in
J774 macrophages. Arch. Int. Pharmacodyn. Ther., 330(2) : 225-240.

Egan, A. F. (1983). Lactic acid bacteria of meat and meat products. Antonie van
Leeuwenhoek, 49 : 327-336.

Eigler, A., Moller, J., and Endres, S. (1995). Exogenous and endogenous nitric oxide
attenuates tumor necrosis factor synthesis in the murine macrophage cell line RAW
264.7. Immunology, 154 : 4048-4052.

Elson, C. O. and Dertzbaugh, M. T. (1994). Mucosal adjuvants. Chapter 33. In Handbook
of mucosal immunology.(P.L.Ogra, M.E.Lamm, J.R.McGhee, J .Mestecky, W.Strober,
and J .Bienestock ed.). Academic Press. New York.

Epple, H. J., Kreusel, K. M., Hanski, C., Schulzke, J. D., Riecken, E. D., and From, M.
(1997). Differential stimulation of intestinal mucin secretion by cholera toxin and
carbachol. Eur. J. Physiol, 433 : 638-647.

225

Festing, M. F. W. (1979). Inbred strains in biochemical research. Oxford University
Press. New York, NY.

Fuller, R. (1991). Probiotics in human medicine. Gut, 32 : 439-442.

Gammage, R. B. and Berven, B. A. (1996). Indoor air and human health. 2nd ed. Lewis
publishers. New York.

Gasser, F. (1994). Safety of lactic acid bacteria and their occurrence in human clinical
infections. Bull. Inst. Pasteur, 92 : 45-67.

Gasson, M. J. (1993). Progress and potential in the biotechnology of lactic acid bacteria.
FEMS Microbial. Rev, 12 : 3-20.

Gerritse, K., Posno, M., Schellekens, M. M., Boersma, W. J. A., and Claassen, E. (1990).
Oral administration of TNP-Lactobacillus conjugates in mice: a model for evaluation of

mucosal and systemic immune responses and memory formation elicited by transformed
lactobacilli. Res. Microbial, 141 : 955-962.

Gilliland, S. E. (1985). Bacterial starter cultures for foods. CRC Press, Inc. Boca de
Raton, FL.

Gilliland, S. E. (1989). Acidophilus milk products: a review of potential benefits to
consumers. J. Dairy Sci, 72 : 2483-

Gilliland, S. E. (1990). Health and nutritional benefits from lactic acid bacteria. FEMS
Microbial. Rev., 87 : 175-188.

Gomez-Flores, R., Rodriguez-Padilla, C., Mehta, R. T., Galan-Wong, L., Mendoza-
Gamboa, E., and Tamez-Guerra, R. (1997). Nitric oxide and TNF-on production by
murine peritoneal macrophages activated with a novel 20kDa protein isolated from
Bacillus thuringensis var. thuringensis parasporal bodies. J. Immunol., 158 : 3796-3799.

Gonzalez, S. N., Cardozo, R. Apella M. C., and Oliver, G. (1995). Biotherapeutic role of
fermented milk. Biatherapy, 8 : 129-134.

Guencheva, F ., Popova, P., Davidkova, G., Mincheva, V., Mihailova, S., Bogdanov, A.,
Pacelli, E., and Auteri, A. (1992). Determination of cytokine release after in viva and in

226

vitra administration of Deodan (a preparation from Lactobacillus bulgaricus LB51) by
the rabbit pyrogen test. Int. J. Immunopharmacal., 14(8) : 1429-1436.

Gumina, R. J., Freire-Moar, J., DeYoung, L., Webb, D. R., and Devens, B. H. (1991).
Transduction of the IFN-y signal for HLA-DR expression in the promonocytic line THP-l
involves a late-acting PKC activity. Cell. Immunol., 138 : 265-279.

Ha, C-L., Lee, J. H., Zhou, H. R., Ustunol, Z., and Pestka, J. J. (1997). Effect of yogurt
ingestion on mucosal and systemic cytokine gene expression in the mouse. Unpublished
data.

Halpem, G. M., Vruwing, G., Van de Water, J ., Keen, C. L., and Gershwin, M. E. (1991).
Influence of long-term yoghurt consumption in young adults. Int. J. Immunother., VH :
205-210.

Hanson, L. A., Ahlstedt, S., Dahlgren, U, and Soderstrom, T. (1983). The gut and the
immune response. Symposia of the Swedish Nutrition Foundation XV .(Bo Hallgren ed.).
Almqvist & Wiksell International. Stockholm.

Hashimoto, S., Nomoto, K., Nagaoka, M., and Yokokura, T. (1987). In vitra and in viva
release of cytostatic factors from Lactobacillus casei-elicited peritoneal macrophages
after stimulation with tumor cells and immunostirnulants. Cancer Immunol. Immunother. ,
24 : 1-7.

Hatcher, G. E. and Lambrecht, R. S. (1993). Augmentation of macrophage phagocytic
activity by cell-free extracts of selected lactic acid-producing bacteria. J. Dairy Sci., 76 :
2485-2492.

Heel, K. A., McCauley, R. D., Papadimitriou, J. M., and Hall, J. C. (1997). Review:
Peyer's patches. J. Gastraenteral. Hepatal., 12 : 122-136.

Henderson, B. and Wilson, M. (1996). Cytokine induction by bacteria: beyond
lipopolysaccharide. Review article. Cytokine, 8(4) : 269-282.

Hendrzak, J. A. and Brunda, M. J. (1995). Interleukin 12 biologic activity, therapeutic
utility and role in disease. Lab. Invest. , 72 : 619-637.

227

Herzenberg, L. A., Weir, D. M., and Blackwell, C. (1997). Weird's handbook of
experimental immunology. Blackwell Science, Inc. Malden, MA.

Heumann, D., Barras, C., Severin, A., Glauser, M. P., and Tomasz, A. (1994). Gram
positive cell walls stimulate synthesis of tumor necrosis factor alpha and interleukin 6 by
human monocytes. Infect. Immun.., 62(7) : 2715-2721.

Hilton, E., Isenberg, H. D., Alperstein, P., France, K., and Borenstein, M. T. (1992).
Ingestion of yogurt containing Lactobacillus acidophilus as prophylaxix for candida]
vaginitis. Ann. Inter. Med, 116 : 353-357.

Hirvonen, M. R., Nevalainen, A., Makkonen, N., Monkkonen, J ., and Savolainen, K.
(1997). Streptomyces spores from mouldy houses induce nitric oxide, TNF-a and IL-6
secretion from RAW 264.7 macrophage cell line without causing subsequent cell death.
Environ. Toxicol. Pharmacol, 3 : 57-63.

Hitchins, A. D., McDonough, F. R., Wong, N. P., and Hargrove, R. E. (1983). Biological
and biochemical variables affecting the relative values for growth and feed efficiency of
rats fed yogurt or milk. J. Food Sci, 48 : 1836-1840.

Hosono, A., Lee, J ., Ametani, A., Natsume, M., Hirayama, M., Adachi, T., and
Karninogawa, S. (1997). Characterization of a water-soluble polysaccharide fraction with
immunopotentiating activity from Bifidabacterium adalescentis M101-4. Biasci. Biotech.
Biachem. , 61(2) : 312-316.

Hugenholtz, J. (1993). Citrate metabolism in lactic acid bacteria. FEMS Microbial. Rev.,
12 : 165-178.

Hughes, D. B. and Hoover, D. G. (1991). Bifidobacteria: their potential for use in
American Dairy Products. Food T echn., April 1991 : 74-83.

Isolauri, E., Juntunen, M., Rautanen, T., Sillanaukee, P., and Koivula, T. (1991). A
human Lactobacillus strain (Lactobacillus casei sp. strain GG) promotes recovery from
acute diarrhea in children. Pediatrics, 88 : 90-97.

Isolauri, E., Joensuu, J., Suomalainen, H., Luomala, M., and Vesikari, T. (1995).
Improved immunogenicity of oral D x RRV reassortant rotavirus vaccine by
Lactobacillus casei GG. Vaccine, 13(3) : 310-312.

228

Jackson, R. J., Fujihashi, K., Xu-Amano, J., Kiyono, H., and McGhee, J. R. (1993).
Optimizing oral vaccines: induction of systemic and mucosal B cells and antibody

responses to tetanus toxoid by use of cholera toxin adjuvants. Infect. Immun, 61 : 4272-
4279.

Janeway, C. J. and Travers, P. (1997). Irnmunobiology: The immune system in health and
disease. 3rd ed. (J aneway, C. J. and Travers, P. ed.). Garland Publishing Inc. New York.

Jin, L. Z., Ho, Y. W., Abdullah, N., Ali, M. A., and Jalaludin, S. (1996). Antagonistic
effects of intestinal Lactobacillus isolates on pathogens of chicken. Lett. Appl.
Microbial, 23 : 67-71.

Jin, L. Z., Ho, Y. W., Ali, M. A., Abdullah, W., and Jalaludin, S. (1996). Effect of
adherent Lactobacillus spp. on in vitra adherence of Salmonellae to the intestinal
epithelial cells of chicken. J. Appl. Bacterial, 81 : 201-206.

Kaila, M., Isolauri, E., Soppi, E., Virtanen, E., Laine, S., and Arvilommi, H. (1992).
Enhancement of the circulating antibody secreting cell response in human diarrhea by a
human Lactobacillus strain. Pediatr. Res, 32 : 141-144.

Kaklij, G. S., Kelkar, S. M., Shenoy, M. A., and Sainis, K. B. (1991). Antitumor activity
of Streptococcus thermophilus against fibrosarcoma: role of T—cells. Cancer Lett., 56 :
37-43.

Kawabata, T. T., Burleson, G. R., Ernst, P. V., and Ullrich, S. R. (1995).
Irnmunotoxicology of regional lymphoid tissue: the respiratory and gastrointestinal tract
and skin. F undam. Appl. T axical., 26 : 8-19.

Kawasaki, E. S. (1991). Amplification of RNA. PCR Protocols: A guide to Methods and
Applications.(M.A. Innis, D.H. Gelfand, H. Gelfand, J.J. Sninsky, and T.J. White ed.).
Academic Press. San Diego.

Kim, S. H., Kim, T. H., and Han, D. V. (1996). The main peritoneal source of precursors
of the murine intestinal IgA plasma cell. Anticancer Res. , l6 : 2821-2824.

Kim, Y. M., Lee, B. S., Yi, K. Y., and Paik, S. G. (1997). Upstream NF-KB site is
required for the maximal expression of mouse inducible nitric oxide synthase gene in

interferon-y plus lipopolysaccharide-induced RAW 264.7 macrophages. Biachem. Biophy.
Res. Commun, 236 : 655-660.

229

Kirchner, H. (1986). The interferon system as an integral part of the defense system
against infection. Antiviral Research, 6 : 1-17.

Kishi, A., Kazuko, U., Matsubara, Y., Okuda, C., and and Kishida, T. (1996). Effect of
the oral administration of Lactobacillus brevis subsp. coagulans on interferon alpha
producing capacity in humans. J. Am. Coll Nutr., 15(4) : 408-412.

Kitazawa, H., Matsumura, K., Itoh, T., and Yamaguchi, T. (1992). Interferon induction in

murine peritoneal macrophages by stimulation with Lactobacillus acidophilus. Microbial.
Immunol., 36(3) : 311-315.

Kitazawa, H., Tomioka, Y., Matsumura, K., Aso, H., Mizugaki, M., Itoh, T., and
Yamaguchi, T. (1994). Expression of mRNA encoding IFN alpha in macrophages
stimulated with Lactobacillus gasseri. FEMS Microbial. Lett. , 120 : 315-322.

Klaenhammer, T. R. (1982). Microbiological considerations in selection and preparation
of Lactobacillus strains for use as dietary adjuncts. J. Dairy Sci, 65 : 1339-1349.

Klaenhammer, T. (1988). Bacteriocins in lactic acid bacteria. Biachimie, 70 : 337-356.

Klaenhammer, T. (1993). Genetics of bacteriocins produced by lactic acid bacteria.
FEMS Microbial. Rev., 12 : 39-86.

Klein, J. B., Schepers, T. M., Dean, W. L., Sonnefeld, G., and McLeish, K. R. (1990).
Role of intracellular calcium concentration and protein kinase C activation in IFN-y
stimulation of U937 cells. J. Immunol., 144(4) : 4305-4311.

Klinman, D. M., Yi, A. K., Beaucage, S. L., Conover, J., and Krieg, A. M. (1996). CpG
motifs present in bacterial DNA rapidly induce lymphocytes to secrete interleukin 6,
interleukin 12, and interferon-y. Prac. Natl. Acad Sci. USA, 93 : 2879-2883.

Kohwi, Y., Hashimoto, Y., and Tamura, Z. (1982). Antitumor and immunological
adjuvant effect of Bifidobacterium in mice. Bifidabacteria Micraflara, 1 : 61-66.

Kopeloff, N. (1926). Lactobacillus acidophilus. The Williams and Wilkins Press.
Baltimore, MD.

230

Krieg, A. M., Yi, A. K., Matson, S., Waldschmidt, T. J., Bishop, G. A., Teasdale, R.,
Koretzky, G. A., and Klinman, D. M. (1995). CpG motifs in bacterial DNA trigger direct
B-cell activation. Nature, 374 : 546-549.

Kroese, F. G. M., Butcher, E. C., Stall, A. M., Lalor, P. A., Adams, 8., and Herzenberg,
L. A. (1989). Many of the IgA producing plasma cells in murine gut are derived from
self-replenishing precursors in the peritoneal cavity. Int. Immunol. , 1 : 75-84.

Laskin, J. D., Heck, S. E., and Laskin, D. L. (1994). Multifunctional role of nitric oxide in
inflammation. T EM, 5 : 377-382.

Laegreid, A., Kolsto Otnaess, A. B., and Fuglesang, J. (1986). Human and bovine milk:
comparison of ganglioside composition and enterotoxin-inhibitory activity. Pediatr. Res,
20(5) : 416-421.

Le Roy, D., Morand, P., Lengacher, S., Celio, M., Grau, G. E., Glauser, M. P., and
Heumann, D. (1996). Streptococcus mitis cell walls and lipopolysaccharide induce
lethality in D-galactosamine-sensitized mice by a tumor necrosis factor dependent
pathway. Infect. Immun, 64(5) : 1846-1849.

Lee, S. Y., Vedamuthu, E. R., Washarn, C. J., and Reinbold, G. W. (1974). An agar
medium for the differential enumeration of yogurt starter bacteria. J. Milk F aad Technol,
37 : 272-278.

Lee, J. H., Ametani, A., Enomoto, A., Sato, Y., Motoshima, H., and Ike, F. a. K. J.
(1993). Screening for the immunopotentiating activity of food microorganisms and
enhancement of the immune response by Bifidabacterium adalescentis MlOl-4. Biasci.

Biotech. Biachem. , 57(12) : 2127-2132.

Lee, Y. K. and Salminen, S. (1995). The coming of age of probiotics. Trends Food Sci.
T echnol. , 6 : 241-245.

Lesourd, B. M. (1997). Nutrition and immunity in the elderly: Modification of immune
responses with nutrition treatments. Am. J. Clin. Nutr. , 66 : 478S-484S.

Lin, H., Wei, R. Q., and Bolling, S. F. (1995). Tumor necrosis factor-alpha and interferon
gamma modulation of nitric oxide and allograft survival. J. Surg. Res, 59(1) : 103-110.

231

Lindgren, S. E. and Dobrogosz, W. J. (1990). Antagonistic activities of lactic acid
bacteria in food and feed fermentations. FEMS Microbial. Rev., 87 : 149-164.

Link-Amster, H., Rochat, F ., Saudan, K. Y., Mignot, 0., and Aeschlirnann, J. M. (1994).
Modulation of a specific humoral immune response and changes in intestinal flora
mediated through fermented milk intake. FEMS Immunol. Med. Microbial, 10 : 55-64.

Lortal, S., Valence, F., Bizet, C., and Maubois, J. L. (1997). Electrophoretic pattern of
peptidoglycan hydrolases, a new tool for bacterial species identification: application to 10
Lactobacillus species. Res. Microbial, 148 : 461-474.

MacMicking, J ., Xie, Q., and Nathan, C. (1997). Nitric oxide and macrophage function.
Annu. Rev. Immunol., 15 : 323-350.

Malin, M., Suomalainen, H., Saxelin, M., and Isolauri, E. (1996). Promotion of IgA
immune response in patients with Crohn's disease by oral bacteriotherapy with
Lactobacillus GG. Ann .Nutr. Metab, 40 : 137-145.

Mao, Y., Yu, J. L., Ljungh, A., Molin, G., and Jeppsson, B. (1996). Intestinal immune
response to oral administration of Lactobacillus reuteri R2LC, Lactobacillus plantarum
DSM 9843, Pectin and Oatbase on methotrexate-induced enterocolitis in rats. Micrab.
Ecol. Health Dis., 9 : 261-270.

Marin, M. L., Tejada-Simon, M. V., Lee, J ., Ustunol, Z., and Pestka, J. J. (1997a). Effects
of Lactobacillus spp. on cytokine production by RAW 264.7 macrophages and EL-4
Thymoma cell lines. J. Food Prat, 60(11) : 1364-1370.

Marin, M. L., Lee, J. H., Murtha, J ., Ustunol, Z., and Pestka, J. J. (1997b). Differential
cytokine production in clonal macrophage and T-cell lines cultured with Bifidobacteria. J.
Dairy Sci, 80 : 2713-2720.

Marin, M. L., Tejada-Simon, M. V., Lee, J. H., Murtha, J., Ustunol, Z., and Pestka, J. J.
(1997c). Stimulation of cytokine production in clonal macrophage and T cell models by
Streptococcus thermophilus: comparison with Bifidabacterium spp. and Lactobacillus
bulgaricus. J. Food Prat. , (submitted)

Marteau, P. and Rambaud, J. C. (1993). Potential of using lactic acid bacteria for therapy
and immunomodulation in man. FEMS Microbial. Rev., 12 : 207-220.

232

Marteau, P., Vaerman, J. P., Dehenin, J. P. Pochart P., Desjeux, J. F., and Rambaud, J. C.
(1996). Effects of Lactobacillus acidophilus strain LAl on serum concentration and
jejunal secretions of immunoglobulins and serum proteins in healthy humans. SOMED
21St International congress on microbial ecology and disease.

Martin, C. A. and Dorf, M. E. (1990). Interleukin-6 production by murine macrophage
cell lines P388D1 and J774A.1: stimulation requirements and kinetics. Cell Immunol.,
128(2) : 555-568.

Mateos Garcia, A. I. Contribucion a1 estudio de la microflora contaminante del yogurt.
1984. Universidad Complutense de Madrid. Dissertation.

McBurney, M. I. (1994). The gut: central organ in nutrient requirements and metabolism.
Can. J. Physio]. Pharmacol, 72 : 260-265.

McCann, T., Egan, T., and Weber, G. H. (1995). Assay procedures for commercial
probiotic cultures. J. Food Prat, 59(1) : 41-45

McDonough, F. E., Wong, M. P., Wells, P., Hitchins, A. D., and Bodwell, C. E. (1985).
Stimulation of rat growth by yogurt: effect of vitamins and minerals. Nutr. Rep. Int, 31 :
1237-1247.

McGhee, J. R., Mestecky, J., Elson, C., and Kiyono, H. (1989). Regulation of IgA
synthesis and immune response by T cells and interleukins. J. Clin. Immunol., 9 : 175-
199.

McGroarty, J. A. (1993). Probiotic use of lactobacilli in the human female urogenital
tract. FEMS Immun. Med. Microbiol, 6 : 251-264.

McKay, L. L. and Baldwin, A. (1990). Applications for biotechnology: present and future
improvements in lactic acid bacteria. FEMS Microbial. Rev., 87 : 3-14.

Metchnikoff, E. (1907). The prolongation of life. G.P. Putnam and Sons. The
Knickerbocker Press, New York.

Miettinen, M., Vuopio-Varkilla, J ., and Varkila, K. (1996). Production of human tumor
necrosis factor alpha, interleukin 6 and interleukin 10 is induced by lactic acid bacteria.
Infect. Immun., 64(12) : 5403-5405.

233

Migliore-Samour, D., Floc'h, F ., and Jolles, P. (1989). Biologically active casein peptides
implicated in immunomodulation. J. Dairy Res, 56(3) : 357-362.

Mitsuoka, T. (1990). Role of intestinal flora in health with special reference to dietary

control of intestinal flora. Microbiology Application in Food Biotechnology.(B.H. Nga
and Y.K. Lee ed.). Elsevier, Essex.

Moineau, S., Boutin, Y., and Goulet, J. (1989). Effect of fermented milks on immune
response in mice. J. Dairy Sci, 72(suppl.1) : 174-178.

Moineau, S. and Goulet, J. (1991). Effect of feeding fermented milks on the pulmonary
macrophage activity in mice. Milchwissenschaft, 46 : 551-554.

Mosmarm, T. (1983). Rapid colorimetric assay for cellular growth and
survival:application to proliferation and cytotoxicity assays. J. Immunol. Methods, 65(1-
2): 55-63.

Muscettola, M., Massai, L., Tanganelli, C., and Grasso, G. (1994). Effects of lactobacilli
on interferon production in young and aged mice. Ann.. N. Y. Acad. Sci, 717 : 226-232.

Nader de Macias, M. E., Apella, M. C., Romero, N. C., Gonzalez, S. N., and Oliver, G.
(1992). Inhibition of Shigella sonnei by Lactobacillus casei and Lactobacillus
acidophilus. J. Appl. Bacterial, 73 : 407-411.

Nader de Macias, M. E., Romero, N. C., Apella, M. C., Gonzales, S. M., and Oliver, G.
(1993). Prevention of infections produced by Escherichia caIi and Listeria
manacytagenes by feeding milk fermented with lactobacilli. J F aad Prat ., 56 : 401-405.

Nanno, M., Shimizu, T., Mike, A., Ohwaki, M., and Mutai, M. (1988). Role of
macrophages in serum colony-stimulating factor induction by Lactobacillus casei in
mice. Infect. Immun., 56(2) : 357-362.

Nes, I. F., Diep, D. B., Havarstein, L. S., and Brurberg, M. B. (1996). Biosynthesis of
bacteriocins in lactic acid bacteria. Antonie van Leeuwenhoek, 70 : 113-128.

Nielsen, S. S. (1994). Introduction to the chemical analysis of foods. Jones and Bartlett
Publishers, Inc. Boston, MA.

234

Nikaido, H. and Vaara, M. (1985). Molecular basis of bacterial outer membrane
permeability. Microbial. Rev., 49(1) : 1-32.

Nomoto, K., Miake, S., Yokokura, T., Mutai, M., and Yoshikai, Y. (1985). Augmentation
of host resistance to Listeria manacytagenes infection by Lactobacillus casei. J. Clin.
Lab. Immunol., 17 : 91-97.

Nomoto, K., Yokokura, T., Tsuneoka, K., and Shikita, M. (1991). Radioprotection of
mice by a single subcutaneous injection of heat-killed Lactobacillus casei after
irradiation. Radiat. Res, 125 : 293-297.

Nousiainen, J. and Setala, J. (1993). Lactic acid bacteria as animal probiotics. Chapter 12.
In Lactic Acid Bacteria.(S. Salminen and A. von Wright ed.). Marcel Dekker, Inc. New
York.

Novakovic, S. and Boldogh, I. (1994). In vitra TNF-or production and in viva alteration of
TNF-(1 RNA in mouse peritoneal macrophages after treatment with different bacterial
derived agents. Cancer Lett., 81 : 99-109.

Ogra, P. L., Mestecky, J ., Lamm, M. E., Strober, W., McGhee, J. R., and Bienenstock, J.
(1994). Handbook of Mucosal Immunology.(Pearay Ogra et al. ed.). Academic Press. San
Diego.

Ohno, N., Hashimoto, T., Adachi, Y., and Yadomae, T. (1996). Conformation
dependency of nitric oxide synthesis of murine peritoneal macrophages by B-glucans in
vitra. Immunol. Lett., 52 : 1-7.

Okitsu-Negishi, S., Nakano, I., Suzuki, K., Hashira, 8., Abe, T., and Yoshino, Z. (1996).
The induction of cardioangitis by Lactobacillus casei cell wall in mice.l. the cytokine
production from murine macrophages by Lactobacillus casei cell wall extract. Clin.
Immunol. Immunopathol. , 78(1) : 30-40.

Paavonen, T. (1994). Hormonal regulation of immune responses. Ann. Med, 26 : 255-
258.

Pascual, D. W., Kiyono, H., and McGhee, J. R. (1996). Mucosal Immunity. Molecular
and cellular aspects of immune protection to enteric infections. Chapter 2. In Enteric
Infections and Immunity.(L.J. Paradise et al. ed.). Plenum Press. New York.

235
Paubert-Braquet, M., Xiao-Hu, G., Gaudichon, C., Hedef, N., Serikoff, A., Pouley, C.,
Bonavida, B., and Braquet, P. (1995). Enhancement of host resistance against Salmonella

typhimurium in mice fed a diet supplemented with yogurt or milks fermented with
various Lactobacillus casei strains. Int. J. Immunoth, XI(4) : 153-161.

Pelczar M.P., Chan, E. C. S., and Krieg, M. R. (1993). Microbiology: concepts and
applications. McGraw-Hill, Inc. New York.

Perdigon, G., Nader de Macias, M. E., Alvarez, S., Medici, M., Oliver, G., and Pesce de
Ruiz Holgado, A. (1986a). Effect of a mixture of Lactobacillus casei and Lactobacillus
acidophilus administered orally on the immune system in mice. J Food Prat, 49(12):

986-989.

Perdigon, G., Nader de Macias, M. E., Alvarez, S, Oliver, G., and Pesce de Ruiz
Holgado, A. A. (1986b). Effect of perorally administered Lactobacilli on macrophage
activation in mice. Infect. Immun., 53(2) : 404-410.

Perdigon, G., Nader de Macias, M. E., Alvarez, S., Oliver, G., and Pesce de Ruiz
Holgado, A. A. (1987). Enhancement of Immune response in mice fed with Streptococcus
thermophilus and Lactobacillus acidophilus. J. Dairy Sci, 70 : 919-926.

Perdigon, G., Nader de Macias, M. E., Alvarez, 8., Oliver, G., and Pesce de Ruiz
Holgado, A. (1988). Systemic augmentation of the immune response in mice by feeding
fermented milks with Lactobacillus casei and Lactobacillus acidophilus. Immunology,
63: 17-23.

Perdigon, G., Alvarez, S., Nader de Macias, M. E., Roux, M. E., and Pesce de Ruiz
Holgado, A. (1990). The oral administration of lactic acid bacteria increase the mucosal
intestinal immunity in response to enteropathogens. J. Food Prat, 53 : 404-410.

Perdigon, G., de Jorrat, M. E., de Petrino, F., and Valverde de Budeguer, M. (1991a).
Effect of oral administration of Lactobacillus casei on various biological firnctions of the
host. Food Agric. Immunol., 3 : 93-102.

Perdigon, G., Alvarez.S., Nader de Macias, M. E., Savoy de Giori, G., Medici, M., and
Nunez de Kairuz, M. (1991b). Behaviour of natural and heated yogurt in the immune
system and preventive capacity on enteric infections. Milchwissenschaft, 46(7) : 409-476.

236

Perdigon, G., Alvarez, S., and Medici, M. (1992). Systemic and local augmentation of the
immune response in mice by feeding with milk fermented with Lactobacillus acidophilus
and/or Lactobacillus casei. Dyn. Nutr. Res. Base], Karger, 1 : 66-76.

Perdigon, G., Medici, M., Bibas bonet de Jorrat, M. E., Valverde de Budeguer, M., and
Pesce de Ruiz Holgado, A. (1993). Irnmunomodulating effects of lactic acid bacteria on
mucosal and humoral immunity. Int. J. Immunoth, 19 : 29-52.

Perdigon, G., Alvarez, S., Gobbato, N., Budeguer, M. V., and Pesce de Ruiz Holgado, A.
(1995). Comparative effect of the adjuvant capacity of Lactobacillus casei and
lipopolysaccharide on the intestinal secretory antibody response and resistance to
Salmonella infection in mice. Food Agric. Immunol., 7 : 283-294.

Pestka, J. J ., Gaur, P. K., and Chu, F. S. (1980). Quantitation of aflatoxin B1 and aflatoxin
B1 antibody by an enzyme-linked immunosorbent microassay. App]. Environ. Microbial,
40: 1027-1031.

Pestka, J. J ., Moorman, M. A., and Warner, R. L. (1990). Altered serum immunoglobulin
response to model intestinal antigens during dietary exposure to vomitoxin
(deoxynivalenol). T axica]. Lett., 50 : 75-84.

Pestka, J. J. (1993). Food, diet and gastrointestinal immune function. In Advances in
Food and Nutrition Research, vol. 37. J .E. Kinsella. ed., Academic Press, Inc., San Diego,
CA.

Pisetsky, D. S. and Reich, C. (1993). Stimulation of in vitra proliferation of murine
lymphocytes by synthetic oligodeoxynucleotides. Ma]. Biol. Rep., 18 : 217-221.

Pisetsky, D. S. (1996). Immune activation by bacterial DNA: A new genetic code.
Immunity, 5 : 303-310.

Popova, P., Guencheva, G., Davidkova, G., Bogdanov, A., Pacelli, E., Opalchenova, G.,
Kutzarova, T., and Koychev, C. (1993). Stimulating effect of Deodan (an oral preparation
from Lactobacillus bulgaricus LB51) on monocytes/macrophages and host resistance to
experimental infections. Int. J. Immunopharmacal., 55 : 25-37.

Portier, A., Boyaka, N. P., Boudougodo, F., Dubarry, M., Huneau, J. F., Tome, D., and
Dodin, A. a. C. M. (1993). Fermented milks and increased antibody responses against
cholera in mice. Int. J. Immunoth, IX : 217-224.

237

Pritchard, G. G. and Coolbear, T. (1993). The physiology and biochemistry of the
proteolytic system in lactic acid bacteria. FEMS Microbiol. Rev., 12 : 179-206.

Puddu, P., Fantuzzi, L., Borghi, P., Varano, B., Rainaldi, G., Guillemard, E., Malomi,
W., Nicaise, P., Wolf, S. F., Belardelli, F., and Gessani, S. (1997). IL-12 induces IFN-y

expression and secretion in mouse peritoneal macrophages. J. Immunol., 159 : 3490-
3497.

Puri, P., Rattan, A., Bijlani, R. L., Mahapatra, S. C., and Nath, I. (1996). Splenic and
intestinal lymphocyte proliferation response in mice fed milk or yogurt and challenged
with Salmonella typhimurium. Int. J. Food Sci. Nutr., 47 : 391-398.

Radler, F. (1975). The metabolism of organic acids by lactic acid bacteria. In Lactic acid
bacteria in beverages and food.(J.G. Carr, C.V. Cutting, and GO Whiting ed.). Academic
Press. New York.

Rafter, J. J. (1995). Review: The role of lactic acid bacteria in colon cancer prevention.
Scand. J. Gastraenteral., 30 : 497-502.

Rasic, J. L. and Kurmann, J. A. (1983). Bifidobacteria and their role. Birkhauser Verlag.
Basel, Switzerland.

Rasic, J. Lj. and Sad, N. (1987). Culture media for detection and enumeration of
Bifidobacteria in fermented milk products. Bul]. IDF, 252 : 24-31.

Reeves, P. G., Nielsen, F. H., and Fahey, G. C. (1993). AIN-93 purified diets for
laboratory rodents: final report of the American Institute of Nutrition Ad Hoc Writing
Committee on the Reforrnulation of the AIN-76A rodent diet. J. Nutr. , 123:1939-951.

Rettger, L. G., Levy, M. N., Weinstein, L., and Weiss, J. E. (1935). Lactobacillus
acidophilus and its therapeutic application. Yale University Press. Haven, CT.

Ringden, O., Rynnel-Dagoo, B., Waterfield, E. M., Moller, E., and Moller, G. (1977).
Polyclonal antibody secretion in human lymphocytes induced by killed staphylococcal
bacteria and by lipopolysaccharide. Scand. J. Immunol., 6 : 1159-1169.

Robbins, J ., Schneerson, R., Szu, S., Fattom, A., Yang, Y., Lagergard, T., Chu, C., and
Sorensen, U. S. (1989). Prevention of invasive bacterial diseases by immunization with

238

polysaccharide protein conjugates. New strategies for oral immunization.(J. Mestecky and
J .R. McGhee ed.). Springer-Verlag. New York.

Robinson, R. K. (1991). Therapeutic properties of fermented milks. Elsevier, Essex.

Saito, H., Tomioka, H., and Nagashima, K. (1987). Protective and therapeutic efficacy of

Lactobacillus casei against experimental murine infections due to Mycobacterium
fortuitum complex. J. Gen. Microbial, 133 : 2843-2851.

Saito, H. (1988). Enhancement of host resistance to bacterial and viral infections by
Lactobacillus casei. Bifidabacteria Micraflora, 7 : 1-18.

Salnrinen, S. and von Wright, A. (1993). Lactic Acid Bacteria. Marcel Dekker, Inc. New
York.

Sambrook, J., Fritsch, E. F ., and Maniatis, T. (1989). Molecular cloning. A laboratory
manual. 2th ed. Cold Spring Harbor Laboratory Press, New York.

Sanders, M. E. (1993). Effect of consumption of lactic cultures on human health. In
Advances in Food and Nutrition Research, vol. 37. (J .E. Kinsella ed.). Academic Press
Inc. San Diego, CA.

Saucier, L., Julien, M., Cheour, F., Letarte, R., and Goulet, J. (1992). Effect of feeding
lactic acid bacteria and fermented milk on specific and nonspecific immune responses of
mice infected with Klebsiella pheumoniae AD-1.J. F aad Prat, 55(8) : 595-600.

Schiffrin, E. J ., Rochat, F., Link-Amster, H., Aeschlimann, J. M., and Donnet-Hughes, A.
(1995). Immunomodulation of human blood cells following the ingestion of lactic acid
bacteria. J. Dairy. Sci, 78 : 491-497.

Schiffrin, E. J., Brassart, D., Servin, A. L., Rochar, F ., and Bonnet-Hughes, A. (1997).
Immune modulation of blood leukocytes in humans by lactic acid bacteria: criteria for
strain selection. Am. J. Clin. Nutr., 66 : 5158-5208.

Schmucker, D. L., Heyworth, M. F., Owen, R. L., and Daniels, C. K. (1996). Impact of
aging on gastrointestinal mucosal immunity. Dig. Dis .Sci., 41(6) : 1183-1193.

239

Scrimshaw, N. S. and San Giovanni, J. P. (1997). Synergism of nutrition, infection and
immunity: an overview. Am. J. Clin. Nutr., 66 : 464S-477S.

Sekine, K., Toida, T., Saito, M., Kuboyarna, M., Kawashima, T., and Hashimoto, Y.
(1985). A new morphologically characterized cell wall preparation (whole peptidoglycan)
from Bifidobacterium infantis with a higher efficacy on the regression of an established
tumor in mice. Cancer Res, 45 : 1300-1307.

Sekine, K., Kawashima, T., and Hashimoto, Y. (1994). Comparison of the TNF-or levels
induced by human-derived Bifidabacterium langum and rat-derived Bifidabacterium
animalis in mouse peritoneal cells. Bifidabacteria Micraflara, 13(2) : 79-89.

Sekine, K., Watanabe-Sekine, E., Toida, T., Kasashima, T., Kataoka, T., and Hashimoto,
Y. (1995). Adjuvant activity of the cell wall of Bifidabacterium infantis for in vivo
immune responses in mice. Immunopharmaca] . Immunotoxicol, 16(4) : 589-609.

Sellars, R. (1991). Acidophilus products. Therapeutic Properties of Fermented
Milks.(R.K. Robinson ed.). Elsevier, Essex.

Shanahan, F. (1994). The intestinal immune system. Chapter 15. In Physiology of the
Gastrointestinal Tract, 3rd ed. (L.R. Johnson ed.). Raven Press. New York.

Sharpe, M. E. (1979). Lactic acid bacteria in the dairy industry. J. Soc. Dairy Tech,
32(1): 9-18.

Shin, H-S. (1997). Factors that influence viability of Bifidobacteria in milk. Michigan
State University. Master’s thesis.

Silliker, J. H., Elliot, R. P., Baird-Parker, a. C., Boyan, F. L., Christian, J. H. B., Clark, D.
8., Olson, J. C., and Roberts, T. A. (1980). Microbial ecology of foods. Academic Press.
New York.

Slonriany, B. L. and Somiany, A. (1992). Mechanism of Helicobacter pylori
pathogenesis: focus on mucus. J. Clin. Gastraenteral, 14(suppl. 1) : 8114-8121.

Smeyanov, V. V., Maltseva, N. N., Bossart, W., and Korshunov, V. M. (1992). Influence
of Lactobacillus acidophilus salca on immunological parameters of totally

240

decontaminated mice under the conditions of general germ-free isolation. Zhurna]
Mikrabialagii Epidemiologii i Immunabiolgii, 1 1 - 12 : 12-1 5.

Snider, D. P. (1995). The mucosal adjuvant activities of ADP-Ribosylating bacterial
enterotoxins. Crit. Rev. Immunol., 15(3 and 4) : 317-348.

Solis Pereyra, B., Falcoff, R., F alcoff, E., and Lemonnier, D. (1991). Interferon induction
by Lactobacillus bulgaricus and Streptococcus thermophilus in mice. Eur. Cytokine
Netw., 2(4) : 299-303.

Solis Pereyra, B and Lemonnier, D. (1993). Induction of human cytokines by bacteria
used in dairy foods. Nutrition research, 13 : 1127-1140.

Solis-Pereyra, B., Aattouri, N., and Lemonnier, D. (1997). Role of food in the stimulation
of cytokine production. Am. J. Clin. Nutr., 66 : 5218-5258.

Speck, M. L. (1980). Preparation of lactobacilli for dietary uses. J. Food Prat, 42(1) : 65-
67, 72.

Stacey, K. J ., Seet, M. J ., and Hume, D. A. (1996). Macrophages ingest and are activated
by bacterial DNA. J. Immunol., 157 : 2116-2122.

Steele, J. L. and Unlti, G. (1992). Impact of lactic acid bacteria on cheese flavor
development. F aad T echn., November 1992 : 128-135.

Stewart-Tull, D. E. S. and Davies, M. (1985). Immunology of the bacterial cell
envelope.(Stewart-Tull, D. E. S and Davies, M. ed.). John Wiley and Sons. New York.

Stiles, M. E. (1996). Biopreservation by lactic acid bacteria. Antonie van Leeuwenhoek,
70 : 331-345.

Svetic, A., Finkelman, F. D., Jian, Y. C., Dieffenbach, C. W., Scott, D. E., McCarthy, K.
F ., Steinberg, A. D., and Gause, W. C. (1991). Cytokine gene expression after in vivo
primary immunization with goat antibody to mouse IgD antibody. J. Immunol., 147(7) :
2391-2397.

Takada, H., Tsuimoto, M., Kotani, S., Kusumoto, S., Inage, M., Shiba, T., Nagao, S.,
Yano, 1., Kawata, S., and Yokogawa, K. (1979). Mitogenic effect of bacterial cell walls,

241

their fi'agments and related synthetic compounds on thymocytes and splenocytes of
guinea-pigs. Infect. Immun., 25 : 645-652.

Takahashi, T., Oka, T., Iwana, H., Kuwata, T., and Yamamoto, Y. (1993). Immune
response of mice to orally administered lactic acid bacteria. Biasci. Biotech. Biochem,
57(9) : 1557-1560.

Takeda, K., Saito, T., Kitazawa, H., Uemura, J ., and Itoh, T. (1997). Mitogenic activity of
whole cells and cell wall components of Lactobacillus acidophilus group lactic acid
bacteria on murine spleen and Peyer's patch cells. Milchwissenschafi, 52(1) : 21-25.

Tanaka, R. (1995). Clinical effects of Bifidobacteria and Lactobacilli. Hebora University
Seminar Monograph, 8 : 141-157.

Tejada Simon, M. V. and Seisdedos Dominguez, M. T. (1986). Leches fennentadas:
Yogur. Universidad Complutense de Madrid. Thesis.

Teraguchi, S., Uehara, M., Ogasa, K., Mitsuoka, T. (1978). Enumeration of bifidobacteria
in dairy products. Jap. .1. Bacterial, 33(6) : 753-761.

Teuber, M. (1991). Cultured dairy products in human nutrition. Bull. IDF, 266 : 2-24.

Tomioka, H., Sato, K., and Saito, H. (1990). Effect of ofloxacin combined with
Lactobacillus casei against Mycobacterium fortuitum infection induced in mice.
Antimicrob. Agents Chemother., 34 : 632-636.

Tortuero, F., Rioperez, J., Fernandez, E., and Rodriguez, M. L. (1995). Response of
piglets to oral administration of lactic acid bacteria. J. F aad Prat, 58(12) : 1369-1374.

Trapp, C. L., Chang, C. C., Halpem, G. M., Keen, C. L., and Gershwin, M. E. (1993).
The influence of chronic yogurt consumption on populations of young and elderly adults.
Int. J. Immunother., IX(1) : 53-64.

Truneh, A., albert, F., Golstein, P., and Schmitt-Verhulst, A. M. (1985). Early steps of
lymphocyte activation bypassed by synergy between calcium ionophores and phorbol
ester. Nature, 313 : 318-320.

242

Ueda, N., Kayashita, J ., Moriguchi, S., and Kishino, Y. (1990). Effect of dietary protein
levels on immune function of exercised rats. Nutr. Res, 10(4) : 429-437.

Vallejo, J. G., Baker, C. J., and Edwards, M. S. (1996). Roles of the bacterial cell wall
and capsule in induction of tumor necrosis factor alpha by type HI group B streptococci.
Infect. Immun., 64(12) : 5042-5046.

Wells, J. M., Robinson, K., Chamberlain, L. M., and Schofield, K. M. Le Page R. W. F.
(1996). Lactic acid bacteria as vaccine delivery vehicles. Antonie van Leeuwenhoek, 70 :
317-330.

Werfel, T., Ahlers, G., Schmidt, P., Boeker, M., and Kapp, A. (1996). Detection of a
kappa-casein-specific lymphocyte response in milk-responsive atopic dermatitis. Clin.
Exp. Allergy, 26(12) : 1380-1386.

Wesselingh, S. L., Levine, B., Fox, R. J ., Choi, S., and Griffin, D. E. (1994). Intracerebral

cytokine mRNA expression during fatal and nonfatal alphavirus encephalitis suggests a
predominant type 2 T cell response. J. Immunol., 152 : 1289-1297.

Wheeler, J. G., Bogle, M. L., Shema, S. J., Shirrell, M. A., Stine, K. C., Pittler, A. J.,
Burks, A. W., and Helm, R. M. (1997). Impact of dietary yogurt on immune function.
Am. J. Med. Sci, 313(2) : 120-123.

Wilson, D., O'Byme, E. M., Blancuzzi, V., Schlosser, M., Borman, C. H., and
DiPasquale, G. (1993). Pathogenesis of Lactobacillus casei-induced polyarthritis in
Lewis rats: 2. time related changes in organ weights and liver enzymes. Agents Actions,

39 : Cl98-C200.

Witsell, D. L., Garrett, C. G., Yarbrough, W. G., Dorrestein, S. P., Drake, A. F., and
Weissler, M. C. (1995). Effect of Lactobacillus acidophilus on antibiotic-associated
gastrointestinal morbidity: A prospective randomized trial. J. 0talarynga]. , 24 : 230-233.

Wood, B. J. B. and Holzapfel, W. H. (1995). The Lactic Acid Bacteria: The genera of
lactic acid bacteria.(B.J.B. Wood and W.H. Holzapfel ed.). Blackie Academic and
Professional. London.

Work, E. (1971). Cell walls. Chapter chapter XIII. In Methods in Microbiology. Ed. by
JR. Norris and D.W. Ribbons. Academic Press, New York.

 

243
Xu-Amano, J., Kiyono, H., Jackson, R. J., Staats, H. F., Fujihashi, K., Burrows, P. D.,
Elson, C. O., Pillai, S., and McGhee, J. R. (1993). Helper T cell subsets for
immunoglobulin A responses: oral immunization with tetanus toxoid and cholera toxin as

adjuvant selectively induces Th2 cells in mucosa associated tissues. J. Exp. Med, 178 :
1309-1320.

Yamamoto, K., Miwa, T., Taniguchi, H., Nagano.T., Shirnarnura, K., Tnaka, T., and
Kumagai, H. (1996). Binding specificity of Lactobacillus to glycolypids. Biachem.
Biahpys. Res. Commun, 228 : 148-152.

Yarnashita, M., Niki, H., Yamada, M., Mue, S., and Ohuchi, K. (1997). Induction of
nitric oxide synthase by lipopolysaccharide and its inhibition by auranofin in RAW 264.7
cells. Eur. J. Pharmacol, 338 : 151-158.

Yarnauchi, F. and Suetsuna, K. (1993). Immunological effects of dietary peptide derived
from soybean protein. J. Nutr. Biochem, 4(8) : 450-457.

Yasui, H., Mike, A., and Ohwaki, M. (1989). Irnmunogenicity of Bifidabacterium breve
and change in antibody production in Peyer's patches after oral administration. J. Dairy
Sci, 72 : 30-35. .

Yasui, H. and Ohwaki, M. (1991). Enhancement of immune response in Peyer's patch
cells cultured with Bifidabacterium breve. J. Dairy Sci, 74 : 1187-1195.

Yi, A. K., Chace, J. H., cowdery, J. S., and Krieg, A. M. (1996). IFN-y promotes IL-6 and
IgM secretion in response to CpG motifs in bacterial DNA and oligodeoxynucleotides. J.
Immunol., 156 : 558-562.

Yohe, H. C. and Ryan, J. L. (1986). Ganglioside expression in macrophages from
endotoxin responder and nonresponder mice. J. Immunol. , 137(3921) : 3927-

Yokokura, T., Nomoto, K, Shimizu, T., and Nomoto, K. (1986). Enhancement of
hematopoietic response of mice by subcutaneous administration of Lactobacillus casei.
Infect. Immun., 52 : 156-160.

Zhang, Y., Broser, M., and Rom, W. M. (1994). Activation of the interleukin 6 gene by
Mycobacterium tuberculosis and lipopolysaccharide is mediated by nuclear factors NF-
IL-6 ans NF-KB. Prac. Nat]. Acad. Sci. USA, 91 : 2225-2229.

244

Zhou, H. R., Yan, D., and Pestka, J. J. (1997). Differential cytokine mRNA expression in
mice after oral exposure to the trichothecene vomitoxin (deoxynivalenol): Dose response
and time course. T oxica]. App]. Pharmacol, 144 : 294-305.

nrcurcnu STATE UNIV. LIBRARIES
1|llllllllllllll111mllllllllllllllllllllllllllllll
31293017127501